Audiences crowding into the multiplex cinemas of Mountain View and Santa Clara, California, are witnessing the return of a Silicon Valley phenomenon that had faded away. Along with the previews, they're once again being treated to big-screen help-wanted ads for engineers. It's one more hint that the two-year tech swoon has bottomed.

Investors have been anticipating this for months, last week's skittishness notwithstanding. Since hitting a three-year low on September 21, the tech-dominated Nasdaq 100 has gained 41%, and the closely watched Philadelphia Semiconductor Index has surged 49%.

The indexes have been propelled by modest but tantalizing evidence that the worst is behind us. Memory-chip prices have been increasing. Fourth-quarter personal computer sales were better than anticipated. DVD players, digital cameras and game consoles from Microsoft, Nintendo and Sony enjoyed robust sales. And there's growing buzz about wireless computer networking, next-generation cell phones and digital media, such as music, video and photography.

The next wave of technology spending lies ahead, and the fate of the American economy depends to a large degree on how quickly it develops. The outlook is more for a ripple than a tsunami. The double-digit growth rates of the late 1990s will be replaced this year by a spending increase in the low single digits. But things will gradually get better throughout 2002. And there will be significant growth in some key sectors, such as security, storage and integration of big corporate software programs across the Web.

Other areas, notably telecom equipment and semiconductor manufacturing equipment, are destined for more rough sledding. But if the economic recovery arrives on schedule in the next quarter or two, a pickup in tech outlays could gain momentum in the second half, setting the stage for better days in 2003 and beyond.

Mark Zandi, chief economist at, says the primary determinants of corporate capital spending all should improve incrementally this year. For 2002, Zandi sees corporate earnings bottoming, the initial-public-offering market for stocks picking up a bit and factory utilization gradually improving from the current moribund levels. While he's not expecting a boom year, Zandi thinks we've got the ingredients in place for a modest uptick in spending.

Other data suggest the same thing. A November survey of more than 1,000 information technology professionals, conducted by Gartner and Soundview, found technology budgets expected to grow 1.5% in 2002. Others offer higher 2002 estimates. Forrester Research, for example, sees 2% growth, and Giga Information Group sees 4% growth.

Compaq Computer Chief Executive Michael Capellas is even more optimistic. He sees 6%-7% growth in information technology spending this year. But he warns that buyers will be far more cautious than in the bubble years. "Now it's about bulletproofing the infrastructure, stabilizing the baseline." Capellas told Barron's. "Buyers are asking, 'How do I get more from what I have, tie systems together, improve reliability and sustainability.' I don't see a lot of interest in new large-scale projects."

Looking further out, forecasters have a cheerier view. Forrester expects 9.7% growth in IT spending in 2003, with a return to double-digit growth in 2004, driven by wider availability of broadband and wireless Internet access. A similar, if more modest, forecast comes from UBS Warburg: 6% revenue growth for tech companies in 2003, 10% growth in 2004.

Just as past technology waves have been driven in turn by mainframes, minicomputers, personal computers and the Internet, so too will the next wave be built on a new wave of products and services. So far, however, there's no single clear-cut driver, no readily available buzzword, no killer application.

The elements for the next wave are beginning to emerge, nonetheless. At home, at work, everywhere we go, we will be connected to the 'Net. We're heading toward anytime, anywhere wireless computing. The PC isn't going away -- killing off technologies is surprisingly hard to do. But we'll increasingly use a range of other devices to communicate, create and work. Established companies like Microsoft and Apple Computer -- and new ones like Moxi Digital -- have made heavy bets that computing will more often be done away from traditional desktop PCs and laptops.

For investors awaiting the turn, patience and agility will be required, as the punk economy is keeping a lid on capital spending. Tom Mangan, head of Arthur Andersen's advisory service for chief information officers, says corporate tech buyers, once willing to take the long view, now demand quick returns on investments in new technology. "Right now," Mangan says, "ROI is king."

Fed Chairman Alan Greenspan echoed the same sentiment in a speech in San Francisco earlier this month. "When capital spending eventually recovers," he said, "its growth is likely to be less frenetic than that which characterized 1999 and early 2000, when outlays were boosted by the dislocations of Y2K and the extraordinary low cost of capital faced by many firms."

The numbers tell the tale. From 1992 through 2000, according to Forrester Research, communications equipment shipments measured in dollars increased 185%, while software investments expanded 201% and semiconductor shipments grew 237%. In that same span, nominal GDP grew just 62%. Meta Group Chief Executive Dale Kutnick notes that over the past six years, the percentage of corporate revenues spent on information technology has increased 50%, a trend that will not likely continue at the same pace.

Indeed, some think it could take as much as two years for the next wave to gain momentum. Pip Coburn, technology strategist at UBS Warburg, sees some big trends coming: the switch from PCs to handheld devices, the emergence of a myriad of Web-based services for businesses and consumers, the chip industry's move to larger, more efficient silicon wafers, and the shift from narrowband communications to broadband, among others. But he warns that the odds that more than one or two of those taking hold before 2004 are small. "This is not just about waiting for the economy to turn," Coburn wrote in a report this month. "We are facing the end of numerous dramatic secular trends which are fading before new ones develop."

In some areas of technology, things will continue to get worse before they get better. For instance, chipmakers continue to slash their capital budgets, which means continued difficulties for the semiconductor equipment makers, such as Applied Materials and Novellus (see "Intel Sneezes"). Just last week Intel announced that it would cut capital spending this year by 25%, to $5.5 billion. Goldman Sachs expects a 20% slide in semiconductor industry capital spending this year, after a 38% slide in 2001; Lehman Brothers foresees a 30% drop.

The picture is equally grim for makers of telecommunications equipment, as carriers continue to suffer from a glut of bandwidth. Mark Wolfenberger, tech strategist at Credit Suisse First Boston, expects a 30% drop in capital spending by carriers. Sprint, for instance, says its 2002 capital spending budget will be down 40% from last year.

Soundview's Arnie Berman, pointing to recent earnings warnings from equipment makers like Lucent Technologies, Juniper Networks and Ciena, and service providers like Sprint and Qwest Communications, observes that "the trough in telecom spending is even deeper than most observers expected."

Nor is telecom equipment the only area where budget-cutting rules the day. Many segments of the economy are planning to reduce IT spending. Of 21 industry sectors tracked by Meta Group, just seven -- including, ironically, the telecom industry itself -- expect to spend more on technology this year (see chart above).

Despite the new-found religion of return on investment, the biggest growth this year will come in a category that does little to boost earnings: security. Arnie Berman, technology strategist at Soundview Technology, says the dominant question being asked by IT mangers has switched from "Which of these new projects can we do without?" to "What do we need to do to keep our business safe no matter what?"

Often an afterthought, security has taken center stage in the wake of the September 11 terror attacks, the recent anthrax scare and ongoing fears over worms, viruses and hackers. Security budgets are ballooning. J.P. Morgan estimates IT security spending will grow 43% this year.

Of course, investors have been onto this trend for months now. Stocks like Check Point Software, Internet Security Systems, Symantec, RSA and Network Associates have all doubled or better from their September lows. It's not hard to see why: There will be strong demand this year for a range of security products and services -- firewalls, anti-virus, intrusion detection, network scanning, even biometric security systems using face and fingerprint recognition.

With a finger to the wind, Microsoft Chairman Bill Gates last week announced a major shift in strategy for the software giant, which emphasizing security above all. In an e-mail to employees, he wrote that security is now "more important than any other part of our work."

Disaster-recovery planning should also get a boost in 2002, for obvious reasons. The leading player is SunGard Data Systems, which recently acquired Comdisco's rival disaster recovery unit. IBM is the only major competitor. Berman asserts the real beneficiary of new spending on disaster recovery planning will be the providers of hardware needed to set up duplicate data centers -- companies like Cisco Systems, EMC and IBM. Compaq's Capellas notes that security concerns have boosted sales of his company's storage systems and high-end computers, including the pricey fault-tolerant systems sold by the division that used to be Tandem Computers.

Another hot sector this year is data storage. There's a particular buzz about networked storage -- centralizing information rather than having it all over the network. Andersen's Mangan says there are both economic and security reasons for doing that. Concentrating operations into fewer data centers means fewer facilities, reduced headcount and fewer points of network vulnerability.

But the picture is complex. EMC, the sector's long-time leader, has come under assault from a host of new competitors big and small. According to IDC, a market research firm, Compaq has become the worldwide leader in disk storage revenue, as well as terabytes sold and units shipped. There's also evidence that venture capitalists' affection for storage has created an over-populated market. Capellas confirms that storage pricing has come down "dramatically." Even so, there is opportunity for storage software providers to help companies manage their capacity more efficiently. EMC, it's worth noting, has been refocusing on software.

Another hot theme is Web integration services -- the push to provide Web links to big corporate software applications, like those from SAP, Seibel, i2 and PeopleSoft. Precursor Group, a Washington-based research boutique, asserted in a recent report that businesses will "spend a large share of their IT budgets on software to integrate the expensive platforms they have built," so that they can "capture the promised productivity."

Similar motivations will propel the outsourcing trend -- Giga expects 10%-20% growth there, boosting the fortune of companies like EDS, Accenture and Computer Sciences that can operate large corporate technology departments.

The focus on outsourcing should also help the contract manufacturers, though it's probably a bit early to call for a turnaround. Companies like Flextronics, Solectron and Jabil Circuit suffered mightily in 2001 as key customers cut orders for PCs, cell phones and the like in the face of reduced demand. Solectron, for instance, reported that revenues for the quarter ended November 30 plunged by $2.5 billion, or 44%, to $3.2 billion, from a year earlier.

The lower revenue reflects weak end-markets, rather any weakening of the outsourcing trend, however. Indeed, cost pressures could actually accelerate reliance on outside manufacturers. IBM, for instance, this month farmed out PC production in Europe and the U.S. to Sanmina-SCI, shedding 900-plus workers in the process.

In the meantime, though, contract manufacturers must contend with the fact that PC demand this year will likely remain soft. Optimists note that corporate America is due for a major PC replacement cycle. In the past, companies have tossed their old hardware every three years, and many PCs were purchased in 1999 ahead of Y2K. But few new applications require additional processing power. So expect the cycle to stretch out.

On the consumer side, PC sales beat expectations in December, driven by aggressive pricing. But there's little reason to expect a boom. Hopes that the introduction of Windows XP would trigger a spending boom have so far not panned out, though Microsoft says sales have met expectations. Compaq's Capellas expects a slight uptick in overall PC units this year, but a 3% drop in revenues for the sector as prices fall. Giga has a gloomier view, anticipating 10% drop in PC revenues this year.

Even without strong growth in PC sales, worldwide semiconductor sales should grow modestly in 2002, up 6% according to the Semiconductor Industry Association. One reason for that is a shifting inventory picture. Last year many companies focused on reducing chip inventory; now some manufacturers apparently fear they've gone too far, and are stocking up again. That's one reason DRAM prices have tripled over the last two months.

Ultimately, higher end-market demand for PCs, cell phones and other tech goods will be needed to keep the chip industry growing. Nonetheless, Soundview's Berman figures inventory restocking could help chip sales for another six months or more. Taking capacity out of the system will help as well. It is certainly a factor in the memory sector, where both Toshiba and NEC plan to exit the commodity DRAM business. And Micron Technology, the No. 2 maker of memory chips, is in talks to some of the chipmaking assets of Korea's Hynix Semiconductor, the No. 3, which would likely result in the closure of some chip plants. Capellas, for one, thinks that the elimination of some chipmaking capacity, combined with higher sales in the second half, could set the stage for some component shortages in late 2002.

And while spending on telecom equipment is weak, there are some promising areas elsewhere in networking. One of those is voice-over-IP, or the use of the Internet and corporate data networks to carry voice calls. Both Merrill Lynch and UBS Warburg list IP, or Internet protocol, voice technology as a hot trend for 2002. Meta agrees, noting that this is the year voice over IP emerges from the early adopter phase. The logic of voice over IP is simple: It is cheaper to run a single corporate network that can carry voice and data rather than separate networks for each.

Meanwhile, Capellas and others think 2002 will mark the start of big growth in wireless networks, both in homes and offices. Capellas, in particular, expects wireless home networks and the new "media servers" from Microsoft, Apple and others to link together speakers, cameras, printers, DVD players and other consumer electronics components. Says Capellas: "This is real, it will be the next big killer app: the integration of content distribution."

Leading the way in the wireless networking arena is a standard called 802.11 ("The Next Big Thing," November 12, 2001). And it's not just for consumers: Andersen's Tom Mangan says he's seeing increased use of wireless local area networks in offices, allowing for more flexible office configurations and work habits. Mangan notes, though, that companies need to be careful to address the additional security issues raised by wireless connections.

The evolution of wireless networks ties in to the fortunes of two other key sectors, cell phones and handheld computers.

In the near-term, the cell-phone sector faces daunting challenges. It's a maturing market. Analysts say 2002 will be the year when cell phone replacement sales will top sales to new users; Sprint last week said fewer new wireless customers than expected signed up in the fourth quarter, and Wall Street has been slashing expectations for the wireless service providers. Analysts at Merrill Lynch, Lehman Brothers and ABN Amro have all reduced their ratings on the sector over the last two weeks, battering stocks like Ericsson and Nextel.

For the makers of cell phones to prosper, wireless providers will have to come through on upgrades to the new technologies known as 2.5G and 3G, since they'll require new phones. So far, few places have new networks.

But the new technology is coming, and new services will follow. UBS Warburg, for instance, is hot on a technology called multimedia messaging services, or MMS, which involves sending video or audio messages over handheld phones. According to UBS, the technology will launch in some European and Asian markets this year; it's already available in Japan. By 2003, UBS says, every phone market-leader Nokia makes will be MMS-enabled.

Mark Anderson, a hedge-fund manager who also pens the Strategic News Service, a weekly technology newsletter, thinks 60% or more of the one billion outstanding cell phones will be replaced over the next 24 months. UBS expects handset sales this year to grow 10%, to 430 million units, and another 16%, to 500 million, in 2003. That's up from about 390 million last year and 408 million in 2000.

After a disappointing 2001, handheld computers should grow substantially; UBS projects 26% unit growth in the handheld market this year, with 90% growth in 2003. The old standalone personal digital assistants from Palm, Handspring, Compaq and others are evolving into much richer tools, with wireless networking ability that gives them full e-mail and Web access -- plus cell phone capability. Palm will introduce new hardware this year, and Research in Motion will offer a new version of its popular BlackBerry device that includes a cell phone.

That said, don't be surprised if the big winners in handhelds turns out to be the familiar duo of Microsoft and Intel ("Knockout Blow," November 12, 2001). UBS Warburg expects that Compaq and Hewlett-Packard, which make handhelds based on Microsoft's PocketPC operating system, will both pass Handspring in market share this year. And while Palm is expected to remain the market leader, UBS expects that the firm's market share will continue to erode. Intel, which makes the StrongARM processor used by PocketPC-based handheld makers like Compaq, HP, NEC, Casio and Toshiba, will also benefit.

Consumer electronics stores are hopping. Expect DVD and digital camera sales to remain strong. Also, this will be a huge year for the videogame business, thanks to the recent introduction of Microsoft's Xbox and Nintendo's GameCube, which are taking on Sony's PlayStation 2. All three boxes have been selling well; UBS Warburg sees a combined 72 million consoles being sold this year, nearly twice last year's total.

The sector still has a few drivers coming, including the introduction later this year of the Xbox in Europe and Asia and the GameCube in Europe. And both Sony and Microsoft are expected to roll out networks for their consoles, allowing gamers to compete head-to-head, surf the Web, check e-mail and download new games.

No question, the most innovative products in technology right now target consumers, rather than businesses -- think Apple's table-lamp shaped new iMac, the media server from startup Moxi, the new videogame consoles and the more powerful handheld devices. Microsoft, for one, seems positioned to play a key role in all of these categories, and could be a big winner this year.

To really succeed, however, tech companies need robust corporate spending -- and another cycle is coming. As Alan Greenspan continues to preach, technology offers businesses the chance to enhance returns. Slowly, as the economy improves, and the last excesses are wrung out of the late 1990s spending glut, the cycle will begin anew. And as this next wave builds, investors who position themselves carefully are likely in for a profitable, if sometimes treacherous, ride.

Attention Must Be Paid

Technology investors face a conundrum. On the plus side, there are signs that demand for technology goods and services has bottomed, and the economy seems poised for recovery. The negative side, however, is that tech stocks have already staged a furious run that anticipates a robust recovery -- leaving them vulnerable to almost any disturbance, as last week's market demonstrated.

The technology sector of the Standard & Poor's 500 index closed last year at nearly 46 times consensus 2002 profit estimates, which are admittedly depressed. This level of valuation relative to earnings has not been seen since the height of the bubble. Steve Milunovich, technology strategist at Merrill Lynch, concluded recently that 70% of tech stocks are overvalued based on consensus 2003 estimates. Jimmy Chang, chief technology strategist at U.S. Trust, is blunter. "The fact that we've hit bottom is no justification for paying outrageous valuations," he snaps.

Chang contends the recent rally was driven by portfolio managers underweight in tech stocks who fear lagging their peers. "The more the stocks rally, the more people fear missing out," he says. "No one has any conviction in the prices they're paying."

So what's an investor to do?

For one thing, reset expectations. David Readerman, growth stock strategist at Thomas Weisel Partners in San Francisco, doesn't expect tech spending to return to the levels seen in 2000 until 2003. And even then, he thinks there is little chance of getting back to the average 12% growth in IT spending seen from 1995 to 2000. But stocks are acting like the recovery will come sooner. "There is a gap between analyst-modeled numbers and what the market is saying," he says.

Chang, despite his caution, has a few suggestions. Microsoft, he says, should flourish even in a weak PC environment, given its many new initiatives in areas off the desktop, such as the Xbox gaming console and PocketPC operating system for handheld computers. Chang also likes outsourcing plays, like Computer Sciences and EDS.

Readerman has plenty of picks for the long haul, based on the belief that wireless, video conferencing, distributed computing and streaming media will be important trends this year. His picks include chipmakers Analog Devices, Broadcom and Intel; Sanmina-SCI, a contract manufacturer, and KPMG Consulting. In the telecom sector, he likes Nortel Networks and Sprint PCS.

Roger McNamee, of Integral Capital Partners, in Menlo Park, California, says he's feeling "constructive" about tech stocks, though he has no great hopes for first half earnings. "It's hard to imagine the technology market just runs from here without any stumbles," he says. "December and March quarter numbers will be terrible, and June won't be any great shakes."

McNamee sees the biggest promise in consumer technology. But traditional blue chips, such as EMC, Cisco Systems and Intel are not on the list of his biggest bets. "The big guys are going to produce underwhelming growth," he says. "In the second half of the 'Nineties, you could ride the big franchises. Great stocks and great companies were synonymous. Temporarily, the linkage has been broken."

Among McNamee's current holdings are a pair of disk-drive component makes, Read-Rite and Hutchinson Technology. Those would appear to be well-informed choices, given that he's a partner in Silver Lake Partners, which owns the disk-drive maker Seagate Technology. He's also bullish on the contract manufacturer Flextronics. And his bets on the security trend are ISS and Symantec.

Mark Anderson, a hedge-fund manager and newsletter editor based in Friday Harbor, Washington, is another Microsoft fan, based on optimism about PC demand, as well as the software giant's potential in non PC-devices. As a bet on wireless, Anderson likes Vodafone and Nokia. He's also keen on Samsung, the Korean electronics giant, which he thinks can be a leader in handhelds. And he's bullish on Dell.

Merrill's Steve Milunovich, while no optimist on tech spending, also likes Dell, in anticipation of a PC upgrade cycle. In chips, he picks Maxim and Microchip Technology; as a wireless play, he's fond of Qualcomm. And in networking, Milunovich recommends Extreme Networks and Riverstone Networks.