2002 Top Ten: Components Catastrophes
Components vendors have been hit hard by the industry recession, as that old, familiar tune plays again: Too much hype led to oversupply, and when the orders dried up, the vendors were stuck with huge inventory backlogs and unsustainable cost structures.
How things have changed. In 2000 and 2001, systems vendors were hoping to make a tidy sum by spinning out their components businesses as separate entities. Finances aside, this was felt to be necessary to compete on a level playing field with the likes of JDS Uniphase Corp. (Nasdaq: JDSU; Toronto: JDU), which had set itself up as the first big independent supplier of components.
By 2002, components had become such a huge millstone around the necks of systems vendors that most would happily give the businesses away -- and some of them practically did. JDSU itself seems to be hanging on by its fingertips. As a result, there have been radical changes in the competitive landscape.
But perhaps there is light at the end of the tunnel. If analysts are to be believed, then the industry has finally hit bottom, and inventory backlogs are close to being exhausted. That only leaves one small problem to be worked out in 2003: There are still more vendors than even a normal, healthy market can sustain.
So with that moderately optimistic thought in mind, let's review this year's most memorable components calamities.
No. 10: More Massive Job Losses
JDS Uniphase's latest workforce tally is 8,000, down from 20,000 in mid 2001 (see JDS Woes Kick Up More Intel Talk). Alcatel Optronics (Nasdaq: ALAO; Paris: CGO.PA) has announced plans for a 58% staff reduction -- reducing the number of employees to under 500 (see Alcatel Optronics to be Slashed). Chip vendors are not immune from layoffs either: Broadcom Corp. (Nasdaq: BRCM) will cut staff by 500, or roughly 20% (see Axe Falls at Broadcom); Vitesse Semiconductor Corp. (Nasdaq: VTSS) cut 200 of its 1160 staff (see Vitesse Drops Some Packets). Need we go on?
No. 9: JDSU Recalls EDFAs
Early in the year, evidence came to light that JDS Uniphase Corp. (Nasdaq: JDSU; Toronto: JDU) had to replace hundreds of defective Erbium Doped-Fiber Amplifiers (EDFAs) in gear that its customer had shipped to at least one carrier (see JDSU in EDFA Recall). The problem was attributed to degradation of epoxy in the path of a high-power optical signal. JDSU claimed that the problem had been discovered and remedied some time ago (see JDSU Says EDFAs Fixed). But industry observers speculate that the problem could run much deeper -- epoxy is still used in components, as a suitable alternative isn't always available.
No. 8: Bell Labs at Sea
Bell Labs has been the source of many key inventions in communications, such as the transistor and the laser -- and where would we be without those? But the former world-leading research institute got cast adrift when Lucent spun out its components business as Agere Systems (NYSE: AGR/A), a transaction that was eventually completed in June of this year. Who is Bell Labs researching for now, and more importantly, who is paying for it? According to Bell Labs' Website, the division boasts 16,000 employees -- which, going by the latest headcount targets, accounts for nearly half of Lucent's remaining employees! (See Post-Bubble Arrogance.)
No. 7: WaveSplatter?
If one company represents a microcosm of what's happened to this industry, it has to be WaveSplitter Technologies Inc. Starting from humble origins as a producer of passive widgets like couplers and splitters, the company jumped on the latest craze -- Arrayed Waveguide Gratings (AWGs) -- and pumped itself up for an IPO, loading itself with top-level executives in preparation. Then pop! -- it missed the IPO window and the market dried up. Today, WaveSplitter has "cut back to the bare bones," according to its former chief exec, and hopes to start over in Taiwan, making (guess what) boring passive widgets again (see WaveSplutter?).
No. 6: ADC Axes Components
ADC Telecommunications Inc. (Nasdaq: ADCT) made its mark in components by spending $872 million on Swedish startup Altitun for its tunable lasers, then another $80 million on a Danish manufacturer of phase masks, Ibsen Photonics A/S -- viewed as quite a coup at the time (see ADC Scores a Coup on Tunable Lasers). Well, it's easy to be wise after the fact. Ibsen survived (just barely) because its management raised enough cash to buy the company back, but Altitun got the chop (see Ibsen Makes a Comeback and ADC Tunes Out). The remaining part of ADC's components business, a division in Australia making passives, got sold to Verrillon Inc. for an undisclosed -- but probably tiny -- sum (see Verrillon Gets Active in Passives). No. 5: Avanex/Oplink Merger Implodes
The proposed merger of Avanex Corp. (Nasdaq: AVNX), and Oplink Communications Inc. (Nasdaq: OPLK) could have rescued two components players in one fell swoop -- but since they're included on this list, obviously it didn't work out. Both vendors saw strength in size, and their product areas were highly complementary (see Avanex/Oplink Merger: Tip of the Iceberg?). The merger was proposed in April, but it was cancelled in August, with the departure of Avanex CTO Simon Cao and a lawsuit with a Taiwanese contract manufacturer cited as reasons (see Avanex and Oplink: Wedding's Off). Since then, the only course of action open to the pair appears to be more executive shuffling and layoffs (see Avanex Gets New CEO and Oplink 'Restructures' Half Its Staff).
No. 4: Lucent Kills LambdaRouter
Tiny tilting mirrors may sound cool, but they don't sell. After spending several years talking up the LambdaRouter all-optical switch and selling only a couple of units, Lucent Technologies Inc. (NYSE: LU) finally gave up the ghost. The technology will be returned to the lab -- where some say it should have stayed all along (see Lucent Terminates the LambdaRouter).
Perhaps an honorable mention should go to Nortel Networks Corp. (NYSE/Toronto: NT), which in 2000 paid $3.25 billion in stock for Xros, an all-optical switching startup (see Nortel Buys Xros). Nortel had been planning to develop a monster all-optical crossconnect based on Xros's technology, but it canned the project this past March (see Nortel Shuts Optical Switch Effort).
No. 3: Anything to Do With AWGs
Once upon a time, any developer of Arrayed Waveguide Gratings (AWGs) could claim to be ahead of the technology curve. Now these widgets, which split and combine wavelengths or colors of light, seem to spell doom. Why? Based on the mass-manufacturing techniques of the electronics industry, AWG technology is only cheap if produced in large volumes -- and right now, the demand has simply vanished. The investment in manufacturing facilities has suffocated several players, including one of the leaders, Lightwave Microsystems (see Obituary: Lightwave Microsystems), and also contributed to WaveSplitter's troubles. Agere, Nortel, and Infineon Technologies AG (NYSE/Frankfurt: IFX) have also fled from the AWG curse (see Agere Favors Thin-Film Filters, Nortel Nixes Passive Components, and Infineon Spins Off AWG Business).
P.S.: A tidbit that should have made it into our 2002 Top Ten: Forecasting Follies: At the beginning of 2002, ElectroniCast Corp. predicted a 57.8 percent increase in AWG sales this year (see Another Surge for Waveguides?). Whoops!
No. 2: The Nortel/Coretek Saga
Nortel's purchase of Coretek, a tunable laser startup, for $1.43 billion in 2000 was one of the defining moments of the optical bubble -- along with its purchase of Xros (see above). Unlike the Xros product, Coretek's technology actually made it to market; the outcome, however, was the same. Nortel ended up selling the majority of its components business to Bookham Technology plc (Nasdaq: BKHM; London: BHM) for a total consideration of $111 million a few months ago (see Bookham Buys Nortel's Components Biz). The sale excluded the former Coretek, whose technology appears to have been buried -- possibly on purpose (see Coretek Is Closed). Multiple sources suggest that part of the deal with Bookham involved keeping Coretek's tunable laser technology (which would compete with Bookham's) out of the market.
No. 1: Agere's Exit From Opto
The numbers say it all. Agere, the former number two in components, sold its optoelectronics business for a mere $40 million in cash to TriQuint Semiconductor Inc. (Nasdaq: TQNT) (see TriQuint to Acquire Agere's Optics). For Triquint, it's a real bargain, but for Agere, it looks like a disaster. The best that can be said is that it was a better option than closing the business down.
— Pauline Rigby, Senior Editor, Light Reading