Markets are inefficient because they are efficient?

The Efficient Market Hypothesis (EMH) and market efficiency in general has been the subject of much debate in the aftermath of the sub-prime meltdown. Popular conception of EMH has it that asset prices cannot deviate significantly from its fundamentals, an untenable hypothesis given history and especially the recent events.

In this context, a more nuanced understanding of the concept of efficiency may help us appreciate the issue in its right perspective. Efficiency has two dimensions - market prices reflect all publicly available information about the fundamentals and the prices are fundamentally unpredictable. The popular inferences from these two dimensions are that they ensure that "prices are always right" and "it is impossible to beat the market on a sustained basis" respectively.

However there are enough reasons to believe that such highly simplified inferences may not be correct. It needs to be borne in mind that prices reflect only the "publicly available" information, and not "private" (or insider) information that is inevitable in such complex systems. Even more importantly, it also does not reflect the information about small and fleeting mis-pricings that are used by the likes of high-frequency traders. There have been even recent examples of wild market fluctuations largely attributable to the actions of such traders.

Further, as I have blogged about earlier, systemically too prices are extremely vulnerable to even small shocks administered through the random actions of noise traders/trading. Therefore deviations from the fundamentals will be a commonplace feature of equity markets.

In this context, Chris Dillow points to a new working paper by Brock Mendel and Andrei Shleifer where they claim that "rational but uninformed traders occasionally chase noise as if it were information, thereby amplifying sentiment shocks and moving price away from fundamental values". They show that even without large numbers of noise traders and significant sentiment shocks, a small numbers of these noise traders can have an impact on market equilibrium disproportionate to their size in the market. They write about how sophisticated but uninformed investors learn from prices,

"... such investors may entertain more complicated models and use other public information, such as bond ratings, in forming their demands... If ratings agencies usually do a good job of assessing the riskiness of bond offerings, it may be rational for uninformed traders to use these ratings as a rule-of-thumb to assess underlying value. On those occasions when the ratings agencies are wrong, this will induce correlated mistakes among the mass of uninformed traders, which will overwhelm the price impact of any better-informed traders in the market. It is only when the direct news about valuations reaches the uninformed investors that the market would correct itself. In this example, uninformed traders would rationally end up chasing noise thinking that it reflects information."

In other words, markets go wrong because sometimes rational but uninformed traders mistakenly believe that a price rise is driven by informed traders and not by noise traders and these noise traders, however small, occasionally carry enough momentum to tip the scales in favor of their trend. But as Chris Dillow rightly points out, this deviation from fundamentals "is only possible because very often prices really are right and do embody genuine information" and therefore "markets are occasionally inefficient precisely because they are so often efficient".

Interestingly, in view of the considerable range of non-fundamental information on market psychology (animal spirits of the market participants) that prices reflect, Rajiv Sethi has written that EMH be renamed as the invincible markets hypothesis (IMH)!

Update 1 (18/10/2013)

John Cassidy has a nice article that debunks the notion of efficiency with financial markets.