The Rule Maker Portfolio managers must make a decision about optical components manufacturer JDS Uniphase (Nasdaq: JDSU), whether we want to keep it in the portfolio or sell it and reallocate the funds to another Rule Maker.

I recommend selling it.

Keep in mind, however, the context of my argument. I'm not saying JDS isn't a good company. I'm saying it's not an appropriate investment for the Rule Maker. It's simply too risky, too young, too pricey, and too complex for a portfolio that looks to invest in established companies generating lots of cash.

I've followed JDS since 2000, researched the optical components market for The Motley Fool's Industry Focus last year, and even suggested in previous Rule Maker articles that the stock isn't a good fit for the portfolio. I'll flesh out my argument in today's story, and the Rule Maker managers will decide this week whether to sell the stock. I'll return to my analysis of credit card company Capital One (NYSE: COF) next week.

First a little background. Our track record with JDS since we bought it in February 2000 is awful. We paid $100.75 per share for a stock that now trades at about $21.80. We're down 78%.

But my concern with JDS has much less to do with its recent stock performance than its risk, complexity, and failure to mesh with the Rule Maker philosophy. First, I'll provide the best argument I've heard regarding JDS Uniphase's bright future, then I'll present a counterargument.

JDS Uniphase -- An emerging giant?
As the Internet grows, the richness of our experience online is limited by bandwidth alone, not by the processing power of the microchip that runs the device we use to access it. Bandwidth, therefore, is the key. Telecommunications networks must evolve to handle fast-growing Internet traffic, and the technology at the center of our ability to improve network speeds and lower costs is optics.

As JDS Uniphase management explained at a recent investment conference, carriers are spending more and more money on optical equipment. Legacy networks are 23% optical -- mostly in long haul segments -- yet next-generation networks are expected to be 64% optical, nearly triple their proportion today. As optical components fall in price, carriers will be able to expand optical networks from the long haul market (in between cities), to the metro market (inside of cities), and perhaps one day, when we're all flying to work on motor scooters, right to the curb in front of our houses.

JDS is the dominant supplier of active and passive components for the industry. No company in the industry can match its product portfolio's breadth and depth. Moreover, its modules strategy involves aggregating components into packages, which are easier for carriers to integrate into networks. It's a smart strategy. Semiconductor equipment maker Applied Materials (Nasdaq: AMAT) adds value to its machines by constantly increasing the number of functions each machine can perform, which means fewer machines along the assembly line and fewer stops in the assembly process. This results in higher yields and lower costs for customers such as Intel (Nasdaq: INTC).

What's not to like?
I have two problems with keeping JDS in the Rule maker portfolio. First, it's just too difficult to understand what kind of demand exists, long term, for laser diodes, raman amplifiers, interleavers, and other optical components. Unlike Intel, which built an empire on silicon and its core line of microprocessors, JDS must develop expertise across a wide range of products, and across a wide range of materials, such as silicon, gallium arsenide, and erbium. Then, it must integrate these products into modules, and eventually, into a hybrid product that is part chip, part module. There is enormous innovation risk that accompanies this journey. In my opinion, this detracts from our ability to understand the business and get a picture of its future.

While the demand for optical components may be enormous, is it as enormous, say, as the demand for microchips in 1993? The reason I ask is because the company is being valued much the way Intel was in early 1993. At least with Intel in 1993, however, we could see companies and consumers regularly buying new PCs, think about the prospects of a computer on every desk, and ask ourselves whether these projections made sense.

With JDS we're in the dark since it's unclear how quickly or to what extent carriers will be able to afford to build new networks; it's unclear to what extent fiber will spread from long haul markets, to metro markets, to neighborhoods; it's unclear if JDS and the industry will be able to scale production and improve yields enough to drive costs down; it's unclear if JDS will be able to keep its margins up as average selling prices plunge; and, it's unclear if consumers are really interested in paying for the emerging generation of broadband applications.

This last point is worth consideration. On the one hand, it's easy to believe the Internet is in its stone-knives and bear-skins phase. After all, we can't even download movies over the Internet. Surely consumers will clamor for new improvements and new applications, and this will spur bigger investments from telecom carriers.

On the other hand, an investment based on the assumption consumers will pay for more bandwidth all the time -- and that JDS in particular will reap the rewards -- is a bit of a wild swing. Essentially, it's a bet on the future development of the Internet, Internet applications, consumer behavior, and the carriers' ability to fund the optical build out. That's a lot to assume. You may believe these things will happen. You may even have insight as to why it will happen, but we don't pretend it's an easy call.

And there's a counterargument to the need for exploding bandwidth. I spend more than 40 hours a week online and don't need video conference calls, media rich commercials, or splashy graphics. I get held up a lot more often by poorly-organized websites than by Internet bottlenecks. I don't need access to a virtual library when for a few bucks a year I have access to the entire contents of the Encyclopedia Brittanica.

For a great many people, perhaps most, the Internet is a research tool, not an entertainment platform. A T-1 line or cable modem provides fast access to the basic text and graphics services the average Internet user requires. More than this, truthfully, I don't need and wouldn't pay for. As for downloading movies, pay-per-view isn't bad, even if it's not real time and even if I can't get every movie under the sun.

It's unclear to me that there will be sufficient demand for super-high bandwidth services, certainly not enough to start running optical connections into thousands of office buildings or neighborhoods, not considering the cost.

The root of the problem with JDS is this: Investing boils down to probabilities, and the Rule Maker approach isn't based upon venture capital-type bets. This is the second problem I have with JDS. We want to make bets when the odds are overwhelmingly in our favor. I'm not convinced this is the case for JDS, and I don't believe we have to bet on the stars aligning for companies like JDS Uniphase to be successful with this portfolio, not when there are more familiar constellations close by. 

Based on a 151-year history and the industry's strongest brand name, it's a reasonable bet American Express (NYSE: AXP) will find ways to expand its charge and credit card business, to leverage this business to grow its financial advisors and banking units. It's a good bet Johnson & Johnson (NYSE: JNJ) will continue developing profitable pharmaceutical products, and that its consumer products line will keep the brand name fresh in consumers' minds. These are the kind of investments the Rule Maker should be making: Great companies with solid brand names at reasonable prices.

In addition, we have downside protection in these companies. The market understands how they operate and they are clearly valued on the basis of future cash flows. JDS, despite its 80% drop, still has a $25 billion market value, which leaves, in my opinion, plenty of downside given the ground it must cover.

My recommendation is that we sell JDS Uniphase. It will be another step towards getting the portfolio back on track.

Note to reader: In the next five business days, the Rule Maker Portfolio plans to invest this month's $500 installment in cell phone giant Nokia (NYSE: NOK). In the latest quarter, Nokia grew its handset sales 21% to $5.2 billion, while maintaining 20% operating margins. Its competitors simply can't match this kind of profitability. Nokia's network business, which we don't hear much about, grew 35% to $1.8 billion and recorded operating margins of 18%, compared to margins of 4% for rival and market leader Ericsson (Nasdaq: ERICY). Nokia continues distancing itself from the pack.

Richard McCaffery, one of the Rule Maker managers, doesn't own shares in any of the companies mentioned above. The Motley Fool is investors writing for investors.