The Classical Notion of

Competition Revisited
Neri Salvadori and Rodolfo Signorino

The gravitation of market prices toward natural prices in all markets char-
acterized by free competition was widely affirmed by the classical authors,
despite the lack of a formal proof. In the wake of the revival of interest in
classical economics fostered by the publication of Production of Com-
modities by Means of Commodities (Sraffa 1960), since the mid-1970s a
vast literature has blossomed concerning the stability of long-period equi-
librium within multisectoral models of classical inspiration. (For a survey
of this literature, see Bellino 2011.) Besides the formal results achieved on
the subject of stability, the debate on gravitation has stimulated an in-
depth investigation of the classical notion of market competition (Arena
1978; Semmler 1984; Steedman 1984; Duménil and Lévy 1987). This
investigation has also highlighted some unsatisfactory aspects in the lat-
ter, particularly the actual process of market price determination in a situ-
ation of market disequilibrium:

Correspondence may be addressed to Neri Salvadori, Dipartimento di Scienze Economiche,

Università di Pisa, via Cosimo Ridolfi 10, 56124, Pisa, Italy; e-mail: nerisal@ec.unipi.it. This
article was prepared for the Italian National Research Project, cofinanced by the Italian Minis-
ter of University, “Heterogeneous Sectors, Growth, and Technical Change” (2007). A previous
version of this article was presented at the thirteenth annual conference of the European Society
for the History of Economic Thought, Thessaloniki, Greece, April 23–26, 2009, and at the
seventh annual conference of the Italian Association for the History of Political Economy,
Trento, Italy, May 30–June 1, 2010. We wish to thank Alain Alcouffe, Enrico Bellino, Stefano
Fiori, Heinz D. Kurz, Manuela Mosca, Arrigo Opocher, and Maria Pia Paganelli for their com-
ments. The usual caveats apply.
History of Political Economy 45:1 DOI 10.1215/00182702-1965222
Copyright 2013 by Duke University Press
150 History of Political Economy 45:1 (2013)

Little is said concerning the actual process [that governs the function-
ing of the market for commodities]. For example, it is not clear who is
changing prices, what information is used, when this change occurs,
what the outcome is on the market etc. (Duménil and Lévy 1987, 136)
[L’originalità della teoria classica della libera concorrenza] non è però
in grado di nascondere certe insufficienze: l’economia politica classica
si scontra infatti con il problema centrale dell’articolazione tra prezzi di
mercato e prezzi naturali. La soluzione di queste difficoltà richiede una
nuova definizione di questa articolazione. (Arena 1978, 323)1
The aim of this article is to detect and highlight those elements present
in classical texts that may be fruitfully employed to overcome the above
drawbacks of the classical view of market competition. Moreover, as recent
commentators on classical economics have emphasized, in a situation of
disequilibrium, when the quantity of a given commodity brought to the
market differs from Smithian effectual demand, the likely outcome is that
of a dispersion of prices: “Whereas natural price is by definition a singular
magnitude for each kind of commodity, the competitive processes whereby
market prices deviate from natural price under conditions of market
imbalance are consistent with, indeed likely to cause, transactions at non-
uniform prices. That is to say, market price as literally the actual prices
at which transactions occur when there is such market imbalance, is not
in general a singular magnitude” (Aspromourgos 2009, 72; see also pages
85 and 88). Accordingly, we claim that the classical theory of market prices
gives scope to the analysis of the behavior of the agents acting in situa-
tions of market imbalance through some of the analytical tools and formal
results achieved by the modern game-theoretic approach to oligopoly the-
ory, namely, the notion of mixed (i.e., nondeterministic) price strategy.
In what follows, we distinguish and compare two different conceptions
of market competition: the Walrasian notion of perfect competition and
the classical notion of free competition, focusing in particular on Adam
Smith and Karl Marx. We stress that while the Walrasian notion may be
described as an equilibrium state in which atomistic agents treat prices
parametrically, the classical notion is a situation in which agents employ
their market power by setting prices strategically. The absence of a price-

1. [The originality of the classical theory of free competition] is not, however, able to hide
certain inadequacies: classical political economy clashes with the central problem of the
articulation between market prices and natural prices. The solution of such difficulties
requires a new definition of this articulation. [Our translation.]
 Salvadori and Signorino / The Classical Notion of Competition 151

taking assumption in the classical authors is the main reason why the
theoretical difficulties besetting the Walrasian notion outside market-
clearing equilibrium do not plague the classical notion as well. Yet though
for the classical authors price undercutting and outbidding are the typical
phenomena that occur in any market characterized by free competition, it
is fair to say that they went no farther than to provide only some unsys-
tematic guidelines on how to analyze in due detail the competitive process
of market price determination. Among other things, we show that Marx’s
extensive use of metaphors and numerical examples hides the modern tax-
onomy of buyers’ market, sellers’ market, and mixed strategy equilibrium
in the capacity space of a standard Bertrand duopoly model. In particular,
we argue that Marx was conscious that between the buyers’ market, in
which the price is determined by the reservation price of sellers, and the
sellers’ market, in which the price is determined by the reservation price
of buyers, there is something in which the price is not unique and any
equilibrium can concern only distributions of probability (mixed strate-
gies) on the behavior of the traders. We substantiate this fact by using a
formalism derived from the contemporary analysis of Bertrand competi-
tion. (The elements required to identify a necessary and sufficient condi-
tion to separate these three different market outcomes are sketched in an
appendix.) We are aware that our analysis, once duly developed, needs to
be related to the contemporary gravitation literature. Yet we defer investi-
gation of this issue to a future paper.2
The structure of the article is as follows. Section 1 compares two differ-
ent notions of the concept of market competition, the classical notion of
free competition and the neoclassical notion of perfect competition, and
highlights some problematic aspects of the latter, absent in the former.
Sections 2 and 3 assess the classical theory of free competition, as devel-
oped by Adam Smith and Karl Marx, with particular concern for market
price determination. Section 4 investigates how the classical notion of
competition has percolated into the modern literature on so-called Ber-
trand competition. Section 5 concludes.

2. The basic model used to represent the classical gravitation process is the so-called
cross-dual model, in which (1) sectoral outputs change in response to differentials in the rates
of profit and (2) market prices change in response to excess demand of the various commodities.
While the first adjustment process is widely accepted as a fair formalization of the classical
principle of capital mobility, the second appears as a mechanical transposition of the Walra-
sian price adjustment process and is the ultimate culprit behind the intrinsic instability of
cross-dual gravitation processes. The present article proposes a totally different market price
formation and should be able to resolve the instability problem envisaged in this literature.
152 History of Political Economy 45:1 (2013)

1. Two Different Notions of Market

Competition or Just One?
Few commentators would disagree with the following statement: “Although
the concept of competition has always been central to economic think-
ing . . . it is one that has taken on a number of interpretations and meanings,
many of them vague” (Vickers 1995, 3). In particular, John Vickers distin-
guishes the notion of perfect competition—a “seemingly tranquil equilib-
rium state in which well-informed agents treat prices parametrically”—
from the “original and ‘real’ concept” of competition, a “rivalrous behaviour
with respect to prices and other variables in a world characterized by flux,
uncertainty and disequilibrium” (7).
Despite the differences between these two notions, many authors have
come (more or less explicitly) to consider the classical notion of free
competition nothing but a primitive and pre-analytical version of the neo-
classical theory of perfect competition, a version still imbued with casual
empiricism: “It is a remarkable fact that the concept of competition did not
begin to receive explicit and systematic attention in the main stream of
economics until 1871. This concept . . . was long treated with the kindly
casualness with which one treats of the intuitively obvious. . . . ‘Compe-
tition’ entered economics from common discourse, and for long it con-
noted only the independent rivalry of two or more persons” (Stigler 1957,
1). Such an interpretation has deeply influenced many of the leading expo-
nents of neoclassical economics. (See, e.g., Arrow and Hahn 1971, in par-
ticular chapter 1, “Historical Introduction,” and Samuelson 1978.) Even
outside the neoclassical camp this interpretation has found supporters,
such as Nicholas Kaldor (1972, 1241), when he claims that “one can
trace a more or less continuous development of price theory from the
subsequent chapters of Smith [the fourth chapter of the Wealth of Nations]
through Ricardo, Walras, Marshall, right up to Debreu and the most
sophisticated of present-day Americans.” Vickers (1995, 7) himself con-
cludes that “the claim that there are two concepts of competition is some-
what misleading.”
Nonetheless, historians of economic thought who have endorsed an
alternative point of view have not been lacking. Paul McNulty (1967, 397)
not only argued that the classical notion of competition as a behavioral
process is radically different from the neoclassical notion of competition
as an equilibrium state but also went so far as to claim that the neoclas-
sical assumption of individual price-taking behavior is entirely alien to
 Salvadori and Signorino / The Classical Notion of Competition 153

the classical analysis:3 “Smith’s concept of competition was decidedly not

one in which the firm was passive with respect to price but was, rather, one
in which the market moved toward equilibrium through the active price
responses of its various participants.” Therefore, what in the previous
interpretation appears simply as a process of analytical refinement, in
McNulty’s view is no less than “a basic conceptual change” (397). More-
over, McNulty adds that, pace George Stigler, the classical notion of com-
petition is far from being derived from casual empiricism. Smith is the
great systematizer of the analysis of the concept of market competition
carried out by a series of authors before him (most notably, Cantillon and
Turgot), and his specific contribution was to raise the concept of compe-
tition to a “general organizing principle of economic society. . . . After
Smith’s great achievement, the concept of competition became quite lit-
erally the sine qua non of economic reasoning” (McNulty 1967, 396–97;
see also McNulty 1968, 646–47).4
McNulty’s interpretation raises the matter of investigating the theo-
retical conditions that have led to such a dramatic change of meaning in
the original notion of competition. A first step in this direction may be
found in the distinction between two different notions of economic sci-
ence: (1) economics as the science that studies a system of forces and
(2) economics as the science that studies a system of relations—a dis-
tinction introduced by Marco Dardi (1983) that was recently emphasized
by Nicola Giocoli (2005):
According to the system-of-forces (SOF) view, economics is a disci-
pline whose main subject is the analysis of the economic processes

3. From this point of view, McNulty’s interpretation is akin to those of Samuel Hollander
(1973) and John Eatwell (1987). The former claims that “the Smithian conception of competi-
tion must be carefully distinguished from the modern conception which envisages sellers (and
consumers) as ‘price-takers’ rather than ‘price-makers’” (126), while the latter points out that
“the characteristics of ‘perfect’ competition (notably the conditions which ensure price-taking)
are often read back, illegitimately, into Classical discussions of competition” (63). See also High
2001, xiv–xv; and Machovec 1995, chap. 8. While Blaug (2001, 153) defines Arrow and Hahn’s
1971 tribute to Adam Smith as a forerunner of perfect competition (and Pareto optimality)
analysis no less than “a historical travesty,” a more balanced position is endorsed by Michael
Bradley (2010, 238), who argues that “Smithian ‘liberty’ contains the seeds of perfect competi-
tion, but perfect competition is different from ‘perfect liberty’ in some critical respects, particu-
larly the nature of competition and the role of entrepreneurs.”
4. McNulty (1967, 395–96) provides a discussion of Cantillon and Turgot as forerunners of
Smith on competition. An extensive treatment of the evolution of the notions of price, cost,
and competition before Smith may be found in Aspromourgos 2009, 101–10.
154 History of Political Economy 45:1 (2013)

generated by market and not-market forces, including—but by no

means exclusively—the processes leading the system to an equilib-
rium. According to the system-of-relations (SOR) view, economics is a
discipline whose main subject is the investigation of the existence and
properties of economic equilibria in terms of the validation and mutual
consistency of given formal conditions, but that has little if anything to
say about the meaningfulness of these equilibria for the analysis of
real economic systems. (Giocoli 2005, 180)
According to Giocoli, the SOF view was the dominant vision up to the
years between the two world wars, while the SOR view gained popularity
only in the second postwar period when considerable intellectual effort
was devoted to the project of a full axiomatization of economic science
(180). Similarly, in a series of papers Blaug (1997, 2002, and 2003) indi-
cated in the formalist revolution of the 1950s and the parallel rise to dom-
inance of the Walrasian general equilibrium theory the two driving forces
that led to the decline of the classical (and early neoclassical) notion of
competition as a process and its replacement with the modern notion of
competition as an end-state (with the associated first and second funda-
mental theorems of welfare economics).
Thus it may be claimed that the semantic shift in the notion of competi-
tion is part of a more general process of redefining central categories and
concepts of economic analysis started in the 1930s with the rediscovery of
Walrasian general equilibrium theory (Donzelli 1990, chap. 9) and fully
accomplished in the late 1970s with the rediscovery of the Nash equilib-
rium concept (Giocoli 2003, chap. 5): the classical notion of market com-
petition makes little sense outside the SOF view, while the Walrasian
notion of competition perfectly fits the theoretical standards set by the
SOR view.
There are at least two reasons why the distinction between the classical
notion of competition and the neoclassical one and a careful analysis of the
theoretical domain of the latter is not simply a historiographical exercise:
1. The neoclassical theory of perfect competition carries with it some
theoretical difficulties alien to the classical notion of free competition.
2. The classical notion of competition can be made analytically precise
in terms of the modern concept of mixed strategies equilibrium.
The final part of this section is devoted to substantiating point 1 above,
while we defer analysis of point 2 to section 4.
 Salvadori and Signorino / The Classical Notion of Competition 155

As pointed out by Jerry Green (1974), every market equilibrium con-

cept requires the specification of a consistent set of behavioral postulates
that prescribe what happens in equilibrium and what happens outside
equilibrium. In the Walrasian framework, the behavioral postulates that
define the situation of equilibrium differ from those that define the adjust-
ment mechanism in disequilibrium. As regards the former, the behavioral
postulate is that every agent assumes market prices as parametrically
given and, on the basis of such prices and other constraints, maximizes his
or her own objective function. By contrast, in disequilibrium, the behav-
ioral postulate is that a meta-agent, the auctioneer, determines market
prices according to market excess demands. Moreover, no transactions
among the agents are allowed to take place during the adjustment process.5
As a consequence, the price-taking behavior assumption implies that each
individual firm in a given market has no incentive to set a price different
from the ruling market price and has no incentive to carry on transactions
at a price other than the Walrasian market-clearing price. Thus such an
assumption drastically reduces the theoretical domain of the theory to
equilibrium, market-clearing situations.6
To our knowledge, Kenneth Arrow (1959) was the first to highlight the
logical difficulties besetting the neoclassical theory of perfect competition:
the Law [of supply and demand, Arrow’s equation 3: dp/dt = h(S − D)
with h′ < 0 and h(0) = 0] is not on the same logical level as the hypoth-
eses underlying equation 1 [D = f (p) and S = g( p)]. It is not explained
whose decision it is to change prices in accordance with equation 3.

5. From this perspective Edgeworth’s concept of equilibrium (core) and his adjustment pro-
cess in disequilibrium (recontracting) are superior to the Walrasian ones: “The recontracting
process . . . is based on the same behavioral postulate, blocking by coalitions, that is used to
define the solution concept, the core. [The core is defined to be the set of all unblocked alloca-
tions. That is, it is the set of all allocations such that no subset of the participants can improve
the position of all its members by withdrawing from the system and using only the resources of
its members.] This seems to be a desirable property. It is, however, not shared by most studies of
disequilibrium price dynamics because these involve price changes brought about by a market
manager or other artificiality. Prices do not vary as a consequence of the maximizing behavior
of individuals” (Green 1974, 22).
6. Current textbooks also acknowledge this fact: “Strictly speaking, it is equilibrium market
prices that [consumers and producers] will regard as unaffected by their actions. . . . For the
price-taking assumption to be appropriate, what we want is that [consumers and producers] have
no incentive to alter prices that, if taken as given, equate demand and supply (we have already
seen that [consumers and producers] do have an incentive to alter prices that do not equate
demand and supply)” (Mas-Colell, Whinston, and Green 1995, 314n1, 315; emphasis added).
156 History of Political Economy 45:1 (2013)

Each individual participant in the economy is supposed to take prices

as given and determine his or her choices as to purchases and sales
accordingly; there is no one left over whose job it is to make a deci-
sion on price. (Arrow 1959, 43)7
In short, for Arrow, in the perfect competition setup there is no place
left for “a rational decision with respect to prices,” thus implying the con-
clusion that “perfect competition can really prevail only at equilibrium”
(41). The solution proposed by Arrow to study market price dynamics
outside market-clearing equilibrium consists in turning to the theory of
monopoly: “When supply and demand do not balance, even in an objec-
tively competitive market, the individual firms are in the position of
monopolists as far as the imperfect elasticity of demand for their products
is concerned” (46). However, Arrow claims that standard monopoly the-
ory must be modified so as to remove the assumption that monopolists
know perfectly their own demand curve (besides their own cost curves):
“Uncertainty [as to the demand curve] is a crucial consideration in the
theory of monopolistic price adjustment” (44). In such circumstances,
monopolists will vary their own price, in a process of trial and error, until
they find the price that maximizes their expected profits.
For our purpose, the salient points of Arrow’s analysis are the following:
1. Jevons’s law of indifference, which states that there is only one price
ruling in a competitive market, ceases to be valid in disequilibrium:
“Although the broad tendency will be for prices to rise when demand
exceeds supply, there can easily be a considerable dispersion of
prices among different sellers of the same commodity” (46–47).
2. By assuming that competition takes place on both sides of the mar-
ket, that is, competition among sellers and competition among buy-
ers, a buyers’ market may be distinguished from a sellers’ market:
“By a parallel argument each buyer on a market with an inequality
between supply and demand can be regarded as a monopsonist. . . .
In disequilibrium, the market consists of a number of monopolists
facing a number of monopsonists. The most general picture is that
of a shifting set of bilateral monopolies. . . . In general, it is reason-
able to suppose that if the selling side of the market is much more

7. D (S) is the quantity demanded (supplied) of a given commodity X, p its price, and f(p) and
g(p) the demand and supply functions, respectively, while dp/dt = h(S − D) is the time derivative
that formalizes the law of motion of market price in relation to market excess demand.
 Salvadori and Signorino / The Classical Notion of Competition 157

concentrated than the buying side, the main force in changing prices
will be the monopolistic behavior of the sellers. . . . Similarly, if the
buying side of the market is the more concentrated, as in non-union-
ized labor markets, the dynamics will come from that side” (47;
emphasis added).
In the following two sections we make it clear that these two elements of
Arrow’s contribution may be found in Smith’s and, even more explicitly,
in Marx’s treatment of market prices, thus paving the way for a restate-
ment of the classical notion of competition through contemporary game-
theoretic analysis.8

2. The Classical Notion of Free Competition:

Adam Smith
George Richardson (1975, 350–51) has convincingly argued that “compe-
tition features within The Wealth of Nations in two different contexts;
first, in the account given of the balancing of supply and demand in par-
ticular markets, and, secondly, in the explanation of structural and techno-
logical development. Smith offers us in effect both a theory of economic
equilibrium and a theory of economic evolution; and in each of these com-
petition has a key role to play.” In what follows we concentrate on the
static aspect of the Smithian notion of market competition, concerned
with market price determination, leaving aside its dynamic aspect (see
Lavezzi 2003).
As far as the classical notion of market competition is concerned, the
locus classicus is book 1, chapter 7, of Smith’s ([1776] 1976) Wealth of
Nations (WN).9 Smith’s working assumption is that it is possible to clas-
sify the economic forces in action in a given moment into two broad cat-
egories, (1) those erratic and short-lived forces that determine the market
values both of commodity prices and of the distributive variables and
(2) those systematic and persistent forces that determine the natural values
of the same magnitudes. Classical economists generally hold the view

8. Arrow’s 1959 contribution with its emphasis on the role of monopoly theory in explain-
ing competitive price convergence is the ideal starting point for the subsequent neo-Walrasian
literature on disequilibrium trading (see Donzelli 1990, chap. 9).
9. As is well known, David Ricardo (1951, 91) devotes to the distinction between natural
and market magnitudes just the short chapter 4 of his Principles, where he explicitly refers to
chapter 7 of WN, where “all that concerns this question is most ably treated.”
158 History of Political Economy 45:1 (2013)

that only the latter can be the proper subject of scientific inquiry (Cic-
cone 1999, 70).10
The data from which the Smithian argument starts are the natural rates
of wages, profits, and rents that, sectoral specificities apart, depend mainly
on the conditions of prosperity of the economic system under scrutiny, its
“advancing, stationary, or declining condition” (WN I.vii.1). The natural
price of (re)production of the various commodities derives from the sum-
mation of these three elements. The natural price is therefore a magnitude
that is not immediately formed in the market but that, given some appro-
priate conditions, may come true in the market. Indeed, the natural price
constitutes a sort of a floor for the market price in the sense that the latter
cannot remain for long below the former without seriously jeopardizing
the reproduction of the commodity in question:
The competition of the different dealers obliges them all to accept
of [the natural price], but does not oblige them to accept of less.
. . . The natural price, or the price of free competition, . . . is the
lowest which can be taken, not upon every occasion, indeed, but for
any considerable time together. . . . [It] is the lowest which the sellers
can commonly afford to take, and at the same time continue their
business. (WN I.vii.11, 27)
The theoretical importance of natural prices consists in providing a guide
to the theorist for explaining the dynamic path followed by market prices:
“The natural price, therefore, is, as it were, the central price, to which the
prices of all commodities are continually gravitating. Different accidents
may sometimes keep them suspended a good deal above it, and sometimes
force them down even somewhat below it. But whatever may be the obsta-
cles which hinder them from settling in this center of repose and conti-
nuance, they are constantly tending towards it” (WN I.vii.15).
To study the genesis of market prices and the existing relations between
market prices and natural prices, Smith introduces the concept of effec-
tual demand, that is, “the demand of those who are willing to pay the
natural price of the commodity.” It is to be stressed that the relationship

10. Ricardo (1951, 91–92) clearly states that the focus of his analysis is only natural magni-
tudes: “Having fully acknowledged the temporary effects which, in particular employments of
capital, may be produced on the prices of commodities, as well on the wages of labour, and the
profits of stock, by accidental causes, without influencing the general prices of commodities,
wages or profits, since these effects are equally operative in all stages of society, we will leave
them entirely out of consideration, whilst we are treating of the laws which regulate natural
prices, natural wages and natural profits, effects totally independent of these accidental causes.”
 Salvadori and Signorino / The Classical Notion of Competition 159

between the quantity brought to the market and the effectual demand
determines only the market price of a commodity and not also its natural
price.11 Moreover, “demand” and “supply” are treated by Smith as given
quantities and not as functional relationships between price and quantity
characterized by well-defined formal properties, as they would be in the
neoclassical theory (Garegnani 1983; Aspromourgos 2009, 83–84).12
Given the unplanned nature of market economies, at the end of a pro-
ductive cycle, entrepreneurs may not face in the market a demand able to
absorb the whole of their production at the natural price (at least). This
requires the specification of an adjustment mechanism powerful enough
to bring about effective convergence to a situation in which the produced
quantity coincides with the effectual demand: in the absence of such a
mechanism, natural prices may not constitute a reliable guide to explain
the movements of market prices.13
In short, the adjustment mechanism envisaged by classical authors is as
follows. At the end of a productive cycle, the entrepreneur brings to the
market a given quantity of produced commodity resulting from the produc-
tion decisions taken at the beginning of the cycle just concluded. Of course,
this quantity cannot be modified to adjust to the demand actually encoun-
tered on the market. Thus the adjustment variable is constituted by the com-
modity’s selling price. Smith assumes that, in the presence of a gap between
production and effectual demand, a sort of auction starts among the agents
that happen to be on (what we today would call) the long side of the market:
such agents are prepared to offer higher and higher prices (in case of excess
demand) or lower and lower prices (in the case of excess supply).
Once the market price of any commodity happens to be different
from its natural price, this causes an imbalance in the distributive sphere
in the sense that the remunerations of those people who have contributed

11. See Ricardo’s (1951, 382) rejection of the opinion that price depends solely on the
proportion between these two quantities.
12. Differences between the classical and the neoclassical theories of value and distribu-
tion have been emphasized by such authors as Krishna Bharadwaj (1978), Alessandro Ronca-
glia (1978), and Pierangelo Garegnani (1984). Nonetheless, historiographical controversies
are still very much alive: see Blaug 1999 and 2009 versus Kurz and Salvadori 2002 and 2010.
13. This consideration may explain Ricardo’s (1951, 90; emphasis added) emphasis on the
effectiveness of the adjustment mechanism: “When we look to the markets of a large town, and
observe how regularly they are supplied both with home and foreign commodities, in the quan-
tity in which they are required, under all circumstances of varying demand . . . without often
producing either the effect of a glut from too abundant a supply, or an enormously high price
from the supply being unequal to the demand, we must confess that the principle which appor-
tions capital to each trade in the precise amount that is required, is more active than is gener-
ally supposed.”
160 History of Political Economy 45:1 (2013)

to the production of the commodity prove different from their respective

natural values. In the absence of entry/exit barriers and in the presence
of market transparency, the difference between the market price and the
natural price brings about (1) an intersectoral reallocation of economic
resources in search of the highest market remuneration and (2) a variation
in the produced quantity of the commodity in the following periods. This
process comes to a halt only when the produced quantity and demanded
quantity balance in correspondence of the natural price and the market
values of wages, profits, and rent equal their respective natural values.14
Therefore the imbalance in the sphere of circulation (the discrepancy
between the natural price and market price of a commodity) spills over
to the sphere of distribution (the discrepancy between natural values and
market values of wages, profits, and rent) and, finally, to the sphere of pro-
duction (the intersectoral reallocation of productive resources and varia-
tion in quantities produced in the following periods).
The assumed tendency of market values toward their respective natural
values is based on two assumptions: (1) the owners of the employed inputs
consider, besides the outlay costs, also the opportunity costs in their deci-
sions as to where to allocate their economic resources (Aspromourgos
2009, 67, 98) and (2) there are but negligible barriers to the intersectoral
mobility of economic resources (91):
When the price of any commodity is neither more nor less than what
is sufficient to pay the rent of the land, the wages of the labour, and the
profits of the stock employed in raising, preparing, and bringing it to
market, according to their natural rates, the commodity is then sold for
what may be called its natural price.
The commodity is then sold precisely for what it is worth, or for what
it really costs the person who brings it to market; for though in common
language what is called the prime cost of any commodity does not
comprehend the profit of the person who is to sell it again, yet if he sell
it at a price which does not allow him the ordinary rate of profit in his
neighbourhood, he is evidently a loser by the trade; since by employing
his stock in some other way he might have made that profit. . . .
Though the price, therefore, which leaves him this profit is not always
the lowest at which a dealer may sometimes sell his goods, it is the low-

14. However, classical economists were perfectly aware of the existence of profit and wage
rate differentials. For a modern treatment, see Kurz and Salvadori 1995, chapter 11.
 Salvadori and Signorino / The Classical Notion of Competition 161

est at which he is likely to sell them for any considerable time; at least
where there is perfect liberty, or where he may change his trade as
often as he pleases. (WN I.vii.4–6; emphases added)
The above shows that Smith devotes much care to determining natu-
ral values and to the gravitation process of market magnitudes to their
natural counterparts. The same cannot be maintained as regards the
question of market price determination, particularly when the market is
not in a situation of long-period equilibrium. Taking stock of Smith’s
sparse hints on this subject, it is possible to point out what follows.
First, in those markets in which competition is not free (e.g., because
of legal monopoly and/or the presence of a guild, a collusive agreement,
a law or a rule that somehow prevents economic agents allocating their
resources in the sector they prefer) or where there are industrial secrets,
entrepreneurs voluntarily limit the produced quantity so that the market
is left understocked and the market price stays artificially high:
The exclusive privileges of corporations, statutes of apprenticeship,
and all those laws which restrain, in particular employments, the com-
petition to smaller number than might otherwise go into them, have
the same tendency, though in a less degree. They are a sort of enlarged
monopolies, and may frequently, for ages together and in whole classes
of employments, keep up the market price of particular commodities
above the natural price, and maintain both the wages of the labour and
the profits of the stock employed about them somewhat above their nat-
ural rate. (WN I.vii.28)
Conversely, where competition is free and industrial secrets absent,
price undercutting starts as soon as at least two competitors are present in
the market. This process is amplified by increasing the number of com-
petitors, since this fact makes the establishment of a collusive agreement
more unlikely:
The quantity of grocery goods, for example, which can be sold in a
particular town, is limited by the demand of that town and its neigh-
bourhood. The capital, therefore, which can be employed in the gro-
cery trade cannot exceed what is sufficient to purchase that quan-
tity. If this capital is divided between two different grocers, their
competition will tend to make both of them sell cheaper, than if it
were in the hands of one only; and if it were divided among twenty,
their competition would be just so much the greater, and the chance
162 History of Political Economy 45:1 (2013)

of their combining together, in order to raise the price, just so much

the less. (WN II.v.7)
Second, Smith’s explanation for the determination of market prices in
situations of disequilibrium includes not only elements that are seemingly
the fruit of casual observation and that he does not analyze in greater
detail (the wealth of the buyers and their desire to get the commodity ver-
sus the necessity of the sellers to dispose of their own commodities) but
also elements that he instead systematically applies in his analysis of the
various markets. Of the latter, the most significant is the relative number
of sellers and buyers and their relative ability to make a binding agree-
ment. The market price will be high or low depending on whether buyers
are more numerous than sellers (and vice versa): the buyers “bid against
one another” offering higher and higher prices, the sellers “bid against one
another” offering lower and lower prices. The relative number of the buy-
ers in relation to the sellers is therefore the crucial element: every time that
the agents on one side of the market are few and are able to communicate
(e.g., because they operate in the same place such as a town) while the
agents on the other side of the market are many and are unable to com-
municate (e.g., because they are isolated and scattered in the countryside),
the bargaining from which the market price springs will obviously be
more favorable to the former. This is particularly evident in Smith’s analy-
sis of the labor market:
What are the common wages of labour depends every where upon
the contract usually made between those two parties, whose interests
are by no means the same. The workmen desire to get as much, the
masters to give as little as possible. The former are disposed to com-
bine in order to raise, the latter in order to lower the wages of labour.
It is not, however, difficult to foresee which of the two parties must,
upon all ordinary occasions, have the advantage in the dispute, and
force the other into compliance with their terms. The masters, being
fewer in number, can combine much more easily; and the law, besides,
authorises, or at least does not prohibit their combinations, while it pro-
hibits those of the workmen. . . .
When in any country the demand for those who live by wages;
labourers, journeymen, servants of every kind, is continually increas-
ing; when every year furnishes employment for a greater number than
had been employed the year before, the workmen have no occasion to
combine in order to raise their wages. The scarcity of hands occasions
a competition among masters, who bid against one another, in order to
 Salvadori and Signorino / The Classical Notion of Competition 163

get workmen, and thus voluntarily break through the natural combina-
tion of masters not to raise wages. (WN I.viii.11–12, 17)
In the following section we show how Marx draws on and develops
these elements of Smith’s treatment of market prices.

3. The Classical Notion of Free Competition:

Karl Marx
In chapter 3 of Wage-Labour and Capital (Marx [1847] 1933) it is pos-
sible to find a vivid description of price determination in the market for
a raw material, cotton. We think that such a description provides a clue
to the young Marx’s view of the competitive process. The chapter bears
the title “By what is the price of a commodity determined?” and Marx’s
answer is the quite conventional one: “By the competition between buyers
and sellers, by the relation of the demand to the supply, of the call to the
offer” (21). Yet, immediately after, he adds that “the competition by which
the price of a commodity is determined is threefold” (21; emphasis added).
The first element highlighted by Marx is competition among the sellers:
“Whoever sells commodities of the same quality most cheaply, is sure to
drive the other sellers from the field and to secure the greatest market for
himself. . . . [It is competition among the sellers] which forces down the
price of the commodities offered by them” (21). The second element is
competition among the buyers that “causes the price of the proffered com-
modities to rise” (21). These two aspects of competition are not considered
sufficient to fully determine the outcome of the competitive process. In
fact, Marx adds a third and last element:
Finally, there is competition between the buyers and the sellers: these
wish to purchase as cheaply as possible, those to sell as dearly as pos-
sible. The result of this competition between buyers and sellers will
depend upon the relations between the two above-mentioned camps
of competitors—i.e., upon whether the competition in the army of sell-
ers is stronger. Industry leads two great armies into the field against
each other, and each of these again is engaged in a battle among its
own troops in its own ranks. The army among whose troops there is
less fighting, carries off the victory over the opposing host. (21)15

15. Marx’s treatment of competition in this passage echoes James Steuart’s notion of double
competition. Marx was well acquainted with the work of Steuart, with whom he often took issue
(Denis 1999). On Steuart’s notion of double competition, see Menudo and Tortajada 2009.
164 History of Political Economy 45:1 (2013)

We claim that the metaphor of the two armies that, at one and the
same time, are engaged in fighting one another and in their own ranks,
coupled with the suggestion that the result of the battle is eventually
decided by the interplay of these two levels of fighting, paves the way to
an interesting analytical intuition. In our view, Marx’s rhetoric foreshad-
ows the modern notion of a mixed strategy equilibrium: the outcome of
market competition need not be univocally determined, even if optimal
(mixed) strategies are.
To clarify his thought, Marx goes on to provide a concrete example.
Marx’s choice of a raw material market for this didactic purpose is illumi-
nating. In the market of a consumption good it is quite obvious to assume
a multitude of atomistic buyers. In such a case, competition among buy-
ers would be reduced to their reservation prices and, eventually, described
through a demand curve. It may not be so in the case of a raw material
market, where the number of buyers may exceed that of sellers, and, in
some cases, it is even possible to reverse the image of atomistic buyers to
that of atomistic sellers. Marx’s example starts with the analysis of what,
in modern terminology, is called a sellers’ market: “Let us suppose that
there are 100 bales of cotton in the market and at the same time purchas-
ers for 1,000 bales of cotton” (21). The fact that the demand is many (ten!)
times greater than the supply is very likely to be intentional: if there were
100 bales and purchasers for 110, conditions would have not been, in
Marx’s opinion, those of a sellers’ market. On the contrary, 100 to 1,000 is
considered enough to obtain that
the cotton sellers, who perceive the troops of the enemy in the most vio-
lent contention among themselves, and who therefore are fully assured
of the sale of their whole 100 bales, will beware of pulling one another’s
hair in order to force down the price of cotton at the very moment in
which their opponents race with one another to screw it up high. So, all
of a sudden, peace reigns in the army of sellers. They stand opposed to
the buyers like one man, fold their arms in philosophic contentment and
their claims would find no limit did not the offers of even the most
importunate of buyers have a very definite limit. (22)
Obviously, the ratio of 1 to 10 is, in itself, neither a necessary nor a suf-
ficient condition. This is not the place to find a necessary and sufficient
condition in general, yet an attempt to pinpoint the elements required for
such a condition can be made with the help of a simple symbolism. This is
attempted in a very special case in the appendix. Going back to the exam-
 Salvadori and Signorino / The Classical Notion of Competition 165

ple, Marx continues by introducing the buyers’ market: “It is well known
that the opposite case, with the opposite result, happens more frequently.
Great excess of supply over demand; desperate competition among the
sellers, and a lack of buyers; forced sales of commodities at ridiculously
low prices” (22). Marx’s text reveals that, for him, the buyers’ market and
sellers’ market are not contiguous in the sense that between them there is
something, but apart from the metaphor of the two armies, his readers are
just left with the obvious remark that “in the same proportion in which
[competition among the sellers] decreases, the competition among the
buyers increases. Result: a more or less considerable rise in the prices of
commodities” (22).
The Marxian text continues by introducing long-period considerations,
that is, the gravitation of market prices toward prices of production (here
Marx uses the expression costs of production) as a consequence of capital
migration from (into) those sectors where market prices are below (above)
costs of production. Yet in Marx’s view, market prices are not to be dis-
missed lightly as theoretically insignificant. Marx, in fact, goes so far as to
claim that the typical market outcome is a market price above or below
costs of production, while the equality between the two should be consid-
ered an exception:
The determination of price by the cost of production is not to be under-
stood in the sense of the bourgeois economists. The economists say that
the average price of commodities equals the cost of production: that is
the law. The anarchic movement, in which the rise is compensated for
by a fall and the fall by a rise, they regard as an accident. We might just
as well consider the fluctuations as the law, and the determination of the
price by cost of production as an accident—as is, in fact, done by cer-
tain other economists. But it is precisely these fluctuations which,
viewed more closely, carry the most frightful devastation in their train
and, like an earthquake, cause bourgeois society to shake to its very
foundations—it is precisely these fluctuations that force the price to
conform to the cost of production. In the totality of this disorderly
movement is to be found its order. In the total course of this industrial
anarchy, in this circular movement, competition balances, as it were,
the one extravagance by the other. (24)
The reader might think that the elder Marx, equipped with an improved
understanding of the classical notion of prices of production and with a
more mature version of his own theory of labor-value, would not have
166 History of Political Economy 45:1 (2013)

endorsed the foregoing analysis by the young Marx. We think that this is
not the case, as witnessed by book 3, chapter 10, of Capital (Marx [1894]
1909). This chapter, bearing the title “Compensation of the Average
Rate of Profit by Competition. Market Prices and Market Values. Surplus-
profit,” is located in part 2, where Marx is confronted with the (insur-
mountable) problem of reconciling the origin of profit from surplus value
with a uniform rate of profit and a uniform rate of surplus value among
sectors. This is not the place to provide a thorough assessment of this
chapter. It suffices to note the following.
After identifying in II.X.14 the conditions to be met in order that “the
prices at which commodities are exchanged with one another may cor-
respond approximately to their values,” Marx adds in II.X.15: “[The
fact that] the commodities of the various spheres of production are sold at
their value implies, of course, only that their value is the center of gravity
around which prices fluctuate, and around which their rise and fall tends
to an equilibrium.” This sentence has been often quoted in the modern
literature on gravitation. However, it is not clear whether Marx thinks that
the price is unique at each moment of time or, rather, that there is a con-
stellation of prices at each moment of time. The difference is substantial.
If sellers and buyers follow mixed strategies instead of pure strategies,
there is clearly a constellation of prices at each moment of time. Marx also
identifies two simple cases. In the first “demand is so strong that it does
not let up when the price is regulated by the value of commodities pro-
duced under the most unfavorable conditions” (II.X.16); in this case these
conditions determine the market value. In the second, “the mass of the
produced commodities exceeds the quantity which is ordinarily disposed
of at average market-values” and, as a consequence, “the commodities
produced under the most favorable conditions regulate the market value”
(II.X.16). Marx is more interested in the result of this process than in the
analysis of less simple cases.16 However, in II.X.51 he claims:
That side of competition, which is momentarily the weaker, is also that
in which the individual acts independently of the mass of his competi-
tors and often works against them, whereby the dependence of one

16. “No matter what may be the way in which prices are regulated, the result always is the
following: 1) The law of value dominates the movements of prices, . . . 2) The average profit
which determines the prices of production must always be approximately equal to that quan-
tity of surplus-value, which falls to the share of a certain individual capital in its capacity as
an aliquot part of the total social capital” (II.X.17–18). (All references to Capital give part
number, chapter number, paragraph number.)
 Salvadori and Signorino / The Classical Notion of Competition 167

upon the other is impressed upon them, while the stronger side always
acts more or less unitedly against its antagonist. If the demand for this
particular kind of commodities is larger than the supply, then one buyer
outbids another, within certain limits, and thereby raises the price of
the commodity for all of them above the market-price, while on the
other hand the sellers unite in trying to sell at a high price. If, vice
versa, the supply exceeds the demand, some one begins to dispose of
his goods at a cheaper rate and the others must follow, while the buy-
ers unite in their efforts to depress the market-price as much as possi-
ble below the market-value. The common interest is appreciated only
so long as each gains more by it than without it. And common action
ceases, as soon as this or that side becomes the weaker, when each one
tries to get out of it by his own devices with as little loss as possible.
(emphases added)
Here we find a clear echo of the argument used by the young Marx in
Wage-Labour and Capital. It is also clear that the price is not unique at
each moment of time: on the contrary, there is a constellation of prices at
each moment of time. This fact supports our claims that the process
needs to be analyzed under the assumption that buyers and sellers follow
mixed strategies instead of pure strategies.

4. Classical Competition and

Bertrand Competition
In the previous sections we outlined the classical notion of competition
with particular concern for market price determination. In this section we
try to answer the following question: Has the classical notion of competi-
tion percolated into modern theory? A positive answer to such a question
would allow us to make use of some recent results in order to extend the
analysis of market competition within classical economics.
As is well known, in 1883 the mathematician Joseph Louis François
Bertrand wrote a review of Léon Walras’s Théorie mathématique de la
richesse sociale (1883) and Augustin Cournot’s Recherches sur les prin-
cipes mathématiques de la théorie des richesses (1838).17 Bertrand was
highly skeptical of the then recent blossoming of mathematical econom-
ics. In particular, he poured scorn on Cournot’s book:

17. An English translation of Bertrand’s text, originally in Journal des Savants volume 48,
pages 499–508, is provided in the appendix of Magnan de Bornier 1992.
168 History of Political Economy 45:1 (2013)

[Cournot’s] formulae, written only in letters, bristle with unknown

functions; [Cournot] would consider it outside his field if he were to be
more specific. Practical economists must feel that it would be of little
value to study such formulae, be they true or false, so they escape from
this study by merely closing the book. If Cournot’s theory of wealth . . .
has failed to attract any serious attention over the past century, it is
because the ideas are lost under the profusion of algebraic signs. (quoted
in Magnan de Bornier 1992, 647)
As concerns Cournot’s duopoly model, Bertrand claimed that, pace
Cournot, it admitted “no solution under this assumption” (647), that is,
under the assumption that each duopolist tries to undercut the rival in
order to attract buyers and stops doing that only when he or she has noth-
ing more to gain from reducing his or her prices. (Note that for Bertrand
it is Cournot himself who assumed price competition between the two
sellers.) In short, the gist of Bertrand’s criticism is that Cournot failed to
acknowledge that the envisaged downward movement of prices was lim-
ited only by the marginal cost.18 A somewhat similar charge of indetermi-
nacy was raised sixteen years later by a distinguished economist, Francis
Ysidro Edgeworth, in a paper originally published in Italian in 1897 and
translated with some modifications into English in 1925. Edgeworth went
beyond Bertrand insofar as he claimed that the duopoly model admits a
continuous price cycle in the presence of diminishing returns:
[The case of two identical articles] is treated by Cournot as the first step
in the transition from monopoly to perfect competition. He concludes
that a determinate proposition of equilibrium defined by certain quanti-
ties of the articles will be reached. Cournot’s conclusion has been
shown to be erroneous by Bertrand for the case in which there is no cost
of production; by Professor Marshall for the case in which the cost fol-
lows the law of increasing returns; and by the present writer for the case
in which the cost follows the law of diminishing returns. In the last
case there will be an indeterminate tract through which the index of
value will oscillate, or rather will vibrate irregularly for an indefinite
length of time. (Edgeworth 1925, 117–18; emphasis added)

18. In Bertrand’s wording “without limits,” but this is only because in Cournot’s original
example the marginal cost is zero. This is not the place to discuss whether Bertrand’s interpreta-
tion of Cournot is well grounded or whether Cournot actually used prices instead of quantities
as strategic variables: see Magnan de Bornier 1992 and 2001, Dimand and Dore 1999, and Mor-
rison 1998 and 2001.
 Salvadori and Signorino / The Classical Notion of Competition 169

To defend this claim, Edgeworth produced a numerical example in

which two firms compete on prices, but they have capacity constraints.
Edgeworth showed there is no (pure strategy) equilibrium in his exam-
ple and formulated a sort of dynamic solution: firms undercut each other
until the price becomes so low that it is convenient for a firm to quote a
high price and sell only to the residual demand instead of undercutting
the price quoted by the rival. As Edgeworth wrote:
At every stage in the fall of price, and before it has reached its limit-
ing value . . . , it is competent to each monopolist to deliberate whether
it will pay him better to lower the price against his rival as already
described, or rather to raise it to a higher, perhaps the initial, level
for that remainder of customers of which he cannot be deprived by his
rival (owing to the latter’s limitation of supply). Long before the low-
est point has been reached, that alternative will have become more
advantageous than the course first described. (120)
With the development of game theory and its application to oligopoly
models, the so-called Bertrand-Edgeworth competition, explicitly based
on price undercutting, became a fruitful and extensively studied alter-
native to the price-taking competition embodied in the standard perfect
competition model (see Baye and Kovenock 2008). Yet the contemporary
Bertrand competition model is somewhat different from the original for-
mulation and from the classical notion of competition. First, it is a one-
shot game with a mixed strategy equilibrium and not a dynamic process
of price undercutting. This feature magnifies the relevance of two miss-
ing elements in Smith’s and Marx’s writings. First is the problem of firms
quoting the same price: in the case of a tie the firms fixing the lowest
prices must share total demand in one way or another. This requires the
introduction of a specific assumption in this regard. Second is the prob-
lem of how demand is rationed when a firm’s quantity demanded exceeds
its capacity. The introduction of a demand rationing scheme is another
assumption lacking in the classical authors.
However, a comparison between modern Bertrand competition and
Smith’s notion of competition, and even more the story told by Marx in
Wage-Labour and Capital, magnifies some deficiencies of the former.
First, in the former there is a multitude of atomistic buyers described
through a demand curve confronted with a given number of sellers, each
defined by their costs (generally marginal costs are constant and uniform)
and capacity. On the contrary, in Marx we find a sort of symmetry between
170 History of Political Economy 45:1 (2013)

sellers and buyers. It is certainly not difficult to extend the Bertrand com-
petition model to investigate the case in which there is a multitude of
atomistic sellers described through a supply curve confronted with a given
number of buyers, each defined by their reservation price (possibly con-
stant and uniform) and purchasing power. Generalization to a symmetrical
case in which a number of sellers with their costs and selling capacities are
confronted with a number of buyers with their reservation prices and buy-
ing capacities is certainly less obvious. Second, Bertrand competition
theorists have analyzed quite extensively the case of duopoly with constant
marginal cost. Few contributions have investigated the oligopoly (see De
Francesco and Salvadori 2010 and the literature cited therein). When there
are more than two competing firms, many changes are needed. In the case
of two firms, any firm can either undercut the other or avoid doing so. In
the case of three or more firms, any firm can either undercut all other firms
or just some of the firms and not others, or none of them. For example, in
the case in which there are two (either equal or not) large firms and a
smaller firm, the latter can avoid high prices so that the two large firms are
not interested in undercutting it when they undercut each other at higher
prices. The small firm can take advantage of this protection from competi-
tion with larger firms (at high prices) to obtain a larger rate of profit (see
De Francesco and Salvadori 2010, theorem 1(c)). The analysis provided in
the appendix may give an idea of the problems involved.

5. Final Remarks
In this article we attempted to assess the classical notion of free competi-
tion in comparison with the Walrasian notion of perfect competition. We
showed that the latter is plagued with some logical difficulties that drasti-
cally reduce its explanatory power to equilibrium situations. Such difficul-
ties are absent in the classical notion of competition, which, contrary to
the Walrasian one, is not based on any kind of price-taking assumption.
Yet the former also displays some unsatisfactory aspects. In particular,
while the classical authors extensively investigated long-period, natural
values and gravitation, they were more sketchy on market price determi-
nation in situations of market disequilibrium. To fill this lacuna we ana-
lyzed Smith’s and Marx’s views on competition between buyers and sell-
ers. We claim that, taking inspiration from the modern theory of Bertrand
competition, it is possible both to render Smith’s and Marx’s hints for-
mally precise and to provide interesting new questions for modern Ber-
trand competition theorists.
 Salvadori and Signorino / The Classical Notion of Competition 171

Appendix 1
Let N be the number of buyers. Their reservation price is pb. Let B1 ≥
B2 ≥ … ≥ BN be the different quantities of cotton they want to buy and
let B = B1 + B2 + … + BN. Similarly, let M be the number of sellers and c
their reservation price. Let S1 ≥ S2 ≥ … ≥ SM be the different quantities
of cotton they want to sell and S = S1 + S2 + … + SM. Let us assume that
c < pb (the case in which c ≥ pb requires more accurate analysis) and that
there are no buyers with a reservation price lower than pb and no sellers
with a reservation price higher than c (once again, a more accurate analy-
sis would be required otherwise).
If S < B2 + … + BN, that is, if all buyers but the largest one are will-
ing to buy more than the existing amount of cotton, buyer 1 could be
excluded from the purchase and, as a consequence, any other buyer 2,
3, . . . , N could be excluded too. Overbidding among buyers leads the
price of cotton to rise to pb. Therefore the best strategy for each buyer is
to express a demand for cotton at her reservation price and the best strat-
egy for each seller is to quote the price pb. On the contrary, if B > S >
B2 + … + BN, that is, if all buyers but the largest one are willing to buy
less than the existing amount of cotton even if all buyers are willing to
buy more than the existing amount of cotton, buyer 1 knows that she
will certainly buy some cotton. In the limiting case in which all buyers
2, . . . , N have purchased their desired amount of cotton, buyer 1 is a
monopsonist in relation to the sellers who have not sold their cotton. Let
pm be this monopsony price. Obviously, pm = c.19 Clearly, if buyers 2,
3, . . . , N quote price pb for cotton, buyer 1 will quote price pm < pb. But
if buyer 1 quotes price pm for cotton, (some of) the other N − 1 buyers will
outbid her instead of quoting pb. Therefore buyer 1 will overbid on them,
instead of quoting pm. And so on, until the price goes up so much that
buyer 1 will prefer again to quote price pm = c and buy only S – (B2 + …
+ BN ) units of corn. And so on and so forth. In this situation, sellers cannot
“stand opposed to the buyers like one man” and “fold their arms in philo-
sophic contentment”: they must fight each other to sell at a higher price.
Similarly, if B < S2 + … + SM, undercutting among sellers leads the price
of cotton to drop to c. Therefore, the best strategy for each seller is to quote
her reservation price and the best strategy for each buyer is to express a
demand for cotton at price c. On the contrary, if S > B > S2 + … + SM, no

19. Note that if buyers with a reservation price larger than c exist, the monopsony price is
larger than c.
172 History of Political Economy 45:1 (2013)

equilibrium price exists. Nor can an equilibrium price exist if S = B, since

S > B2 + … + BN and B > S2 + … + SM: the larger seller has a realistic pos-
sibility of selling part of her cotton at pb and the larger buyer has a realistic
possibility of buying part of the desired amount of cotton at c.
As a consequence, if B + S1 > S and S + B1 > B, both armies, to use
Marx’s metaphor, are engaged in infighting. The best strategies for both
buyers and sellers are not single (i.e., deterministic) prices but distribu-
tions of probability within a set of prices.
Determining such distributions of probability requires proper analysis,
which is beyond the scope of this article. Besides, to make such calcula-
tions we need to make further assumptions not stated by Marx. In particu-
lar, we need to know how the residual demand (supply) is determined for
sellers (buyers) quoting a price higher (lower) than that quoted by other
sellers (buyers) and what happens in the event of a tie. What is certain is
that buyer 1 will never quote a price lower than c (the monopsony price if
all other buyers are served) or a price higher than pMb, defined by the con-
dition that buyer 1 gets the same profit from buying min {S, B1} at price
pMb and buying S – (B2 + … + BN ) at price c. But, as a consequence, no
buyer will quote prices outside the range [c, pMb]. Similarly, seller 1 will
never quote a price higher than pb (the monopoly price if all other sellers
are served) nor a price lower than pmS, defined by the condition that seller 1
gets the same profit from selling B – (S2 + … + SM) at price pb and selling
min {B, S1} at price pmS. However, as a consequence, no seller will quote
prices outside the range [ pmS, pb]. Both arguments make sure that traders
can quote only prices that are in both ranges [c, pMb] and [ pmS, pb].

References
Arena, R. 1978. “Note sulla concezione classica della concorrenza.” Economia e
Lavoro, no. 2:323−52.
Arrow, K. J. 1959. “Toward a Theory of Price Adjustment.” In The Allocation of
Economic Resources, edited by M. Abramovitz, 41–51. Stanford: Stanford Uni-
versity Press.
Arrow, K. J., and F. H. Hahn. 1971. General Competitive Analysis. San Francisco:
Holden Day.
Aspromourgos, T. 2009. The Science of Wealth: Adam Smith and the Framing of
Political Economy. London: Routledge.
Baye, M. R., and D. Kovenock. 2008. “Bertrand Competition.” The New Palgrave:
A Dictionary of Economics, edited by J. Eatwell, M. Milgate, and P. Newman.
London: Macmillan.
 Salvadori and Signorino / The Classical Notion of Competition 173

Bellino, E. 2011. “Gravitation of Market Prices towards Natural Prices.” In Sraffa

and Modern Economics, edited by R. Ciccone, C. Gehrke, and G. Mongiovi,
58–75. London: Routledge.
Bharadwaj, K. 1978. Classical Political Economy and the Rise to Dominance of
Supply and Demand Theories. New Delhi: Orient Longmans.
Blaug, M. 1997. “Competition as an End-State and Competition as a Process.” In
Not Only an Economist: Recent Essays by Mark Blaug, 66–86. Cheltenham:
Edward Elgar.
———. 1999. “Misunderstanding Classical Economics: The Sraffian Interpretation
of the Surplus Approach.” HOPE 31:213–36.
———. 2001. “No History of Ideas, Please, We’re Economists.” Journal of Eco-
nomic Perspectives 15:145–64.
———. 2002. “Ugly Currents in Modern Economics.” In Fact and Fiction in Eco-
nomics: Models, Realism, and Social Construction, edited by U. Mäki, 35–56.
Cambridge: Cambridge University Press.
———. 2003. “The Formalist Revolution of the 1950’s.” Journal of the History of
Economic Thought 25:145–56.
———. 2009. “The Trade-Off between Rigor and Relevance: Sraffian Economics as
a Case in Point.” HOPE 41:219–47.
Bradley, M. E. 2010. “Adam Smith’s System of Natural Liberty: Competition, Con-
testability, and Market Process.” Journal of the History of Economic Thought
32:237–62.
Ciccone, R. 1999. “Classical and Neoclassical Short-Run Prices: A Comparative
Analysis of Their Intended Empirical Content.” In Value, Distribution, and
Capital: Essays in Honour of Pierangelo Garegnani, edited by G. Mongiovi and
F. Petri, 69–92. London: Routledge.
Dardi, M. 1983. “Piero Sraffa (1898–1983).” Quaderni di Storia dell’Economia
Politica 3:3–14.
De Francesco, M. A., and N. Salvadori. 2010. “Bertrand-Edgeworth Games under
Oligopoly with a Complete Characterization for the Triopoly.” http://mpra.ub
.uni-muenchen.de/18766/.
Denis, H. 1999. “Marx’s Polemics against Steuart.” In The Economics of James
Steuart, edited by R. Tortajada, 76–83. London: Routledge.
Dimand, R. W., and Mohammed H. I. Dore. 1999. “Cournot, Bertrand, and Game
Theory: A Further Note.” Atlantic Economic Journal 27:325–33.
Donzelli, F. 1990. “The Concept of Equilibrium in Neoclassical Economic Theory:
An Inquiry into the Evolution of General Competitive Analysis from Walras to the
‘Neo-Walrasian Research Programme.’” PhD diss., University of Cambridge.
Duménil, G., and D. Lévy. 1987. “The Dynamics of Competition: A Restoration of
Classical Analysis.” Cambridge Journal of Economics 11:133–64.
Eatwell, J. 1987. “Classical Competition.” The New Palgrave: A Dictionary of Eco-
nomics, edited by J. Eatwell, M. Milgate, and P. Newman. London: Macmillan.
Edgeworth, F. Y. 1897. “La teoria pura del monopolio.” Giornale degli Economisti
40:13–31.
174 History of Political Economy 45:1 (2013)

———. 1925. “The Pure Theory of Monopoly.” In Papers Relating to Political Econ-
omy, 1:111–42. London: Macmillan.
Garegnani, P. 1983. “The Classical Theory of Wages and the Role of Demand
Schedules in the Determination of Relative Prices.” American Economic Review
73:309–13.
———. 1984. “Value and Distribution in the Classical Economists and Marx.”
Oxford Economic Papers 36:291–325.
Giocoli, N. 2003. Modeling Rational Agents: From Interwar Economics to Early
Modern Game Theory. Cheltenham: Edward Elgar.
———. 2005. “Modeling Rational Agents: The Consistency View of Rationality and
the Changing Image of Neoclassical Economics.” Cahiers d’Économie Poli-
tique, no. 49:177–208.
Green, J. 1974. “The Stability of Edgeworth’s Recontracting Process.” Economet-
rica 42:21–34.
High, J. 2001. “Introduction: Split Personality; A Brief History of Competition in
Economic Theory.” In Competition, edited by J. High, xiii–xv. Critical Ideas in
Economics. Cheltenham: Edward Elgar.
Hollander, S. 1973. The Economics of Adam Smith. Toronto: University of Toronto
Press.
Kaldor, N. 1972. “The Irrelevance of Equilibrium Economics.” Economic Journal
82:1237–55.
Kurz, H. D., and N. Salvadori. 1995. Theory of Production: A Long-Period Analy-
sis. Cambridge: Cambridge University Press.
———. 2002. “Mark Blaug on the ‘Sraffian Interpretation of the Surplus Approach.’”
HOPE 34:225–36.
———. 2010. “In Favor of Rigor and Relevance: A Reply to Mark Blaug.” HOPE
43:607–16.
Lavezzi, A. 2003. “Smith, Marshall, and Young on Division of Labour and Economic
Growth.” European Journal of the History of Economic Thought 10:81–108.
Machovec, F. M. 1995. Perfect Competition and the Transformation of Economics.
London: Routledge.
Magnan de Bornier, J. 1992. “The ‘Cournot-Bertrand Debate’: A Historical Per-
spective.” HOPE 24:623–56.
———. 2001. “Magnan de Bornier on Cournot-Bertrand: A Rejoinder to Clarence
Morrison.” HOPE 33:167–74.
Marx, K. (1847) 1933. Wage-Labour and Capital. New York: International Publishers.
———. (1894) 1909. Capital. Vol. 3. Chicago: Charles H. Kerr.
Mas-Colell, A., M. Whinston, and J. Green. 1995. Microeconomic Theory. New York:
Oxford University Press.
McNulty, P. 1967. “A Note on the History of Perfect Competition.” Journal of Polit-
ical Economy 75:395–99.
———. 1968. “Economic Theory and the Meaning of Competition.” Quarterly
Journal of Economics 82:639–56.
 Salvadori and Signorino / The Classical Notion of Competition 175

Menudo, J. M., and R. Tortajada. 2009. “Double Competition and Market Stability
in Sir James Steuart.” WP ECON 09.06. http://ssrn.com/abstract=1408122.
Morrison, C. 1998. “Cournot, Bertrand, and Modern Game Theory.” Atlantic Eco-
nomic Journal 26:172–74.
———. 2001. “Magnan de Bornier on Cournot-Bertrand.” HOPE 33:161–65.
Ricardo, D. 1951. The Works and Correspondence of David Ricardo. Vol. 1, Prin-
ciples of Political Economy and Taxation. Edited by Piero Sraffa, with the col-
laboration of Maurice Dobb. Cambridge: Cambridge University Press.
Richardson, G. B. 1975. “Adam Smith on Competition and Increasing Returns.” In
Essays on Adam Smith, edited by A. S. Skinner and T. Wilson, 350–60. Oxford:
Oxford University Press.
Roncaglia, A. 1978. Sraffa and the Theory of Prices. Chichester: John Wiley.
Samuelson, P. A. 1978. “The Canonical Classical Model of Political Economy.”
Journal of Economic Literature 16:1415–34.
Semmler, W. 1984. “On the Classical and Marxian Theories of Competition, Value,
and Prices of Production.” Australian Economic Papers 23:130−50.
Smith, A. (1776) 1976. An Inquiry into the Nature and Causes of the Wealth of
Nations. Vol. 2 of The Glasgow Edition of the Works and Correspondence of
Adam Smith, edited by R. H. Campbell and A. S. Skinner. Oxford: Oxford Uni-
versity Press.
Sraffa, P. 1960. Production of Commodities by Means of Commodities: A Prelude
to a Critique of Economic Theory. Cambridge: Cambridge University Press.
Steedman, I. 1984. “Natural Prices, Differential Profit Rates, and the Classical
Competitive Process.” Manchester School 52:123–40.
Stigler, G. 1957. “Perfect Competition, Historically Contemplated.” Journal of
Political Economy 65:1–17.
Vickers, J. 1995. “Concepts of Competition.” Oxford Economic Papers 47:1–23.