"By "market" is meant the entire complex of institutions which people buy and sell and hire and are hired and borrow and lend and trade and contract and shop around to find bargains"
(Thomas C. Schelling, Micromotives and Macrobehavior, 1978: p.23).
"In fact, the whole world may be looked upon as a vast general market made up of diverse special markets where social wealth is bought and sold. Our task then is to discover the laws to which these purchases and sales tend to conform automatically. To this end, we shall suppose that the market is perfectly competitive, just as in pure mechanics we suppose to start with, that machines are perfectly frictionless."
(Léon Walras, Elements of Pure Economics, 1874: p.84).
(A) Neoclassical Economics as Catallactics
(B) Exchange Processes
(C) Markets and Commodities
(D) The Meaning of Perfect Competition
(A) Neoclassical Economics as Catallactics
Economics, Alfred Marshall tells us, "is a study of mankind in the ordinary business of life; it examines that part of individual and social action which is most closely connected with the attainment and with the use of the material requisites of well-being." (Marshall, 1890: p.1). Unfortunately, this mild and uncontroversial statement is not very informative. Other social sciences -- anthropology, sociology, political science, history, psychology, etc. -- to a greater or lesser extent, also study "mankind in the ordinary business of life". If one sought to pinpoint the feature that distinguishes economics from the other social sciences, one would have to be a bit more precise.
One proposition is that economics, in particular Neoclassical economics, studies the phenomenon of equilibrium exchange, or, as Richard Whately (1832) called it, catallactics. The adjective "equilibrium" is included here without apology and captures the idea that exchange is undertaken only when there is an agreement among participants as to what that exchange should be.
The characterization of economics in general as the "science of exchange" is somewhat of an exaggeration, but as a characterization of Neoclassical economics, it is no exaggeration at all. The quintessential feature of Neoclassical economics, above everything else, is that it addresses all economic phenomena in the same manner: it first reduces the problem to one of pure exchange, and then searches for the equilibrium exchange ratio.
All branches of Neoclassical economics feed into or stem from this central procedure. They can all be classified according to whether they seek to (1) characterize the agents that will be involved in the exchange; (2) characterize the process of exchange itself; (3) characterize the results of the equilibrium exchange. Thus, we see that consumer theory and producer theory fall into category (1). General equilibrium theory, the theory of the core, etc. fall into category (2); capital theory, distribution theory, growth theory, etc. all fall into category (3). Exchange is the Neoclassical forest, the rest of Neoclassical economics is but trees. Equilibrium exchange, then, becomes the Procrustrean bed of Neoclassicism: if a particular phenomenon is to be considered "economics", it must be reducible to equilibrium exchange; if it is not thus reducible, it is not economics.
(B) Exchange Processes
If Neoclassical economics is catallactics, then category (2), the study of the process of exchange, should be the foundation, the heart, the core of Neoclassical theory. Alas, in this respect, Neoclassical economics has been somewhat narrow. In principle, equilibrium exchange among individuals can be achieved in manifold ways: via bilateral bargaining, via auctions, via market specialists, via the medium of abitrageurs, via the haphazard absorption of all improving trades between any people or groups, via coalitions deciding upon aggregate allocations, etc. However, for most of its history, Neoclassical economics has cut out all these options and focused on only one process of exchange: the price-mediated process.
A price-mediated exchange process is what is known popularly as "supply-and-demand" theory. The process is simple: when prices are announced by "the market", agents choose how much they wish to buy of each good (demand) and how much they wish to sell (supply). If prices are equilibrium prices, then the market-wide demand for each good meets the market-wide supply of each good and thus "equilibrium exchange" is said to be achieved. If demand is not equal to supply, then the prices were wrong. What usually follows is that, before any exchange is undertaken, another price is offered up by the market which corrects the discrepancies in demands and supplies (the Walrasian tatonnement process) or agents try to exchange whatever they can at these wrong prices but adjust their behaviors over time by, say, changing expectations or moving out of low-price to high-price activities so that prices eventually come to equilibrium (non-tatonnement processes, including the Marshallian).
The Neoclassical exchange process is thus price-mediated in the sense that there is a "price" that is "given" by the market and which the agents, effectively, treat parametrically. But not all exchange processes are price-mediated: e.g. Edgeworthian coalitional exchange, bilateral bargaining, etc. In price-mediated processes, agents face "the market" which intermittently spits out prices at them which they react to; they demand goods from and supply goods to "the market". In non-price-mediated processes, agents face "each other" and try to trade with "each other" by offering, accepting or rejecting whole series of possible trades with "each other".
This does not mean that these alternative exchange processes don't have prices at all. If we define prices as the ratios of goods exchanged, then all exchange processes end up with prices, i.e. exchange ratios between goods. As Carl Menger reminded us a long time ago, "[p]rices are only incidental manifestations of these activities [exchange], symptoms of an economic equilibrium between the economies of individuals." (Menger, 1871: p.191). This is as true of non-price-mediated as well as price-mediated exchange processes.
The crucial difference is that alternative exchange processes argue that during the exchange process, before all final exchanges have gone through, there are either (1) no prices, but rather disparate bids and offers of exchange or (2) piecemeal exchanges, so that the same goods will be exchanged at different prices simultaneously or over time. This is why these alternative processes are not "price-mediated", even if they are "price-resulting".
The reason for the intense and exclusive Neoclassical focus on the price-mediated exchange priocess is not mere shortsightedness on the part of the Neoclassicals. At least three justifications can be given for it, which we classify and label accordingly.
(1) Old Austrian-Chicago: gathering information about exchange opportunities with other agents is expensive; the prices posted by "the market" convey all the necessary information anyway, thus people will naturally choose a price-mediated process. (cf. Hayek, 1937, 1945; Friedman, 1962)
(2) Lausanne: the exchange process is very complicated, and exchange opportunities are always being exploited by armies of arbitrageurs and competitors, thus individual agents end up having no choice but to take prices as given and thus exchange according to a price-mediated process (cf. Walras, 1874: p.160; Pareto, 1906: p.245)
(3) Edgeworthian: the exchange process can be what it may be, but there are certain, perhaps idealized, scenarios (e.g. when there are infinitely many participants), where that exchange process achieves effectively the same result in the end that a price-mediated process would achieve. Thus analyzing price-mediated processes is a simple, quick way of getting to the solution analytically, even if the "real" underlying process is different. (cf. Edgeworth, 1881; von Neumann and Morgenstern, 1944: p.13-4).
These are the three popular defenses for analyzing price-mediated exchange. The crucial points are noted in italics: either agents actually choose price-mediated exchange or they are forced into price-mediated exchange, or economists simply prefer it as an analytical device. These justifications are not without their disputants: for instance, Maurice Allais (1989), Douglas Gale (1985), M.J. Osborne and Ariel Rubinstein (1994), Duncan Foley (1994) and Louis Makowksi and Joseph M. Ostroy (1998) and modern Austrian economists (stemming from Ludwig von Mises (1949) through Israel Kirzner (1973) to today) have given excellent reasons for not ignoring non-price-mediated exchange processes and offered means by which to make them tractable to analysis. However, as so much of the Neoclassical enterprise hinges on the price-mediation assumption, economists have been quite reluctant to follow these pioneering trails.
[To be completed later]