View From a Height

Markets That Aren't Always so Wise

People who've read my review know I'm a big fan of James Surowiecki's The Wisdom of Crowds. That said, I think his faith in the market's pricing mechanism may be a little overstated. And not just because it seems prone to frequent, er, corrections. Surowiekci himself notes the difficulty of pricing in long-term company performance. What he doesn't note is the way that analysts get around that problem, and how it violates of one his conditions for collective wisdom - independence.

If you're like most people, you probably think that analysts - even Janus-type "smartest-guy-in-the-world" analysts who dive into tunnels and count heads at restaurants come up with their estimates in sort of an isolated, cloistered, ivory-tower-with-oak-paneled-walls sort of way.

Not.

Most fundamdental analysts (and most analysts are fundamentalists) use one of two methods to value a stock: ratios or discounted cash flows. I won't get into a long discussion here, but basically discounted cash flows estimate the expected free cash flow - how much money the firm actually makes - and tells you what that's worth to you now.

So far, so good. Except that nobody acutally relies solely on those numbers. Ratios are things like Price-to-Earnings, Price-to-Sales, Price-to-Free Cash Flow; that sort of thing. Many people who use a discounted cash flow analysis also use a ratio analysis to estimate the value of the "out" years, the years past their 5-year model. And even those who don't, usually use some sort of comparison to industry ratios to make sure they're still in the same ball park, even if it is a little hard to see home plate from here.

Do you see the problem? Ratio analysis already includes the price of the stock. There's no independence here. Analysts - fundamental analysts, anyway - are alway all looking around at each other to make sure they're going in the same direction, which may still be right over a cliff. The technical term for this is "herding," an analysts do it all the time.

Now, you may say that analysts don't set the market price, the market does. And you'd be right. To a point. But analysts are extremely influential in making markets, especially when they act in groups. Why else do Schwab and other public sources of research report analysts estimates? And remember, this sort of modeling is exactly what M&A and investment bankers do to set purchase prices and IPO levels.

There's really no way out of it. Trying to forecast earnings past five years in advance is truly folly, and most analysts, given enough sodium pentathol, with readily admit to that. Still, the company does have expected earnings, and thus value, past that point, and they must be taken into account. The only way to do this is to assume that the market is basically right, and tweak the numbers until they look realistic.

But this may also help account for sudden changes in individual stocks on seemingly minor news, missing earnings by a penny out of 60 cents, or what-not. Not only is everyone re-evaluating their estimates on their own terms, they're also re-evaluating them on everyone else's terms, too.

Which means that a market's wisdom may not be so collective, after all.