PsyFi Search

Saturday 7 August 2010

Tâtonnement: Groping for Stock Equilibrium

Old Saws, New Rocks

One of the oldest saws in the book of economics is the idea of supply and demand; it even pre-dates Adam Smith, with a legacy stretching back to Muslim thinkers of the eleventh century. If people demand the Pet Rock as the next must-have toy but supply is limited then prices of Pet Rocks will go up. If some enterprising rock counterfeiter then floods the market with a supply of good-enough fakes then the demand is likely to fall, and price with it.

This iron rule of economics is actually a bit less rigid than you might think. In fact, it’s rather too wibbly-wobbly to describe it as a rule at all. But at the margins it more or less works which means it’s rather curious that it’s rarely cited as a reason for long-term changes in stock prices. It stands to reason, though, that if lots of people decide they want to buy shares just as companies start to withdraw them from the market that prices should go down. Or does it?

Partial Equilibrium

Supply-demand models are based on so-called partial equilibrium, the idea being that you can look at the market for some goods or other independently of all other variables. This certainly can’t be true most of the time – it’s like working out what the correct price is for lighters without taking into account the availability of matches. Or Boy Scouts with a handy supply of dry sticks and kindling.

Linked to this is the concept of elasticity. If the quantity demanded changes a lot when there’s only a small change in price then it’s elastic. If the quantity demanded changes little when prices change a lot then it’s inelastic. So the market for bread tends to be inelastic because changing the price of it doesn't markedly change the amount consumed. As you might have twigged, many economic theories of the stockmarket assume that stocks are inelastic . If this wasn’t the case then stuff like the Efficient Markets Hypothesis wouldn’t be even vaguely possible.

Pareto Groping

Now the idea that markets are efficient and prices reflect all known information somewhat begs the question about how this occurs. This rather fundamental problem has perplexed generations of economists who’ve come up with a concept to explain it known as tâtonnement which, joyously, translates as “groping” or, as non-economists say, "trial and error”. Basically people fart about guessing until things settle down at the equilibrium point, where supply and demand is matched.

At this equilibrium point the market is presumed to be Pareto-efficient. Any situation is Pareto-efficient if there’s no way of changing it without making at least one person worse off: it’s often used in analysis of welfare situations for hopefully obvious reasons. So at the equilibrium point, where the quantities of supply and demand are in balance the quantity produced is predicted to be Pareto-efficient.

False Trades

The trouble is that this general groping around implies that there’s a fair amount of trading at prices which, technically, are wrong. These false trades will, be definition, happen at the wrong price and, as Neal H. Buchanan describes in How Realistic is the Supply/Demand Equilibrium Story?, there’s no guarantee at all that the market will fix itself:
“The assumption of simple supply-and-demand equilibrium can mask a fundamental problem inherent in the standard approach to economic equilibrium. At any given moment we cannot assume that trades are made at equilibrium prices nor that markets “tend to” any useful conception of equilibrium. Efficiency is not guaranteed and the process of convergence to equilibrium is inherently open-ended.”
Which implies that the ideal of the Pareto equilibrium can't be achieved which means, in essence, that the amount of goods produced are sub-optimal. More – or less – goods are produced than the market really needs. There is no guarantee that the operation of the market will meet the needs of the market. Wave goodbye to the invisible hand.

Supply-Demand Failures

Buchanan’s analysis even extends to assuming that the Law of One Price actually operates – a situation we’ve already seen isn’t true in financial markets – and still shows that the market equilibrium, if it exists, isn’t guaranteed to be at the point efficient market theories insist it should be. The law of supply and demand doesn’t obviously work.

Is this, then, the explanation for why the supply and demand characteristics of stocks doesn’t often appear as an issue in market movements? Well, maybe. Carl Hopman in Are Supply and Demand Driving Stock Prices? analysed limit order data to try and tease apart the various factors. Because not all limit orders are satisfied this is a proxy for an imbalance between supply and demand and this imbalance is highly correlated with stock returns on all time frames up to three months.

So, supply and demand does seem to drive stock returns. However, Hopman goes on to show that 70% of the order flow appears to be driven by uninformed investors who, presumably, are operating on purely behavioural grounds. Given, as we’ve seen, there’s no obvious mechanism for driving valuations back to an equilibrium point it’s entirely possible – and entirely theoretical – that these micro-economic tendencies could cause bubbles and various other troubles.

Stock Retirement

There’s another level, however, at which we can consider supply and demand in the context of the stockmarket. Consider, for instance, the oft-documented tendency of companies whose CEO’s are heavily rewarded with stock options to retire stock from the markets through buybacks rather than rewarding shareholders with increased dividend flows. As we’ve seen, such behaviour improves the returns for option holders at the expense of shareholders but it may have an even more invidious effect at the gross stockmarket level: if stocks are subject to an elastic demand curve then reducing their availability may push prices substantially higher; fertile grounds for stock over-valuation.

There isn’t much evidence either way but that which there is suggests that the demand curve for stocks is actually inelastic. When China removed the restrictions on A-shares in 2005 and allowed them to be freely traded the supply shock that this caused would rationally have been expected to cause prices to fall. In fact, as Li, Liao and Shen show in An Inelastic Demand Curve for Stocks: Evidence from China’s Split-share Structure Reform, the opposite happened – stock prices went up, with an abnormal return of, on average, over 13%.

EMH Reborn

Which, all in all, suggests that the Efficient Market Hypothesists have got something right. However, doubts persist. The other effect of the reforms was to remove a whole series of agency issues to do with the concentrated ownership of these stocks in the hands of a group of people with no especial interest in shareholder democracy. The researchers argue that the differences in ownership across the newly listed firms show that these agency effects were dwarfed by the impact of price inelasticity and that the Efficient Market Hypothesis holds.

The problem here, of course, is that most of the evidence points the other way and a behavioral explanation of this price change seems more likely. Generally, though, we’d better hope that the researchers are right. Consider how the growth of the Cult of the Equity has moved individuals into investing in the stockmarket. As net flows of cash stream into the market an obvious reaction would be for more stock to be supplied to meet this demand. However, at the level of the individual corporation the opposite tends to be happening as stock is removed from the market through buybacks and takeovers. The listing of new stock has been substantially less than the retirement of existing stock for many years, even as the demand for stocks as investment vehicles has soared.

Boom and Bust, Supply and Demand

So it’s plausible – and completely unproven, as far as I can discover – that it’s been the changes in supply of stock coupled with the increase in shareholder demand that have driven stockmarket returns over the last few decades. This article by John Oswim Schroy sets out the case. Meanwhile onset of mass retirement amongst the Baby Boomers is moving the goalposts somewhat, even as we speak.

For once stockmarket investors should hope that adherents of the Efficient Markets Hypothesis are correct. The alternative could be a very long and slow decline.

Related articles: Volatility, the Last Anomaly, Buyback Brouhaha, Finance: Where the Law of One Price Doesn't Apply


  1. I have a hard time seeing how the laws of supply and demand could work in a real way in the stock market. The thing that investors are seeking is not shares but dollar value. Regardless of how many shares are outstanding, dollar value can be (at least temporarily) whatever the investors want it to be. If they don't like the current price, they can just bid it up (only collectively, of course, not individually).

    That reality takes us (it seems to me) from a real world of limits (the supply and demand world) to a fantasy world of no limits (in the short term -- and price crashes when the Reality Principle asserts itself in the long run).


  2. "It stands to reason, though, that if lots of people decide they want to buy shares just as companies start to withdraw them from the market that prices should go down."

    is that correct?