Efficient Markets, Mean Reversion and why chickens come home to roost

A recent article in the Financial times, Clash of the CAPE Crusaders, on the radically different conclusions arrived at by Jeremy Seigel and Robert Shiller when evaluating long-range stock results reminded us of an draft post we had begun a while ago.  This debate between value investors and others has been running for a long time, the latest iteration goes something like this:

EMH camp: Since we know stocks over time out-perform other investments and do not know what stock prices will do tomorrow, it is OK to go ahead and invest in stocks, and you should invest in a broad list of stocks (say an index fund).

Value camp: You should only invest in specific stocks when you have done your analysis, and believe the return from the investment will compensate you for the risk you are taking.

EMH camp: You don’t know what the price of the stock will be tomorrow, so there’s no reason for you to choose to invest today.

Value camp: That is true, but we don’t care about tomorrow, we care about the likelihood of prices being higher 3, 5 or 7 years from now, and that will depend on whether we are buying bargains now, since over time the true value will be reflected in the price. And even if it isn’t reflected in the price, our analysis has given us confidence that this is a good business proposition and the downside is limited. I can show you numerous periods where a purchase of stocks would have resulted in huge losses within relatively short order, and this is generally because prices exceeded value.

EMH camp: But you would have said they were overvalued well before they started to fall, and I would have lost out on some gain, and I know that over the long-term, stocks will do well.

Value camp: True, I admit that I cannot predict when exactly a bubble will end, but I do know when one is underway, and that it will end eventually. I’d rather make a little less, than risk losing a lot.

EMH proponents have always seemed to us like naughty children who hurt themselves periodically while swinging from trees and then blame their parents for not watching out for them. The oft-expressed view that the market always values all investments correctly seems to us a convenient tautology. There are numerous instances of speculative manias running amok, and they generally concern some new technology or market. Wildly optimistic rates of growth are assumed and incorporated into “analysis” by herds of professional investors, while non-professionals rush along in tow.  These speculative mania always end, and with remarkable speed.  We call EMH a tautology since most adherents would say this is perfectly acceptable and a result of new information being incorporated into the marketplace.  Our sympathies lie with the mean-reversion camp and their view that markets periodically overshoot highs and lows, creating opportunities for investors who can anticipate the eventual return to the norm.

We view our role as value-oriented investment advisors to find and understand these under-appreciated pieces of information, to evaluate their impact on investments and position ourselves to take advantage of them.  In Gretsky’s over-used metaphor, we skate to where the puck is going to be.  There are, however, limits to our powers of anticipation. Mean-reversion can take a long-time, and our investors generally have a limited time horizon. This is why, in our role as prudential advisers we generally maintain some investment in stocks in our balanced portfolios. At the same time, we realize that mis-pricings and mean-reversion applies to all asset classes and markets, though the impact may be more extreme in stock markets.  So, for instance, we have been very careful about the types of bonds we have purchased and held for clients over the past few years. We know interest rates cannot be kept at zero forever, and this means long-term bonds offer limited upside and a lot of potential downside. For the past couple of years, we’ve kept bond duration in most client portfolios at very low levels since we have not seen value in long-term bonds.