Market efficiency is a concept used by economists to describe markets with certain theoretical characteristics.  For example, a “weak-form” efficient market is one where historical prices are not predictive of future prices, and therefore excess profits cannot be earned by using strategies based on historical pricing.  A “semi-strong” efficient market is one where public information is reflected in stock price to the point where it is impossible to earn excess returns by trading public information.

Market efficiency is also a legal concept, which, it must be said, only roughly tracks the economic definition.  In particular, in Section 10(b) litigation, an “efficient” market is one that absorbs information with sufficient speed and thoroughness to justify allowing plaintiffs to bring claims using the fraud on the market theory to satisfy the element of reliance.

The exact degree of speed/thoroughness that’s required for Section 10(b) litigation is something of a theoretical muddle (as Donald Langevoort has written extensively about) – although the Supreme Court’s recent Halliburton decision may provide more guidance on that (see discussion here and here).

For now, though, most courts try to assess “efficiency” by reference to what are known as the Cammer factors, taken from the case of Cammer v. Bloom, 711 F. Supp. 1264 (D.N.J. 1989).  These factors include weekly trading volume (with higher volume taken as an indicator of efficiency), the number of analysts reporting on the security, the number of market makers, and the cause-and-effect relationship between the disclosure of new information, and changes in the security’s price.  Some courts also consider additional factors such as the size of the bid-ask spread and the number of institutional owners.

These factors have frequently been criticized as duplicative or uninformative.  See e.g.,  Geoffrey Christopher Rapp, Proving Markets Inefficient: The Variability of Federal Court Decisions on Market Efficiency on Cammer v. Bloom and its Progeny.  Some commenters have speculated that a few of the factors are counterproductive, and in certain markets might indicate less efficiency.  See, e.g.,  William O. Fisher, Does the Efficient Market Theory Help Us Do Justice in a Time of Madness? (2005).

There’s a new paper that tests this claim, and concludes that the commenters are right, and at least some of these factors really are counterproductive.

[More under the jump]

In Arbitrage Risk and Market Efficiency – Applications to Securities Class Actions, Rajeev Bhattacharya and Stephen O’Brien argue that two of the Cammer factors – trading volume, and the number of market makers for a NASDAQ stock – are in fact correlated with less efficiency, and thus are being misused by courts.  Other factors, however, are useful, such as market capitalization, the size of the bid-ask spread, and the number of institutional owners.

The authors’ methodology is to assess the risks associated with arbitrage for a particular stock, on the theory that arbitrageurs are the ones who use public information to keep stock prices at their efficient levels.  When arbitrage risks are high, arbitrageurs will not function as effectively.  Therefore, high arbitrage risk should be associated with a lack of efficiency.  The authors then test the relationship between high arbitrage risk and the Cammer factors, to reach their conclusion that courts, basically, are doing it wrong.

I look forward to seeing whether their methodology and conclusions work their way into actual court decisions on this subject.