I sold my entire position in Cisco (CSCO) last week, and that’s what this piece is about.

I thought that would suffice as an attention-grabber, since suggestions of selling Cisco are akin to blasphemy these days. And while my public declaration to sell the Holy Grail of all tech stocks may make me the Maplethorpe of the investment community to some, I think martyrdom is closer to what history will have in store for those willing to courageously sell into this Tech Stock Mania. So while you may or may not own Cisco, the following rationale may apply to other, similarly swollen tech stocks in your portfolio.

A little history is in order to understanding my affinity for what I still contend is the greatest company of our time. I’d been following Cisco for years and really got interested in it in 1997. Back then, it was clobbering its computer-networking-product competitors like 3Com (COMS). My dilemma at the time—get this—was that I was afraid the stock had come too far, too fast. The stock was trading in the high teens (adjusted for splits) and it had a forward P/E (using 1998 earnings) of 24. The company’s earnings, which had been clipping along at 50% per annum, were slowing down to what was projected to be about 35%. Thus, the PEG—P/E to Growth—was 0.68 (24 divided by 35). [The PEG is one of a number of justification measures of increasing popularity for paying high P/Es for fast growing companies whose original premise was that "lower is better," especially below 1.0.] To make a long story short, I bought the stock because I thought I was getting good growth for cheap and I wanted a Net infrastructure play. Most of my buy trades were in the $17-19 range (again, split adjusted).

My experience has been a very good one. Since my purchase nearly three years ago—and I don’t claim to be an early discoverer of this company—my return has been 660% or about 125% annualized, based on my holding term. And what’s the company done? Just about what I thought it would; it grew earnings by 32% per year. But given that the company’s only grown annually by 32%, and the stock by 125%, the result is multiple expansion. What does that mean? Remember that in 1997 the P/E was 24 and the PEG was 0.68? Today, the forward P/E is 135 and the PEG is 3.8 (earnings growth still projected at 35%). For the next three years to replicate the last three, the stock would have to rise to $1,041 (not adjusted for eventual splits), the forward P/E would be 420, and the PEG would be 12. That won’t happen. Period...end of story..."new paradigms" not withstanding...not in your lifetime.

The more likely scenario is that Cisco continues to execute its business strategy and earning grow annually at 35% before settling down to a future growth rate of about 30%, three years from now. All the while, the stock should gravitate to a reasonable, long-term valuation of between one and two times its growth rate. The math? Three years from now, a PEG of 1.0 equates to a P/E of 30 over $2.48 in earnings and a stock price of $74—resulting in a three-year, annualized loss of 18.6%. Best case scenario, a PEG of 2.0 would be $148 stock price and a three-year, annualized return of 2.6%.

What confuses the novice investor the most is that they read this kind of "relative," price-based rationale, and can’t separate it from the "absolute,"company-based attractiveness. In other words, they say, "What’s wrong with Cisco? They’re a great company! You’re crazy for selling!" That’s not the point. I agree that they’re a very attractive company, in an absolute sense, but in a relative one—recognizing that it trades at four times its growth rate—I say its stock is unattractive looking out three years, my minimum investment horizon.

Let me put it another way. Everything has its price. Let’s say you just built a $300,000 house that you love. Interested in selling? No way, you say! What if I offer you $1.2 million dollars—four times what it should be worth. You’re out of there faster than you can say, "Honey, get the kids out of the way of the movers." This is especially true if you believe you can turn right around and build another one.

I sold most of my shares in the $137 range, less than a dollar from its all-time-high closing price. In all likeliness, the stock will continue to do fine in the short term. It might even set a new all-time-high, tied to good news, like an impending stock split. However, a "best-of-all-worlds-and-then-some" expectation is priced into Cisco (and many like it) and the slightest stumble, missed execution, or earnings disappointment will clobber the stock.

Let me reiterate, I’m Mr. Internet. I spend up to 12 hours a day on it, use it for all of my research, communicate with clients on it, and gain nearly all of my new clients though it—not to mention time spent on it away from work. I was also was early to identify its prospects for investing. But my enthusiasm for Net stocks has been tempered with my fundamental training which prevents me from paying ludicrous prices for stocks. Thus, my investments have mostly been the more moderately priced, Net-infrastructure plays.

I’ve even modified my methodology to account for the "New Era" valuations, but every behavior, taken to an extreme, eventually crosses a threshold of rationality. What’s at the center of the debate today is just where that threshold is, and many "New Era" investors don’t believe there is one. They would argue that tech stock valuation premiums will continue to widen. I believe that the long-term, sustainable valuation level for these stock is about half where it is today, and that we are at the tail end of a Tech Stock Mania, the bulk of whose participants are late-comers and headed for negative results over the next few years.

Therefore, having captured much of the tech run, I’m modifying my strategy with symbolic moves like selling Cisco. Additionally, I’m limiting my overall tech exposure, buying meaty helpings of non-techs, diversifying globally, and focusing my tech picks on contrarian, reasonably priced stocks. In that spirit, if Cisco presents itself again at one-times-its-growth-rate, I’ll readily saddle up. In the meantime, it’s "Adios, Cisco Kid."

At the time of publication, the author was neither long nor short any of the stocks mentioned in this article, either in client accounts or personal ones. Positions may change at any time.