INCREASING RETURNS, THE COMPETITIVE MODEL AND THE ENIGMA THAT WAS ALFRED MARSHALL
| Author | A. L. LEVINE |
| Date | 01 November 1980 |
| Published date | 01 November 1980 |
| DOI | http://doi.org/10.1111/j.1467-9485.1980.tb00931.x |
Scvllrsh
Journul
of
Pvliticul
Economy.
Vol.
21,
No.
3,
November
1980
Q
1980
Scottish Economic Society
0036-9292/80/00190260 802.00
INCREASING RETURNS, THE COMPETITIVE
MODEL AND THE ENIGMA THAT WAS
ALFRED MARSHALL
A. L. LEVINE
The University
of
New Brunswick
I
INTRODUCTION
In his recent paper in this Journal on “Whatever Happened to Marshall’s
Theory of Value?” Brian Loasby raised still another question
:
what happens
to the notion ofequilibrium, in Marshall’s
Principles,
in the face of the analyti-
cal problems that are presented by increasing returns? For Loasby, whatever
one may detect
of
long-run static equilibrium in the increasing returns case,
in the
Principles,
is merely an “extremely convenient analogy” (Loasby,
1978,
p.
7)
or an “expedient” (executed on the level of the industry, not the
firm,
ibid.
p. 10). However, Loasby’s reduction of conventional equilibrium
to the status of “expedient’. in his interpretation of Marshall’s treatment of
increasing returns in the
Principles
should have been carried even further. In
fact, conventional equilibrium-individual-firm
or
industry-should be
expunged altogether from a felicitous rendering
of
Marshall’s intentions
here and simply replaced by “competitive conditions”,’ where such condi-
tions are a matter
of
both numbers of
firms
and product differentation.
Marshall’s attempted accommodation of increasing returns may then be
read as an effort to reconcile the latter with some sort of non-formal “com-
petitive” dispensation rather than with formal static “equilibrium”. Now,
there is nothing startlingly original about this. It would seem to accord with
what we may refer to as the consensal view of the reconciliation exercise.2
Of much greater moment is the following. Not only should Loasby’s
partial excision of “equilibrium” in the increasing returns case, in his
rendering of Marshall’s treatment
of
this case in the
Principles,
be
carried
to the extreme, but the role of the time element-of the adjustment of the
firm through time-in the reconciliation exercise needs to be examined in
some detail, and certainly given primacy of place.3 Moreover (contrary to
!
Date
of
receipt of final manuscript: 5 May
1980.
’
The expression, “competitive conditions”, as a description of the “object”
of
the reconcilia-
tion exercise is used by Shove in his Marshall centenary article. (See Shove, 1942,
p.
304.)
Shove (1942, p. 304), Hague (1958), Newman (1960), and Blaug (1962, p. 391), may all be
said, in various ways to reflect the “consensal view”. (See p.
12,
below, for an account of this
“consensus”.) Newman, however, imparts somewhat of a different twist. (See footnote 3.)
Loasby pointed
to
Marshall’s awareness, in this context, of the intertemporal adjustment
of
the firm, and
so,
too, did Newman (1960) and Jenner (1964-65). However, Newman insisted
that Marshall’s way out of the problem of conceptualizing an equilibrium in the increasing-
returns case is to be found in a macroscopic or industry-level--but still formal- notion
of
equilibrium, “a notion influenced perhaps by Marshall’s interest in biology”. (Newman, 1960,
p.
590.)
260
THE
ALFRED MARSHALL ENIGMA
26
1
the consensal view), Marshall’s biological analogies should be read as
having, and intendedly so, relatively little substantive content in this
reconciliation exercise.
These perceptions of Marshall’s attempt to accommodate increasing
returns to some
sort
of competitive regimen form the subject matter of Parts
I11
and
IV
of the present paper. They are preceded, in Part
11,
by an effort
to determine what exactly Marshall meant by increasing
return^.^
WHAT
DID
MARSHALL
MEAN
BY
“INCREASING
RETURNS?”
Beginning students of microeconomic theory who wish to grasp the mean-
ing of the
static
principle
of
returns to
scale,
as distinct from returns to a
factor
or the phenomena encompassed by the principle of variable propor-
tions, are taught to conceptualize in wholly static fashion an input-output
relation or production function
(Q
=
Q(K,
L,
. .
.))
usually within that
(conceptual) time frame known as the long run. Assumptions concerning
given technology and given factor qualities will preserve the static quality
of the analysis, students are told, while the assumption of constant factor
proportions serves to demarcate the principle of returns to scale from that
(usually) short-run but equally static notion, returns to a factor or the
principle
of
variable proportions.
So
safeguarded by assumptions,
returns to
scale
will mean precisely what the words suggest
:
returns to changes in scale
of operation uninterfered with by changes in factor proportions, technology
or factor q~ality.~
And, of course, students are also taught to distinguish among three condi-
tions of returns to scale-increasing, diminishing and constant-and to look
for the sources of increasing returns to scale in such phenomena as are per-
mitted by the static framework and which usually include (l), those redeploy-
ments of factors (especially labour) which although they might come under
the rubric, increased division of labour and specialization, are not tanta-
mount to a major reordering, and
(2),
indivisibilities (although this latter is
certainly not beyond controversy. See, for example, Chamberlin, 1948,1949,
and Hahn, 1949.)
We
shall, with one exception, confine ourselves to the eighth edition
of
the
Principles.
On the other hand,
it
may be argued that both inside and outside the textbook literature
there is neither unanimity nor even clarity concerning all the implications of “returns to scale”.
Thus, if
one
takes
as the relevant inputs, units
of
factor
services
rather than physical quantities
of factors and postulates changes in the pattern
of
factor specialization
or
whatever other re-
arrangements are permissible within the ambit of static analysis (see the next paragraph in the
text), then the consequences of such
a
modelling
of
a production scenario may be changes in
the proportions in which factor services are combined, and hence an obliteration of that con-
ceptual strict separation between the principle
of
returns to scale and the principle
of
variable
proportions-+ven though the proportions of physical quantities
of
factors remain the same.
(Cp.
W.
D.
Maxwell, 1969, pp.
239-241.
The pagination is that
of
the reprint in Nee].)
To continue reading
Request your trial