This is a discussion of Harold Demsetz's well-known paper on "Competition". Demsetz is one of the luminaries of the "New Institutional Economics" that has received much attention from orthodox economists because it seeks to supplement what are evident "gaps" in neoclassical theory (which we have tried to expose on this site). The Nobel Prize Committee, sensing the importance of this intellectual "prop" to the deficiencies of orthodox bourgeois theory, has well rewarded these "institutional economists" by awarding Nobel Prizes in Economics to the likes of Ronald Coase, Douglass North and lately Oliver Williamson. My notes refer to this Demsetz paper:
Clearly, as we have seen in the chapters on Hayek and Schumpeter, it was their attempt to overcome the insuperable apories of neoclassical theory and of the earlier Austrian tradition that led to their “evolutionary” path, though from different premises.
Equilibrium prices cannot be the result of “perfect competition”, because the economic agents that it describes are perfectly “self-interested in-dividuals” whose “individual endowments” and “utility preference schedules” in terms of the “exchangeable endowments” of every other economic agent must be known by all simultaneously for the relative prices of all exchangeable endowments to be determined by means of simultaneous equations in terms of a ‘numeraire’ (a weighted unit of account) Consequently, “competition” can play no role in the determination of equilibrium prices (those at which all exchange transactions clear and the demand and supply schedules of every agent are maximally satisfied) because the agents need not engage in any “competitive activity or decision” of any description. The “competition” is void of any substantive content and is implicit (tautologously) in the status of each agent as a “self-interested (atomized) in-dividual”.
As Demsetz (following Hayek) rightly points out, because “competition” is implicit in the definition of “self-interested in-dividuals”, economic analysis paid scarce attention to “competitive activity” and ultimately had to resort either to “the invisible hand” or to the empty, “frictionless” and tautologous simultaneous equation of “given” utility schedules among “self-interested in-dividuals” in “perfect competition” or, as Demsetz calls it, “perfect decentralization”, to achieve “equilibrium”. In both cases we have a Leibnizian “pre-established harmony” – the object of Hayek’s attack.
Demsetz skirts around the crux of this problem – the nature of prices! But he never gets to the heart of the matter. For him, the “friction” and “details” of “competitive activity” are relevant only to “the short run” – because they do not affect the determination of “natural prices in the long run” where the assumption of “perfect competition” is legitimate.
Williamson does something analogous by distinguishing four levels of analysis of price co-ordination from “spontaneous, non-conscious” (recall Hayek’s “spontaneous order”), “embedded” institutions (a relic of the Iraqi war?), through to transaction costs and governance regimes, to the “instantaneous” resource allocation and utilization of neoclassical theory. Of course, it is a perfect mystery to divine how these mythological layers of analysis (one is reminded of Auguste Comte’s “positivist stages of civilization”!) will ever lead us to the determination of “prices”.
Indeed, once again, (we applied this critique to Hayek’s analysis) this entire fantastic exercise is possible if and only if we imagine “prices” as “relative prices” referring to a common ‘numeraire’ fixed simultaneously, synchronically and therefore “timelessly” in a marginal “exchange” of incommensurably subjective “utility schedules” (metaphysics indeed!). Were it otherwise, all the “institutions” in the world, no matter how “embedded” or “spontaneous”, could never reveal to us the “prices” that can “co-ordinate” the market economy at any given time! Indeed, as with Hayek, we could only declare that the economy is “co-ordinated” at any given time by simply “decreeing”, by “postulating” that it is so a posteriori, ex post facto! But this “declaration” would amount to an empty, gratuitous and farcical “rationalization” of the status quo!
We will soon reveal the real purpose behind the readiness of “the new institutional economics” to preserve the whole edifice of equilibrium analysis. But it is far more instructive and revealing to follow the iter intellectualis of this approach from the more peripheral premises to the core conclusions.
At the more peripheral level, far from challenging its entire theoretical basis, Demsetz confines himself to a remedial criticism of neoclassical theory aimed at improving its effectiveness. He does this for a dual purpose: firstly, it allows him to leave the entire ethereal and tautological edifice of general equilibrium analysis intact for what concerns the “determination of ‘natural’ prices”, which “the new institutional economics” otherwise would not be able to replace. Secondly and as a direct result, Demsetz perceives quite rightly that without the validity of equilibrium analysis he would not be able to embark on the entire apologetic, rationalizing expedition called “new institutional economics”. Read carefully:
“Part of the difficulties of competition come from our attempt to use the perfect competition model for a purpose for which it is not ideally suited – for analyzing competitive activity. However, the perfect competition model…is required [necessary] to understand the functioning of the price system in a decentralized economy,” (p4).
Of course, if we leave production out of our analysis and we consider only exchange by self-interested in-dividuals of their existing endowments, then the optimal co-ordination of exchange activities must occur in a context where the endowments offered in exchange are in quantities that take into account simultaneously all the exchange plans advanced by all the exchange participants. As a result, as Hayek adroitly pointed out, any kind of practical “competitive activity” is axiomatically excluded from equilibrium analysis! Demsetz is quite wrong to opine, then, that
“[s]tripped of competition, economics would consist largely of the maximizing calculus of an isolated Robinson Crusoe economy” (p1).
Because that is exactly what “equilibrium analysis” is! Far from “competition [being] required to understand the functioning of the price system in a decentralized economy”, equilibrium analysis is tautologous precisely because “competition” is synonymous with “decentralization” (those “self-interested [atomistic] in-dividuals”) so that the former can never explain “the functioning of the price system” or of anything at all! Equilibrium analysis does not contain any substantive or concrete (“real”) notion of “competition” at all. As Hayek and Demsetz himself properly perceive, “competition” is nothing outside of concrete “competitive activities”, which are not defined or (Hayek) even possible according to the axioms postulated by the analysis. Yet Demsetz tries to salvage the theoretical relevance of equilibrium analysis by introducing for it the much narrower concept (indeed an apory, a non-entity!) of “pure price competition”:
“[Equilibrium analysis] determines how the ‘natural’ prices of long-run equilibrium integrate decentralized economic decisions; these natural prices presumably are unaffected by time, uncertainty, and the ‘frictions’ of the short-run. This is accomplished by developing a very special notion of competition – call it perfect price competition,” (p7).
Demsetz is wrong again. The dichotomy that he seeks to draw between long-run natural price equilibrium and short-run “frictions” of perfect price competition is simply inoperable – because the long run is always a succession of short runs so that the conceptual difficulties of the one invariably flow on to the other! If “the natural prices of long-run equilibrium” are obtained “by developing [the] notion of…perfect price competition”, and given that this latter notion is entirely devoid of any “competitive activity” – that is, of any real, concrete socio-historical content or substance -, how on earth can we then accept or conclude that “equilibrium analysis determines the ‘natural’ prices of long-run equilibrium”? How can a supposedly scientific analysis adopt abstractions that have no practical real content but are instead as we have shown tautologous, aporetic and antinomic concepts?
For even when “competition” is given a substantive practical content, as “competitive activity”, it remains an antinomic concept, even more so than before – because the aim of competition is to eliminate competitors, so that its end-goal is the abolition of “competitive activity” and the installation of “monopoly”. It is incorrect then to say, with Demsetz (p4), that “the self-interest” of each participant “limits” the ability of others to reduce “competition” and thus helps maintain price discovery through “decentralized” co-ordination: -
“[Smith] was able to show that these [self-]interests are disciplined by each other and that they are linked sensibly through prices,” (p4)
In theory and in practice, instead, if the goal of all “competitors” is, by definition, to out-compete everybody else, the practical end result will be the end of the competitive regime! (Cf. Arendt’s critique of Hobbes’s political theory in these terms.) “Perfect competition”, already an “empty” concept in equilibrium analysis, now becomes an “antinomic” one because there is no concrete activity that, once permitted in reality, will not result in its antithesis – monopoly or “perfect centralization” (to echo Demsetz).
Conversely, ‘monopoly’ is in ‘antinomic’ relation with ‘competition’ (like ‘being’ and ‘nothing’ in Heidegger). Demsetz brilliantly re-defines the two in terms of “freedom and coercion” (discussed in our ‘Hayek-Robbins’).
But surely the greatest difficulty is one that is essential to isolating the central problem of economic inquiry: if the aim of competition is to out-compete all other competitors, this only begs the question: what is the goal and purpose of out-competing all other competitors? What, in other words, is the “final aim” of “competition” or “competitive activity”? Obviously, it is the acquisition of “wealth” defined as “value”, that is as “socially pro-duced wealth”, as against freely available natural wealth, not for its use value, but rather as a means for the pro-duction and accumulation of more socially produced wealth, that is, for its exchange value.