The current slowdown in the optical components market could signal a period of industry consolidation, but there are still plenty of opportunities on the leading edge of performance and cost reduction, says John Lively, director of optical components research at RHK Inc.

Although the optical components market has seen explosive growth in recent years, a dramatic slowdown hit the market starting in the first quarter of 2001. A report published in May 2001 by RHK s optical components research team forecasts that demand for optical components for DWDM networks (including active and passive devices) will decline by roughly 10% this year versus 2000 revenues, and 2001 revenues will total only $4.5 billion this year. The report identifies the reasons for the decline as over-purchasing of critical components in 2000 (which led to a build up of excess inventory), and a slowdown in 2001 spending on DWDM long-haul networks by North American carriers. Long-term evolution A key question now facing the optical components industry, however, is how it will evolve in the long term. The recent RHK study predicts that the most likely future is that market demand will return to positive growth starting in 2002, with growth rates in the range of 2535% in 20032004. But what will happen to the composition of the industry during this transition from "hypergrowth" to more "normal" growth? Looking at how other industries have evolved is one way to see what could happen in the optical components industry. A comparison with the personal computer (PC) industry would suggest that the optical components industry is headed for a period of industry shakeout and consolidation, while a comparison with the aircraft industry suggests the status quo may prevail for some time to come. Before attempting to predict where the optical components market is head-ing, it is instructive to examine just what the market consists of and how the market has reached its current state. Semiconductor-based optical components can be roughly divided into three major application segments:

1. DWDM optical transport: this has been the fastest growing market segment in recent years, and products in this group include source lasers, photodiode receivers, and pump lasers.
2. SONET/SDH optical transport: as with DWDM, this technology also requires source lasers and detectors.
3. Core and edge IP routers: these also require source lasers and receivers, which are typically packaged as transceivers or transponders.

The components used in these market segments are based on InP and GaAs devices, with the majority fabricated from InGaAs/InP structures. The high-speed, high-bandwidth capability of these devices has been a critical enabler of the rapid capacity expansion of DWDM optical networks. For example, the number of wavelengths carried on a single fiber increased from a single wavelength in 1994 to four in 1995, and by 2001 had reached 176. In addition, over the same time period, the transmission speed of each individual wavelength increased from 2.5 Gbit/s to 40 Gbit/s. The combined effect of the increase in wave-length number and transmission speed has been to increase the potential maximum capacity of a single fiber by a factor of 10, every two to three years, for the last six years. Hypergrowth and expansion The explosive growth in DWDM network capacity was accompanied by a period of "hypergrowth" in demand for the optical components used in those networks. illustrates the historical growth in active optical component sales in the DWDM market, which have been broken out as pump lasers, source lasers, receivers, and external modulators. Since 1998, the aggregate market has increased from just over $500 million to $3.0 billion. The other major optical network market, represented by SONET/SDH, has also experienced very strong growth. While not as rapid as DWDM, SONET/ SDH component sales increased by more than 60% from 1999 to 2000. shows the sales breakdown of both active and passive optical components in 2000, representing a total revenue of $6.3 billion, of which two-thirds was generated by active devices. The impact of this hypergrowth in market demand on industry structure was dramatic. As demand accelerated, severe component shortages appeared. Existing suppliers realized profit margins in the 4050% range. This combination of rapid growth and high margins attracted a number of new entrants to the industry, which ranged from established, related industries like semiconductors and industrial optics, to specialized military and government optics contractors, as well as venture-capital-funded start-ups. The number of the latter is quite impressive over 50 start-ups exist in Europe alone, and probably well over 200 worldwide. An industry transition? As previously noted, in the first quarter of 2001 the music suddenly stopped: system demand growth slowed, and large inventories of purchased parts slowed components sales to a crawl. Such a sudden and large drop in market demand often signals a transition point in the evolution of an industry. This is particularly true with respect to new "disruptive" technologies; a term which certainly describes DWDM optical networks. We can get some ideas for what this could mean for the optical components industry by examining how the composition and structure of other industries has changed as a result of disruptive technologies. One recent example of a disruptive technology is the PC. Not only did it replace the mainframe as the tool of choice for many applications, but it also opened up a whole new range of applications that the mainframe could not address, such as the home computer and small business computer markets. The parallels between the current optical components industry and the PC market period of 1978 to 1984 are striking. During that time, the PC market grew at a CAGR of greater than 100%, and over 150 PC manufacturers came into existence. The early market-share leaders were technological innovators: Apple and IBM. However, in 1985 and 1986, market demand slipped just as new capacity became available. Market growth plummeted to a CAGR of less than 10%, resulting in massive over-capacity, price wars, and widespread industry consolidation. In 1985, Apple Computer, the darling of the early-1980s tech stocks, laid off almost half of its work force. However, as bad as things seemed then, all was not lost. The PC industry is still around. After the rapid decline in growth during 19851986, PC market growth stabilized, leveling off during the 1990s to a growth rate of about 20% per year. During those intervening years there >was significant consolidation, as the 150 would-be manufacturers of PCs were reduced to a dozen or less. Apple and IBM were displaced as market leaders by Dell and Gateway; these companies focused on streamlining their manufacturing and distribution systems, rather than technological innovation. The course of this industry transition is generalized in . The number of manufacturers and industry capacity increases at a ferocious rate during an initial period of rapid market acceptance and wild optimism from investors. Very often, this gives way to an industry shakeout, or period of consolidation, when many companies either merge or fail. After this relatively brief and difficult transition period, the fewer, larger remaining suppliers enjoy a period of more balanced supply and demand, and more rational competition. While the PC industry was going through this evolution, components and subassemblies used by PCs were going through major changes as well. If one views the PC as an analog to an optical network, then optical components are analogous to PC subassemblies such as the disk drive. Over the period from 1980 to 1988, 5.25 inch rigid disk drive shipments increased from zero to some eight million units, an increase somewhat reminiscent of the explosive market growth experienced by optical components over the last several years. In 1989, shipments began to decrease, and in 1995 less than 1 million 5.25 inch disk drives were shipped. The market for rigid disk drives did not go away of course. Instead, faster, higher density, smaller diameter drives were developed. Each new generation experienced a few years of very fast growth, followed by a rapid decrease in demand as the next generation reached the marketplace. This highlights an important lesson for optical components suppliers and investors alike: there will always be high-growth opportunities associated with leading-edge technology, even during periods of reduced overall market growth. Of course, for every clever analogy, there is usually a counter-example, and this case is no exception. Some industry experts, including RHK s own chief analyst, Dr John Ryan, believe the aircraft industry offers a better analogy to the optical components industry. The aircraft industry has undergone tremendous consolidation at the aircraft manufacturer level, until only a few remain, while at the same time the parts suppliers of the aircraft industry have remained numerous and fragmented. This analogy suggests the optical components industry structure could remain unchanged with a plethora of small, niche-focused start-ups surrounding a half-dozen or so broadly based suppliers even as a wave of consolidation sweeps through the telecom carrier and optical transport system levels of the industry. High-growth opportunities still exist Will the optical components industry evolve in the manner followed by the PC industry, and enter a period of consolidation? Or will it more closely resemble the aircraft industry, with a plethora of smaller start-ups coexisting with a half-dozen larger, more broadly based suppliers? Only time will tell. Regardless of which evolutionary path the industry takes during its transition from hypergrowth to a more normal growth phase, it is safe to say there will continue to be high-growth opportunities for leading-edge optical components, where the "leading edge" is defined in terms of both performance and cost reduction. The performance-related opportunities are tied to system trends, such as the transition to 40 Gbit/s transmission; the need for more channels, which require higher pump powers; and the introduction of tunable laser sources, to reduce complexity. In turn, these component trends are being pursued through a variety of semiconductor device improvements, such as higher levels of on-chip integration, the development of 980 nm VCSELs, and the use of MEMS that are coupled with VCSELs to implement tunable laser sources. Current cost-reduction efforts lag behind the leading edge of performance by several product generations because of the industry s focus on rapid product development over the last three years. As a result, there are now many opportunities in manufacturing, integration and automation. Other cost-reduction-related opportunities are tied to innovative product designs, including the use of less expensive uncooled lasers, parallel optics, and plastic housings, which are being developed for the metro market. By focusing on the right opportunities, and investing rationally, optical components companies can weather the current storm, and position themselves to enjoy the period of more rational growth ahead.