How and why you can beat the market part one: irrationality

Thursday, 2nd July 2009

I promised (on both a comment on Interactive Investor and an earlier post here) to explain why I think it is possible to identify areas of market — or how to tell when every one else is wrong.

An efficient market requires that investors have some commonality in their view of what gives an investment value. In practice this means that we assume that investors are rational, risk-averse, wealth maximizers. I will talk about irrationality for now, and deal with the others in future posts.

It is fairly evident, and generally accepted that herd instinct and “fear and greed” play their parts. Few people “can keep their heads while all about are losing theirs”. So far, so boring, but what are people thinking when they lose their heads, and how can we spot irrational valuations.

We can answer the question, because one thing that happens during bubbles are arguments between the sceptical and the herd. There are definitely patterns to the bull (in more ways than one) arguments.

One line of argument is that normal valuation methods (e.g. NPV, the usual valuation ratios, etc.) are irrelevant. This was very evident during the dot com boom when analysts resorted to valuing companies by website traffic and other such measures that were far removed from the ability to generate cash. In the case of property the discrepancy between the NPV of the rental value (even on optimistic views of future rents) and sale prices was largely ignored. N.B. reference to property here are primarily about the UK housing market.

If you cannot justify the price on the basis of the cash flows you forecast, then it is mispriced.

Even more characteristic justification of bubbles is the argument that things are different this time. The dot com boom arguments were varied, including a permanent shrinking of the equity risk premium.. There was no plausible reason advanced for thinking that the change was permanent, and without that it was just a fancy way of saying that the market was expensive by historical standards. What should have been a warning to sell, was twisted into a reason to buy.

The housing boom produced similar arguments. One was that housing was more affordable because of lower interest rates. This was true to an extent, but it is offset by the matching low inflation that meant that debt was no longer eroded by inflation. Another was that immigration would generate sustained increases in demand. This ignored a number of issues including the lack of historical evidence that periods of high immigration produced price rises that were sustainable and the likelihood that current immigrants (unlike previous generations) would return to their countries of origin (especially if the economy weakened).

Things do change, but permanent changes in economies are usually slow. Mean reversion is almost always a far more likely scenario than a step change.

This brings us to another common example of irrationality. The tendency to assume that current trends will continue indefinitely. This is, of course, much discussed by advocates of behavioural investing. This is more difficult to draw an easily applicable lesson from. All I can really say is: check the assumptions on which your long term forecast are based for credible expectations of how current trends will slow or reverse.

Another very strong indicator that markets are irrational, is the existence of arbitrage opportunities or violations of the law of one price that persist. This happened during the dot com boom. Although it appears to be rarer than the other indicators, I suspect it is merely less well reported.

There is one ironic twist to all this. Acting on any of these indicators is usually done in the expectation that the market will, in the future, return to rational prices, to allow one to exit with a capital gain. So the market may be inefficient, but only for a while.

Comments

Printing itself out of trouble : Interactive Investor Blog
Monday, 6th July 2009 5:36PM

[...] says markets may be inefficient, but only for a [...]