The Inevitable Consolidation of the Optical Components Industry
As mergers and acquisitions run rampant through the communications industry, even the highly fragmented optical component industry is consolidating. In this month's column, Fiber Optics Online guest editor William Magill explains why modularization is inevitable and how system performance advancements make it more urgent.
By: William Magill, NationsBanc Montgomery Securities
The planned merger of Uniphase and JDS Fitel brings together two leading players in the optical components sector and marks a consolidation trend in the industry. The marriage is a strong validation of this trend, given that both bride and groom (you pick) have been enjoying single life with strong growth and profits and record-high stock valuations.
The trend toward cost containment by the suppliers' customers favors suppliers with higher degrees of component integration than each company provides by itself. Cross-pollination of their distinctly different product portfolios is the fastest means to achieve the integration customers seek. As single-product companies find it increasingly difficult to compete in the optical component supply market alone, they will seek out partners.
The logic of modularization
WDM systems companies are no more interested in assembling their solutions from discrete components than are automobile companies in piecing together engines from each ring and gasket. Like telecom systems, car engines are comprised of a number of subsystems (intake, ignition, exhaust, cooling, etc.), each made up from 10s or 100s of components. Manufacturers buy these subsystems on an OEM basis, then bolt them together to build engines.
The subsystem approach results in tremendous savings in assembly and test time, labor expense, and inventory management. Likewise, telecom systems designers want to combine optical subsystems for their wavelength division multiplexing (WDM) engines (end terminals and amplifiers), and move away from the laborious assemblage of lasers, isolators, collimator lenses, and so on.
Conflict between higher tech and lower cost
The term "WDM system" is expanding from simple point-point equipment to full-blown network architectures. Products are evolving from low-channel-count (16 or less) terminal multiplexers with mid-span amplifiers to higher-channel-count (40, 80, and more) platforms with routing and traffic management functionality. This evolution places much greater complexity and performance demands on component and system designs.
At the same time, bringing product to market faster and at a lower price than the competition remains paramount in the equipment marketplace. In this environment, systems companies will favor suppliers with integrated solutions (with an emphasis on solutions) versus discrete components.
Consider this hypothetical situation. A systems design engineer, needing post-gain amplification for a WDM product, sits through two sales meetings. Supplier A offers a single amplifier module with guaranteed performance metrics regarding gain, signal-to-noise, spectral range, amplification linearity, and lifetime. Supplier B opens a catalog of two dozen components that may generate the same results if chosen and connected properly—which, of course, will require time for design, assembly, test, and possibly redesign. Option A is clearly the preferred solution, and becomes even more attractive when the buyer considers labor costs.
Modules with both active and passive components
To date, modularization has primarily involved the combination of two or three active components or passive components, with little fusion between the two camps. There is no logical argument to keep actives and passives separate, but an increasingly compelling argument to combine them.
Consider next-generation optical add-drop multiplexers (O-ADMs). While O-ADM designs vary by company, all share a common theme of peeling off or adding wavelengths on a selective, dynamic basis from a bidirectional stream of WDM traffic. This process requires a complex combination of selective filters, such as Bragg gratings and circulators, placed in close proximity to switches, and tunable or wavelength-selectable transmitters (lasers) and receivers (photodiodes).
A tight synchronization between the filtering and switching mechanism (inherently passive) and the light generation and detection mechanism (inherently active) is required for proper O-ADM system operation. It makes sense for these collections of components to come from the same supplier, and, better yet, to come packaged in modules. The system designer, then, only needs to provide the intelligence and control that make these modules work with the rest of the system and network.
Component industry implications
The optical components industry is highly fragmented today, but it will consolidate over time. Small specialty component developers—and there are many— develop much of the industry's leading-edge technology, such as planar devices and tunable sources. These specialty firms will eventually be hard-pressed to sell to systems integrators that favor modules over discrete components. Larger module designers with broad component catalogues will pull the boutique operations into the fold through acquisition and secure access to the most promising technologies while denying access to the competition.
Opportunities for private investment in this environment are particularly exciting, given the mix of funding opportunities and exit possibilities available through merger and acquisition. The trick, as always, will be in finding innovative and distinct technology ideas that are also manufacturable, scalable, and profitable. But that is a topic for another column.
About the author…
New Fiber Optics Onlinecolumnist William Magill, principal with NationsBanc Montgomery Securities LLC, offers a Wall Street perspective on the optical networking industry. He can be reached at wmagill@montgomery.com.