Daily Speculations

The Web Site of Victor Niederhoffer & Laurel Kenner

Dedicated to the scientific method, free markets, deflating ballyhoo, creating value, and laughter;  a forum for us to use our meager abilities to make the world of specinvestments a better place.



Write to us at: (address is not clickable)

Dr. Alex Castaldo

Price Elasticity

In 1957 the Dupont Corporation was found guilty of violating the Clayton Antitrust Act because it owned 23% of the stock of General Motors and was ordered to immediately sell the stock. Lawyers had strongly argued against this potentially ruinous move: isn't the price of GM going to drop by 23% or more as a result, damaging all stockholders? Where are we going to find someone willing to buy such a huge block?

As it turned out, these concerns were completely unfounded and the sale was absorbed by the marketplace with nary a ripple on the price (neither at the announcement nor the actual sale).

As a result of this and other episodes, the idea gained ground that stock market liquidity is for all practical purposes infinite, or as the Chair put it, the demand curve for stocks is horizontal (Price on the y axis, Quantity on the x axis).

The most important proponent of this view was Myron Scholes in his article "The Market for Securities: Substitution versus Price Pressure and the Effects of Information on Share Prices" (1972). IN THE ABSENCE OF ANY NEGATIVE INFORMATION ABOUT THE STOCK a sale would not affect the price. The fact that DuPont was forced to sell by an order of the U.S. Supreme Court guaranteed that DuPont had no negative information about GM and made it very attractive for any pension fund to buy GM for 1/8 below the market, selling Ford or other securities if necessary to raise the cash. All stocks were perfect substitutes and the liquidity of the entire US capital markets was available when called upon.

Academics widely accepted this theory, although eventually (as with all academic theories) they found some exceptions and qualifications. The first exception was found by Shleifer and others in 1986 when they discovered that stocks added to the S&P 500 generally experience positive abnormal returns following the announcement. It seems clear that this announcement contains no new information about the company whatsoever, so under Scholes' theory it should cause no price change. Forced buying from index funds responding to Standard & Poor's should be met by other investors glad to sell for a penny above market and switch into other (S&P or even non S&P) stocks. It is exactly the Dupont/GM story in reverse, with S&P playing the role of the Supreme Court; an outside entity orders a transaction for non-economic information-neutral reasons, but this time the theory does not work. (Formally, a theory can be rejected as soon as one small exception is found, but many people do find this S&P effect rather small and insignificant and continue to believe the horizontal curve theory is mostly true.

Can other examples be found where a stock price changes purely in response to quantity demanded in the absence of any information? The future Nobel winners are working in this).

A second qualification arises in the short term. Economics allows that markets may be less elastic in the short term than the long term; it takes some time to find substitutes. Academics who study the behavior of prices from transaction to transaction call themselves Market Microstructure theorists and they do find (not surprisingly) that stock market liquidity is not infinite in the short run (they look at periods of a few seconds or a few minutes). In fact they have measured a parameter they call "Kyle's lambda" which measures the impact on price of a demand of one share. They find that it is not zero, that it is different for different stocks, that it changes throughout the day, and generally that it behaves sensibly as a measure of liquidity.

But Market Microstructure does not tell you anything about stock valuation, only about transaction to transaction price response. It doesn't tell you what stocks to buy, but it might suggest that if you want to buy a large block you might do it near the open or near the close when Kyle's lambda is lower. (Or you might use some of the fancy trading algorithms that Mr. Haag talks about). In these theories the ultimate value of the stock is still determined by information, not quantity.

See Victor Niederhoffer's post on price elasticity>>>

Go to more posts by Alex Castaldo >>>