By Richard Mack, vice president and general manager, KMI Corp.; George Miller, senior analyst, KMI Corp.
The major issues facing manufacturers of cable and equipment for optical networking include the following:
How bad are things...how weak is the market...how bad is it going to get?
Why are we in this pickle?
When will business improve?
How bad are things?
In terms of growth relative to the previous year and in terms of the actual demand so far this year relative to what was expected, 2001 is unquestionably the worst year in fiber optics since fiber optics was an industry. It's the worst year in telecommunications since KMI has been tracking it (about 20 years).
Quantification of how bad
North America will have negative growth in fiber optic cable installations. In an analysis KMI Corp. completed in August, we said -10 percent. Since then, we are hearing that some of the large customers for fiber, such as the regional Bell companies, are making further reductions in their orders for 2001, and the market will be even weaker.
The North American market for DWDM equipment, which is a major driver for many photonic components and subsystems, will drop 33 percent from $5.9 billion in 2000 to $4.0 billion this year. This market includes both new systems and channel cards, and the new systems portion of this business will fall 50 percent.
The worldwide markets are not much better. Some countries and some regions have different prognoses, but other large fiber optics markets in Western Europe, Korea, and parts of Latin America are showing similar weakness. China and some countries in Latin America will do better this year, but the worldwide markets for cable and equipment will post growth rates lower than ever before. For cable, worldwide installations will grow 7 percent this year, and for DWDM equipment, worldwide shipments will drop 17 percent.
Why are we in this pickle?
The market overspent in recent years; many carriers overbuilt with fiber optic cable and equipment relative to the near-term revenues they could capture using that fiber and equipment. Now there is a correction to spending levels more commensurate with revenues for telecom services. Some of this overspending was financed with debt and equity placements, and the amount of money due on loans is straining many carriers, causing some to shut down.
Why did we overspend (in some cases "over-borrow") or overbuild?
Deregulation provided opportunities for new telecom businesses, and independent investors took interest. The growth in bandwidth demand associated with Internet traffic seemingly offered boundless business opportunities. But the investors and new telecom operators failed to match their investment with the opportunity for several reasons. Bandwidth growth proves unusually difficult to measure and forecast. Further, the bandwidth used to transport Internet traffic cannot be billed to the customer with the same price structure as voice circuits, even though comparable quality is required.
In general, the market for hauling Internet data was a brand new business and, as such, it attracted too many entrants. In the early going of a brand new business, it proved difficult to see how many of these entrants could achieve reasonable returns along with the incumbents from the voice business who could convert more of their network capacity to haul data. So some retrenching is inevitable, though painful.
When will stronger growth resume, or how much did we overspend?
Some are even asking how it could be possible for any near-term rebound, with such corollary questions as:
* Is there a glut of capacity?
* Is there a glut of fiber installed?
* Are bad telecom loans similar to the bad real estate loans that brought down the savings and loan banks about 10 years ago?
* Is the overbuilding of fiber networks similar to the overbuilding of railroads 100 years ago, which resulted in dramatic restructuring of the railroad industry?
This is the tricky part; it's where some statements have been misunderstood and pulled out of context, then repeated in misleading forms. We've seen some extreme changes in market capitalization, investment strategy, and business plans, and these tend to draw attention to the worst-case situations. At the same time, some products and companies are experiencing positive growth and hitting their targets.
The bandwidth capacity business follows a different model than the railroad business, despite analogies that have appeared in the daily press. Fiber plant is installed with the intention of lasting decades -- with full knowledge that only a small percentage of the fiber will be used in the first years following installation, and some of it may never be used. There is no poor business decision making here -- no miscalculation. The idea is that since you don't know how much capacity you'll need in the future, it's better to err on the side of having too much rather than too little. And because of the cost structure associated with fiber cable, having more is of little consequence in the context of the installation costs and the cost of electronics and manpower to light that fiber and make the bandwidth available. The glut now is not one of fiber, cable or bandwidth; it's a glut of carriers trying to achieve profits in a market unable to provide profits to all of them.
Furthermore, not all routes and all network segments suffer from vast excess capacity or over-investment. On some routes, there are network operators who can envision a need to add more fiber or cable within a few years. At a June 2001 KMI conference on undersea cable systems, Global Crossing Development Co. executive advisor William Carter addressed this point specifically. He compared forecasts for capacity demand with the capacity provided by current transatlantic and transpacific cable systems and concluded that demand could again exceed supply by 2002-2003.
No carriers have postulated that bandwidth demands will decrease; no carriers that we are aware of are offering slower Internet connections, less reliable phone service, or more restricted network access as part of their service portfolios.
A United Telecom Council's study on fiber rates supports the view that there is no ubiquitous, uniform glut of capacity. "The conventional wisdom is wrong," said the council's president and chief executive Bill Moroney. "There is no glut. There is, in fact, an ever-increasing demand for fiber where not enough exists today -- in metro markets." The UTC's position is that there are still opportunities for new networks or new operators to prove-in offering fiber-based services. The recent entry of Consolidated Edison Communications into New York City's carrier's carrier business is but one example.
Fiber optics industry participants can't avoid slogging through the inevitable shakeout that follows a period of unrestrained exuberance. It will (and already has) affected both carriers and manufacturers. And it's not over yet. Unfortunately, as a therapist puts it, "the only way out of it is through it."
But fiber optics will get through it. Bandwidth demand is still growing, and nothing provides bandwidth like fiber. Growth in fiber optic products will resume, but not entirely in the same application segments and for the same products that grew well in the late 1990s. Fully competitive, long-distance, fiber networks for intercity backbones are fairly well built out in North America, Europe, and several other geographic markets. The focus is shifting to metropolitan and access networks. The lengths are not as great but the fiber counts are higher, and the quantity, type, and opportunities for node electronics are different.
Some of this shift in product types is already apparent, especially for transmission equipment. Even during this difficult year (2001), there are signs of strength in the demand for products tailored to metropolitan applications. Looking further into the present decade, there is vast potential in the residential access market, although slow progress at present.
The norm for cable demand growth in the 1990s was in the range of 20 percent to 30 percent (worldwide). These rates may never by the norm again, but positive annual growth will resume before the current five-year period is up, and the CAGR going forward from there will probably be in the range of 10 percent to 20 percent.
The growth rates for fiber optic transmission and routing equipment will be higher than those for cable because they will be driven by bandwidth growth -- conceivably reaching levels of 50 percent or better in some product areas. But again, the rates of 100 percent or better enjoyed by some transmission and multiplexing product families in the late-1990s will probably not be encountered again for backbone applications.
About KMI Corp.:
KMI Corporation, a fiber optics market research firm, provides reports on worldwide fiber optics markets, bi-monthly newsletters, fiber optics route maps, and conferences on global, regional and undersea fiber optics markets. For more information, visit www.kmicorp.com.