Stock Prices and Alpha

Aside from the considerations we discussed last week, there is this to remember about stock performance, the performance of a stock as an investment is not (entirely) a matter of whether it rises or falls in price. A stock also entitles its holder to a portion of whatever dividends the issuing company may declare. Suppose, then, that we do the arithmetical magic to factor in the dividend stream as if it were being paid out day by day, and we included that along with the stock price move as the performance of XYZ. Once we do this, we’ll want to be sure that we’re matching the stock price against a broad market benchmark that also includes dividends as part of performance.

Fortunately, these are readily available. The S&P Index, for example, comes in three variants: one that considers solely the price of component stocks; one that factors in dividends (the “total return” index); and a third that subtracts the tax on those dividends (“net return.”)

We might make discussion easier by giving a name to the difference between the stock performance figure for a particular stock (or of your personal portfolio of stocks) on the one hand, and of the broad market’s equivalent on the other. Let us call this difference alpha.

Positive alpha is a good thing. It means you're doing better than 'the house.' Negative alpha means you're doing worse.

Now: a curve for a particular stock’s alpha as we’ve defined it will often look a lot like the Bell curve. One significant implication of this: alpha nets out to zero. That is worth saying again: alpha nets out to zero. In graphical terms, there is as much space beneath the left-hand (negative alpha) side of the pertinent bell curve as there is beneath the right-hand (positive alpha) side of the curve.

This is a good moment to observe that stock picking, betting your nest egg on the future of individual stocks, or on some expensive “active manager's” ability to buy individual stocks, is usually a bad idea. It is cheaper and safer to invest indirectly through broad based (passively managed) funds or exchange traded funds (ETFs).  It is best to bet on that index against which alpha is measured, rather than betting on some stock’s alpha itself.  Why? Because (a) have I mentioned this yet? alpha nets out to zero and (b) if you're trying to beat the market, you're paying a broker and/or advisers for that zero.

Let us ask another sort of question now. If stock price moves are random (or to the extent that they are): why are they random?

One widespread view is that stock prices jump about in a “random walk” fashion precisely because the stock market is accomplishing its job. This is also, with important qualifications, my own view: the market is a venue for the efficient allocation of capital, according to rational expectations, and the randomness arises from that fact. This is a somewhat counter-intuitive idea. After all, in most contexts in our day-to-day life randomness is a bad thing. When I get up in the morning and get into my car for the drive to work and turn the key in the ignition, I want the car to start. Deterministically. Certainly. I don’t want there to be any uncertainty or randomness about it!

But the stock markets are a rather different sort of mechanism. Or, rather, they aren’t a sort of mechanism at all, but a meeting place for a lot of different vitalisms. And in this context, it might help us to consider a simple seasonal development, like the arrival of Halloween. Presumably in parts of the world where Halloween is celebrated, the demand for pumpkins rises rather suddenly every October and falls even more suddenly at the end of that month.

“Aha!” says the naïve investor, “so I can get rich! All I have to do is find a way to bet all my money and all the money I can borrow on the prospect that pumpkins will rise in price this coming October. My bet will pay off, and I’ll be rolling in enough dough to take full advantage of the big cranberry crunch the following month.”

Sorry, that doesn’t work. If you voice such a speculation out loud, your neighbor surely says, “if it were that easy, everyone would do it.” And he’s right. Think of the market as a collective intelligence: Mr. Market. Now, surely, Mr. Market knows that people are going to want to buy pumpkins by late October and are going to want cranberries by late November. The prices of these products and all derivatives of those prices factor in such matters. So, sorry, you don’t get to outsmart Mr. Market that way.