OFC/NFOEC: Mergers Haven't Gone Far Enough
That theme was repeated a few times by panelists at yesterday's The Optical Society (OSA) Executive Forum, a day-long conference held alongside OFC/NFOEC.
"It continues to be a problem for our industry that we have so many participants," said Jerry Rawls, CEO of Finisar Corp. (Nasdaq: FNSR), during one of the afternoon panel sessions.
As an example, he cited the chase to develop coherent receivers for 100-Gbit/s long-haul links. Rawls said he'd found himself in a conversation with someone who listed eight companies or partnerships going after that market. "There's no chance eight groups will make money on that," he said.
"We have to eliminate weak business models," agreed Terry Unter, CEO of Mintera Corp.
But this is something the industry has known for a long time. Optical components have been characterized by lots of midsized players struggling just to reach break-even levels. (See Troubles Linger for Optical Components.)
Some companies got taken out by mergers during the past two years. But those deals have involved companies whose products didn't overlap much -- such as Bookham acquiring Avanex to form Oclaro Inc. (Nasdaq: OCLR); the Finisar Corp. (Nasdaq: FNSR) purchase of Optium; or Opnext Inc. (Nasdaq: OPXT) buying Stratalight. (See Finisar & Optium Challenge JDSU, Bookham, Avanex Defend Their Deal, and Opnext Steps Up With StrataLight.)
Round 2 of consolidation has to be more predatory, said Natarajan "Subu" Subrahmanyan, managing director of The Juda Group, a division of Sanders Morris Harris .
"Consolidation is good, but consolidation just for breadth is not," Subrahmanyan said. "It's about taking out competition, and I think the next round of consolidation has to be about that."
But the improving economy might dissuade companies from looking for deals, some panelists noted. Finisar and Oclaro have seen revenues rebound from a tough 2009; Finisar even saw investors take an overallotment on its recent stock offering.
Better business means components companies aren't so desperate for mergers. "I had a guy tell me he felt further away from deals last month than he did a year ago," said Bob Flanagan of Oppenheimer & Co. Inc.
"Now that we're out of the dark ages and into the sunlight again, I don't know what the catalyst is," said Morgan Keegan & Company Inc. analyst Paul Bonenfant.
One thing those mergers did accomplish was to get some components makers more vertically integrated. Avanex, for example, had divested its manufacturing; by becoming Oclaro, it got that capability back via Bookham.
That brought panelists to the idea of vertical integration for the systems companies -- that is, OEMs building their own optical components, just as they used to. "If it was good for the optics guys, maybe it's also good for the optical equipment guys," Bonenfant said.
Some systems vendors are reportedly considering the idea. (See Can Vendors Build Their Optical Components?)
Panelists hesitatingly, and awkwardly, started calling it "reverticalization."
Ray Conley, a partner at Palo Alto Investors, was particularly in favor of the idea. It could give OEMs a chance to develop standout capabilities before a wave of commodity gear arrives, he said.
And it could even improve the design of systems, Conley argued. Commoditized piece-parts are great in the digital world, where a chip behaves the same way no matter what system it's in. But photonics is analog, with optical budgets that can get upset unpredictably as outside components get brought in. "When you're dealing with an inherently analog process, a systems-level design gives you a whole layer of optimizations that a components [view] was not designed to achieve," he said.
The theory has its doubters. Subrahmanyan noted that it's tough to believe a group inside a large company could innovate better than a focused, independent components company.
Moreover, the carriers might not stand for vertical integration. Scott Wilkinson, vice president of product management and systems engineering for Hitachi Communication Technologies America Inc. (Hitachi-CTA) , told Light Reading that his company toyed with developing its own 40-Gbit/s components, only to get a tongue-lashing from one carrier. The carriers love the low prices they pay for lookalike parts that conform to multisource agreements (MSAs) and aren't keen on seeing a components rebellion start up.
But Conley argued that vertical integration might the best way for systems vendors to get out of a continual price war.
"Huawei Technologies Co. Ltd. and ZTE Corp. (Shenzhen: 000063; Hong Kong: 0763) are taking advantage of commoditization of components and are winning because it's a price game," said Conley. "If the other vendors want to have an advantage in a full value-added game, they're going to have to have full vertical integration."
— Craig Matsumoto, West Coast Editor, Light Reading