An Alternate View of Markets
He or she is taught that there exist microeconomic, and macroeconomic aggregate, supply and demand functions which codetermine the combination of real output and price for any particular good or service, and across the entire economy. He or she is taught that these forces push markets to equilibrium by draining surpluses through underproduction or price cuts -- or in the face of scarcity, expanding production or hiking prices. He or she is taught to think about markets as, in a word, orderly.
But what if that's wrong? Or, more precisely, what if the orthodox view of markets as self-organizing and equilibrating systems captures but the smallest sliver of their behavior? What if the vast majority of market behavior does not fit into such a model? What if the truth is that markets are ultimately more disorderly, more behavioral, more unstable, and more path-dependent than a primarily supply-and-demand framework allows us to understand?
Brittle, fragile, and chaotic is an entirely different characterization of markets, so much so that I find these the two visions difficult to reconcile. To get there, you may have to adjust your view of human nature -- we can't be, or maybe can't all be, intemporal optimizers in such a world. By teaching the orderly former so early on and so unequivocally, and to the near-total exclusion or extreme delay of the disorderly latter, it's worth asking if the emphasis on the supply-and-demand framework blinds economists, or if it leads them to dependably misperceive the economic systems they study.
It's worth asking, in fact, if economists have it backwards, if rather than living in a neoclassical world with some non-neoclassical phenomena happening in the footnotes -- imperfect information, money illusion, noise traders, loss and risk aversion, herding instinct, etc. -- we live in the non-neoclassical world with a limited amount of neoclassical phenomena sprinkled on top.
Basic behavioral intuition would seem to reinforce the notion that we are cognitively biased to see a neoclassical economy and confine the disorder to footnotes, considering the well-documented and broad tendency of humans to perceive ordered phenomena where there is none, or their persistent overestimation of confidence in order or pattern.
In some subfields of economics -- that is, besides the direct behavioral/cognitive/neuro research -- supply-and-demand is not seen as quite so regnant. I'm thinking, in particular, of the economics of exchange rates.
Here I find a rather close analogy to the view that supply-and-demand explains relatively little: the rejection of perfect and continuous purchasing power parity (PPP) hypothesis, which said that exchange rates should always adjust such that identical goods available in different countries cost the same amount, i.e. such that there is no room for further arbitrage trade.
Kenneth Rogoff, in a 1996 paper quoted in this PPP literature review in the Journal of Economic Perspectives, wrote: "While few empirically literate economists take PPP seriously as a short-term proposition, most instinctively believe in some variant of purchasing power parity as an anchor for long-run real exchange rates." That is, perhaps, a similar destination for this alternate view of markets: supply-and-demand "as an anchor for long-run" prices and quantities, hardly operating "as a short-term proposition." This dimmer view of PPP comes from a recognition of a variety of influences which consistently push foreign-exchange markets out of PPP equilibrium: interest rate differentials, real growth differentials, risk premia, all of which influence capital flows and trade balances.
It is not challenging, as a theoretical matter, to render a supply-and-demand equilibrium unstable through the addition of other forces. Let's look at one stylized models with some interesting, but I think plausible, assumptions.
Consider the possibility that price can influence demand -- more specifically, a microeconomic demand function which is determined to a significant extent by recent changes in price. When prices are rising, the entire demand function moves out; when prices are falling, the entire demand function falls back. There is, in other words, a component of demand which responds to the direction of price changes; in a traditional supply-and-demand model, the demand function is independent of price, and it is quantity demanded which is dependent upon price. In the case of the housing market, where demand is appears strongly governed by expected appreciation or depreciation of that asset, that modification of the model strikes me as reasonable, especially in the context of risky, limited arbitrage. Under these assumptions, partial equilibria are unstable, sensitive to movements in prices, and the market as a whole looks much like our brittle-fragile-chaotic story, rather than self-ordering and equilibrating.
Without implicating him in any way, I thank Miles Kimball for helpful comments and suggestions via email in advance of this post.
Update (8/4/12): Noah Smith makes a similar point in an old post, writing that:
[t]he whole notion of thinking of each interesting feature of the economy as a "friction," and then of considering only one or two "frictions" at a time, has been very detrimental. For one thing, it makes it hard to develop a useful model of the economy, since the actual economy contains many, many "frictions" (so many that the "frictions" together are usually more important than the "frictionless" dynamics that supposedly "underlie" them). Also, the "one friction at a time" approach makes it very difficult to generate any alternatives to the classical "core theory" of Walrasian general equilibrium.