|
|
|
September 21, 2006
|
|
The Randomness of Efficient Markets (part 2)
|
|
In search of the holy grail
|
Last week we asked whether the financial markets are so random as to make it impossible to earn excess returns on the basis of past performance? Or whether the market is so efficient that investors can never make money on mis-pricing opportunities, because those opportunities never really exist.
There are many arguments against the latter point. The mis-pricing that occurred during the 1990 tech bubble, or perhaps today, as we near the end of a real estate bubble in the US, and perhaps even, a commodity bubble in Canada.
That left us with a question that has no simple answer. Efficient market theorists will explain away bubbles by arguing that efficient markets eventually correct excesses. The disconnect in that position is how one defines “eventually.” If nothing else, it introduces time – and how one defines time - as the one constant that closes the gap between those who believe in efficient markets and those who do not.
If excesses are “eventually” corrected, how long do excesses exist? If one believes excesses create opportunity, timing is also an issue, because the opportunist must calculate a point at which it is appropriate to make a trade.
And for both sides, there remains a question as to whether one can actually take advantage of excesses. Can one successfully buy or sell markets, sectors or stocks that are driven to excess because of irrational behavior? Certainly the stats do not provide any support for that view.
In my opinion, markets are efficient in the sense that it is difficult to manage through excesses with any success. I also believe that efficiency for the market, sectors and stocks exist as a range of prices rather than a specific number, which means that the price on any given day has more to do with noise and randomness.
A range of prices provides an efficient channel if you will, with the upper and lower band defining excesses, and the current price being just noise (i.e. randomness). That approach would work for active managers tying to find entry and exit points, but still does nothing to help one time the market.
But assuming a “normalized” channel is still just the beginning. If you believe as I do that financial markets react to things we know, things we think we know and things that are unknowable, then there is always the possibility that the top and bottom of the channel will change because of an unknowable event. And like it or not, unknowables are what induces significant changes to the value of individual stocks and, to a lesser extent, sectors and the market.
Think about that for a moment. Inco (symbol N, traded TSX) was a recent example of a stock that traded in a range based on what the market knew or thought it knew about an equilibrium price – i.e. a price that matched supply with demand – for nickel. The fair value for Inco based on that analysis was somewhere between $50 and $60 per share.
Now add into the equation an unknowable. In this case, a string of offers and counter-offers from companies interested in buying Inco. The impact on the company’s stock price to offers that could not have been known ahead of time, was significantly greater, than any impact the market’s assessment of fair value had on the basis of what it knew or thought it knew about the price and demand for Inco’s core product.
Having said that, we come full circle back to the portfolio. When investors, working with advisors or one their own, construct a portfolio, they are really building a model to manage risks associated with events that are not predictable, and that give rise to excesses, driven by irrational behavior.
A well constructed portfolio should perform on the basis of things we know. For example, we know that equity assets grow over long periods, because equities in their purest sense, mirror growth in a capitalist economy. We know that capitalist economies – as measured by the GDP - grow over time because if they did not, no investment including bonds, real estate and cash, would have any long term value.
We build portfolios on the basis of the individual investors long term objectives and their ability to tolerate short term risk. We diversify to manage the risk associated with unknowables, by diversifying across asset classes (reduce risk within the market), across geographic regions, sectors, and investment styles (i.e. passive indexing or active value and growth styles).
|
|
