Will they ever come back? Stocks such as JDS Uniphase ( JDSU), BroadVision ( BVSN ), Applied Micro Circuits ( AMCC), Ciena ( CIEN) and Network Appliance ( NTAP) were the most sparkling bubbles in the frothy market that came to an end in March 2000. Since then, many have been reduced to the status of penny stocks. And each rally has been but a momentary respite in that descent. But it's hard to forget that these companies once seemed to be the leaders of entire market segments or the pioneers of whole new industries. Certainly not all of that was illusion, was it? Any investor who still owns shares in any of these companies surely hopes not. And many who made a profit from holding the shares -- and who therefore know the stories of these companies -- are just waiting for the slightest sign of a recovery before buying back in. Both these groups want to believe in a comeback for these bubble stocks. Of course, the potentially huge profit from such a turnaround doesn't do anything to discourage such hopes either. If BroadVision is headed back to $90, isn't loading up at $4 a no-brainer?
Don't Let Nostalgia Nab You
But are such hopes anything more than wishful thinking? How do you tell the difference between a realistic chance to get in on the ground floor and nostalgia that will bury a portfolio under further losses? Try asking why any individual stock would come back now. Just because a stock once soared isn't enough reason by itself to expect that it will soar again. Nor is it enough that a stock is massively cheaper than it once was. Just like any other stock, these former stars need a catalyst before they can head higher. Do they have what it takes right now -- or even in the next year or so? In most cases, the answer is no. Either conditions in the industry haven't improved noticeably yet, or the company's individual prospects have indeed declined in the almost two years since the bubble popped. At best, the turnaround is unacceptably distant. At worst, it won't ever arrive. There are some exceptions. Among technology blue-chips and former bubble favorites, Intel ( INTC), Nvidia ( NVDA) and GlobeSpan Virata ( GSPN) all have current catalysts that make them worth a look now -- and I'll get to them at the end of this column. But first the bad news. Take a look at former bubble favorite (and a former Jubak's Pick) BroadVision, for example. In the fourth quarter of 2001, the company reported solid progress in cutting costs that reduced losses below Wall Street expectations. And the company achieved remarkable success in holding onto existing customers. But BroadVision delivered disappointing news when it came to developing new sales. Sales fell another 6% from the third quarter of 2001 and came up about $3 million short of Wall Street expectations for the period. That led Wall Street analysts to postpone their projections of break-even for BroadVision to the fourth quarter of 2002 from the second quarter of the year. For the year, revenue could well be 25% or more below previous projections. With almost $200 million in cash as of the end of the December quarter, BroadVision certainly has enough in the bank to get back to break-even. But with the company projected to lose 9 cents a share in 2002 and to earn just a penny in 2003, it's hard to argue that this is a must-own stock.
Waiting, Waiting, Waiting for Recovery
A number of other bubble stocks are in comparable positions because of conditions in their industry. JDS Uniphase, for example, is waiting for a recovery in capital spending in the telecommunications industry. But service providers continue to cut budgets in 2002 -- although not at the pace they slashed spending in 2001, and it looks as if Corning's ( GLW) projection of no spending recovery until 2003 remains a solid estimate. In other cases, expected catalysts haven't materialized on schedule. For Qualcomm ( QCOM), for instance, the company's potential addressable market is set to expand from roughly 20% of the wireless personal-communications services (PCS) market, according to Morgan Stanley, to about 100% with the introduction of new, third-generation phones. Unfortunately for Qualcomm, however, the rollout of third-generation equipment is taking longer than expected, and that has slowed Qualcomm's drive to expand its market -- with predictable consequences for the stock. And in still other situations, a lack of consolidation has hurt investing opportunities. For example, overcapacity in the DRAM computer memory market drove the price of a 128-megabit chip to less than $1 at the end of 2001 from $18 in 2000. What's needed to restore prices is to get supply and demand back in sync. But weaker companies -- such as Korea's Hynix ( HXSCF) -- have stubbornly hung in rather than exiting or selling to stronger players. (It's critical to the usefulness of this kind of analysis that you put a company in the correct industry for comparisons. Cisco ( CSCO), for example, does indeed sell to the deeply troubled telecommunications service provider industry. But the bulk of Cisco's sales go to what's called the enterprise market - businesses buying equipment for their own networks -- and that market is much healthier and much closer to showing improvement than the telecom sector. Cisco, I'd argue, is closer to a real turnaround than JDS Uniphase, for example.) Some former star performers face even tougher sledding. Not only do conditions in their industries remain difficult, but competitors have changed the technology balance in the sector. For example, Intersil ( ISIL) and Agere ( AGR.A) have, until recently, split the market for wireless local area network chips. But during the time since the bubble burst, companies such as Broadcom ( BRCM), RF Micro Devices ( RFMD) and Intel have all targeted the market. In some segments the pace of technology change has been nothing short of revolutionary. While companies such as Advanced Micro Devices ( AMD), Atmel ( ATML) and Intel have been waiting for demand to catch up to supply and improve prices for flash memory chips, a new, still-private start-up Matrix Semiconductor has raised capital to create a product that could shrink the size of a flash memory chip enough to reduce costs by 50% to 75%. The first flash memory chips using Matrix's technology could hit the market as early as the second half of 2002. And in the meantime, flash memory chip makers continue to struggle with overcapacity and sluggish demand.
Buying Opportunities in Sight
Not everything is bleak in techland. Some companies are making progress toward recovery despite general conditions in their market segments. These companies have individual catalysts that could power their stocks higher even though their industries are stuck in neutral, at best. Intel's catalyst, for example, is the company's impending switch to 12-inch silicon wafers this year. Even with PC unit growth projected to increase by just 5% to 8% in 2002, that manufacturing switch by itself could be enough of a catalyst to power Intel's stock. The use of 12-inch wafers will reduce the cost of a Pentium 4 processor by about 30%, according to Michael Murphy, editor of the California Technology Stock Letter. That will let Intel make a Pentium 4, currently priced at $160, for $40 (including testing and packaging). There's no way that won't improve the company's margins. In other cases, companies with dominant market positions in one market segment are using that leverage to open up new opportunities. The typical way to do this is by adding new functions to the dominant chip (or chipset) that were previously performed by other chips from other makers. For example, Nvidia has used its market clout in graphic processors to go after the chipset market against companies such as VIA Technologies for chips made with Advanced Micro Devices technologies and is expected to go after the much bigger market for Intel chipsets later in 2002 or in 2003. And finally, a few companies have managed to force the necessary consolidation in their own industry segment and create an internal catalyst as well. GlobeSpan, for instance, has merged with competitor Virata to produce a single company with offerings in the physical layer and the networking layer for DSL (digital subscriber line) chips. That combination should help the company expand market share by selling a full chipset, yet with lower sales and marketing costs. But GlobeSpan has also been moving to improve margins by moving established customers to higher-performance central office chipsets -- that not so incidentally carry higher margins for GlobeSpan. The results of these moves were apparent in the third quarter of 2001 when GlobeSpan's gross margins climbed to 56%, a 7 percentage point improvement over gross margins in the second quarter. When you see numbers like that for a company that has the kinds of catalysts that GlobeSpan has put together, you've moved well beyond the realm of wishful thinking and nostalgia for valuations in the bubble market of 2000. Yes, GlobeSpan did trade at $140 a share way back then. But if you buy it now, you're not making a bet that the shares will move up just because they're so much cheaper than they once were. Rather, you're buying it because of current catalysts that have the power to move the stock higher from its recent price of about $12, even if that level is far below its former heights. I'd revisit the rest of the bubble market favorites three or six months from now to see if any of them have moved from the realm of wishful thinking to real catalysts. I wouldn't write off all these stocks. I just think it's too early for most of them.