Is there really a fiber glut?
Source: Network World
Publication date: 2001-09-10
Arrival time: 2001-09-12
Face-off Face-off
Two industry experts debate whether fiber oversupply is fact or fallacy,
ith the free-fall in prices in long-distance telecom over the last year, it's hard to deny there's a fiber glut. Microeconomic rules about supply and demand suggest that when prices destabilize, supply exceeds demand. The steep decline in long-distance prices this year leads to one conclusion: fiber oversupply. After the passage of the Telecommunications Act of 1996,fiber-optic buildouts increased at an unprecedented rate. Many new service providers entered the telecom market and, along with incumbent players, built out fiber coast to coast.Way too many players built out way too much fiber. From 1997 to 2000, the number of miles of fiber-optic cable deployed more than quadrupled.The number of fiber route miles increased from 3 million in 1997 to 12 million by the end of last year More significant was the incremental change in fiber miles added to the installed base of fiber miles each year. New fiber buildouts increased by 239% in 1998, 72% in 1999 and 116% in 2000, according to research done by Friedman, Billings, Ramsey & Co. That's a lot of fiber. At the same time, technological advances in dense wavelength division multiplexing allowed network designers to pump through 160 channels of information per each fiber-optic cable. Network engineers estimate the utilization rate of the new lit fiber is low, perhaps 40%. Carriers usually need network efficiency levels of 60% to 70% per channel to operate profitably.
The service providers underestimated how much fiber would be deployed between 1998 and 2000. Then, making the situation worse, they overestimated demand.They built it and nobody came.
The commodity bandwidth trading market is a leading indicator for pricing direction in the long-distance industry. Prices in the commodity bandwidth market have dropped 80% in nine months, which is more than the average price drop for the telecom industry Another leading indicator of pricing for long-distance is trading volume on popular routes. In the commodity bandwidth market, strong trading volume gives prices a chance to stabilize. But in August, transactions in the commodity market slowed to a crawl. Bandwidth traders claim it is difficult to find any customers for any routes, especially the most popular (that is, overbuilt) routes, such as New York to Los Angeles and New York to London.
As we look into historical capital spending, we realize that 2000 may have been the peak year for fiber buildouts. Carriers will have to pay hefty interest rates to continue to aggressively spend on capital. Their expensive capital structure may raise the bar to break- even and make it difficult to defend the business proposition for more fiber buildouts.
More online!
* Log on to Network World Fusion to register your opinion. FBR's Susan Kala and TeeChoices Dane Brier will add their thoughts to
the discussion. DocFinder 5922
NO By Daniel Briere
You're sitting in bumper-to-bumper traffic on 1-880 in San Jose. Nothing but stopped cars as far as you can see. Then a radio announcer says there's enough space on the nation's highways to handle a fiftyfold increase in traffic, and because of this, financial analysts expect stock in road construction companies and equipment to be depressed for some time.
Arguments like this - but in the fiber world - have sucked the wind out of the optical marketplace. But this argument is flawed, outdated and out of touch, for three reasons.
First, telecom is complexly inter-related. Failures in one part of the industry spread to other parts, and probably no part of telecom will affect the industry's future more than fiber optics.
Second, gluts imply a stagnant market, which mandates slow or no sales, meaning companies have to wait for sales to pick up. As research and development and other growthoriented programs pull back, more of today's network problems will stick with us longer and get much worse.
Third, glut only matters to you if it's in your area. If your regional long-haul links are congested, that means more route- arounds and more latency. Empty roads in North Dakota hardly matter when you're stuck in San Jose traffic.
TeleChoice put together a route-by-route view of what's going on among the major cities in the U.S., indicating:
* Much of industry analysis focuses on unlit fiber in the ground; however, the critical supply variable is available capacity - lit fiber.
* A provider's decision to make additional capacity available is evaluated on a route-by-route basis, not a nationwide basis.
* Having excess fiber in the ground is not a bad thing, despite analysts' arguments.Available capacity must always lead demand by at least 30% per route to avoid the severe capacity constraints and long service waiting lists users in other countries face.
* On 14 of the 22 routes TeleChoice analyzed, indications are that current demand equals or exceeds 70% of total supply across all carriers.
* Even considering unlit fiber, there's not as much as many believe. In two
out of three viable future industry scenarios we've evaluated, the industry completely runs out of fiber on all major intercity routes by 2004.
This is not a reason to start writing our telecom wills.The smart carriers are using tools to accurately predict what's going on. But financial analysts touting nationally averaged numbers and specific application growth figures are painting an inaccurate picture, squeezing the optical space and stifling innovation that will dramatically affect the services you and I ultimately have in our areas.
Briere is CEO of TeleChoice, a market strategy consultancy for the telecommunications industry. He can be reached at dbriere@telechoice.com.
YES By Susan Kalla
Kalla is a senior vice president and equities analyst covering the telecommunications industry at Friedman, Billings, Ramsey, & Co. in Arlington, Va. She can be reached at skalla@fbr.com.
Copyright Network World Inc. Sep 10, 2001
Publication date: 2001-09-10 |
