Selling the Cycles

We've all heard the admonition "trees don't grow to the sky." It's true, at least in degrees. After all, most trees do grow toward the sky, and some grow quite far and persistently, like a sequoia. Others go for the meteoric spurt before withering back to Earth, like an albizia, and then there's everything in between.

Saying that something doesn't grow to the sky doesn't offer much insight on when to sell stocks. In hindsight, we could see ExxonMobil (NYSE:XOM) was a sequoia and JDS Uniphase (NASDAQ:JDSU) was an albizia. If you owned ExxonMobil in 1995, you'd probably still want to own today. The same can't be said for JDS Uniphase.

Rules to sell by
Because investing is a discipline, investors look for rules on when to sell, often attempting to gauge what's a sequoia and what's an albizia. Of course, there is no one set of rules that signal a sell, just like there is no one set of rules that signal a buy. All rules have their strengths and weaknesses.

Some investors lean on strict selling rules, such as if a stock drops 10% or more below its purchase price, it's a sell. Problem is, an investor can be whipsawed into insolvency by such an ironclad rule. There have been few times in my investing career when I've bought a stock and not seen it drop 10% or more after purchase.

That's one of many reasons investors will tether sell decisions to a more flexible pole, such as a price-to-earnings multiple. It's not illogical to assume a stock sporting a high P/E multiple could be under the spell of unrealistic expectations if the multiple exceeds the overall market multiple by a wide margin. Over time, the premium is unlikely to be sustained, so it's reasonable to consider selling a stock that trades at a P/E premium based on earnings expectations.

Then again, a P/E premium can be sustained much longer than we'd expect. I thought for sure JDS Uniphase was overvalued at $200 a share in August 1999, by every measure of value, including P/E multiple, based on my expectations. About six months after my logical deduction, JDS Uniphase was trading at $1,100 a share. Eighteen months later, it was still trading above $200. That's quite the opportunity cost for any investor who sold at $200, and quite the catastrophe for any investor who sold short at $200. 

A more conscientious investor might look for general weakening of company fundamentals -- contracting margins, expanding days/sales outstanding, wilting EPS, flat-lining revenue growth -- before selling. It seems intuitive: Weakening fundamentals often portend bad things. The principal shortcoming of waiting for observable weakness, though, is that stock prices often weaken before the fundamentals.

How about selling when intrinsic value is reached? Sure, but what's intrinsic value? Many investors have latched onto Warren Buffett's putative calculation that relies on discounting future cash flows, which values a stock much like a bond. It's far from a precise science. Ask 50 investors to determine projected cash flows and a current discount rate, and I can guarantee you'll receive 50 different answers. I tend to be leery of intrinsic value calculations based on discounted cash flows, including my own.

Take the middle road
Most stocks lie somewhere in between an ExxonMobil and a JDS Uniphase. Many are established, sturdy, seemingly cyclical-moving, companies. They're not spectacularly moat-endowed, nor are they duds. They ebb and flow with economic happenstance. More important to me, their price appears to hold some memory to suggest a buying and selling season.

I would classify the following six companies as being neither sequoias nor albizias, but cyclicals based on price behavior. And since I think they are cyclicals, I think investors should consider taking profits after the stout run they've posted over the past two years. That's not to say there's anything egregiously wrong with these companies. To the contrary, they still sport reasonable margins and reasonable ratios, and they're still good businesses.  

Then why sell? Trees don't grow to the sky. If you agree that they are cyclicals, then at least considering selling is logical, since it's the reverse of considering buying. After all, the most remunerative time to buy a cyclical stock is after its share price has declined on a string of underwhelming financial results. Doesn't it make sense, then, to sell a cyclical stock after its share price has advanced on a string of encouraging financial results?