SPECIAL REPORTS: Optical Communications Market
U.S. carrier revenue peaked in 1998, but capital expenditures did not reach their height until 2000, leading to capex cuts and an unstable market.
By MEGHAN FULLER
When asked about their survival strategies in the current capital-expenditure (capex) crunch, equipment vendors' answers range from investing in a lifetime supply of antacids to a firm belief in the power of prayer. But is the situation really so dire?
If you're the kind of person who sees the glass as half full, then yes, there's hope; carriers have already completed the bulk of their capex cuts. If you tend to see the glass as half empty, brace yourself; capex spending could get lower, and the major carriers aren't likely to make significant purchases of new gear until 2003, at the earliest. In the meantime, what is an equipment vendor to do?
Roots of the problem
The current economic malaise can be traced to the Telecommunications Act of 1996, which deregulated the industry and triggered the creation of competitive local-exchange carriers (CLECs). Sensing unlimited opportunity, the capital markets were all too happy to fund them.
At the same time, Internet traffic was growing exponentially-200% per year at its height-and the carriers all saw revenue increases. Expecting traffic growth and the ensuing dot-com boom to boost revenues even higher, service providers ramped up their spending.
Revenue peaked in 1998, but capital expenditures did not peak until 2000. Further complicating matters, many of the new-builds were long-term projects. "Even by 2000, when people were realizing that things were slowing down, telecom companies were still locked into their 2-3-year contracts to complete their networks," explains Melanie Swan, director of the Telecom Economics Program at RHK Inc. (San Francisco).The bulk of those networks was completed in 2001, when carrier capex spending was $77 billion, down from the previous year's monster $97 billion, but still high nonetheless. "What we expect this year," contends Swan, "is somewhere in the range of $46-51 billion, which is a 38% decrease from the $77 billion last year" (see Figure 1).
Who's cutting spending, and by how much?
The "Big 7" carriers-Verizon, SBC Communications, BellSouth, Qwest Communications, AT&T, WorldCom, and Sprint-issued guidance on capital expenditures early this year, and in some cases, more than once. In February, BellSouth knocked an additional $500,000 off its already-reduced 2002 estimates, citing continued weakness in demand, primarily in the domestic telecommunications market. Sprint also lowered the expected capex of its FONS Group (its core wireline telecom operation) from $3.4 billion to $3 billion. According to Communications Industry Researchers (CIR) Inc. (Charlottesville, VA), Verizon could announce further spending cuts as well.
The good news, in terms of overall spending levels, is that most carriers have already adjusted to more manageable ratios of capex to revenue, says David Gross, senior analyst of optical-networking research at CIR. The historical norm of capex to revenue is 20%; a carrier with $20 billion in annual revenues that spends $4 billion a year on capital should be able to service any debt, make interest payments, and cover operating expenses with relative ease.
In the optical heyday of 1999-2000, however, some carriers spent upwards of 30-40% of revenue on capital expenditures, which is unheard of, claims Muayyad Al-Chalabi, director of executive strategic partnerships at RHK. "No business does that," he asserts.
Much of the frenetic spending was not done by the traditional incumbent local-exchange carriers (ILECs), but rather by the CLECs and emerging interexchange carriers (IXCs). Level 3, for example, spent 400% of revenue on capital expenditures, which meant nearly all of its capex had to be financed. That said, even the Big 7 U.S. carriers needed to make reductions to bring their spending levels into conformance with the 20% ratio (see Table on page 12).
But it's not just a question of the spending levels returning; there are other factors at play in today's telecom economy. "They are too many companies out there that have a decent amount of cash in the bank but no customers or a very limited number of customers," says Gross. "Right now, there's just a very painful, slow dwindling of cash, and I think the market will be better served when the companies that really are going to survive re-emerge as the leaders."
Further compounding the problem is excess inventory. The demise of the CLECs in particular has led to the indefeasible rights of use (IRUs) they had signed with wholesale providers suddenly becoming excess capacity, which then becomes inventory that has to be worked off.
How they'd like to spend money
Yet, despite substantial reductions in capex, excess inventory, and market instability, the Big 7 carriers will continue to spend-the question is, on what?
For major carriers, the objective is to determine where they want to go in terms of technology and spend accordingly, claims Ray Smet, vice president of network transformation at BellSouth (Atlanta). "You're skating toward where the puck is going, so to speak," he reasons. To accomplish that, BellSouth has divided its investment into three categories: target, leverageable, and non-leverageable.
Smets defines target technologies as anything that is technically feasible and economically viable, citing broadband, optical, and packet technology as examples. "If a technology is ready and the economics are right and the market is ready to accept it, we clearly would like to move our capital investment in that direction," he explains.
That said, the bulk of a carrier's investment is in the leverageable category and includes the purchase of such technologies as circuit switches, short copper links instead of fiber, and other items carriers would traditionally buy. These items are still a good investment says Smets, though he adds that carriers manage these assets very carefully. "You really do want to limit your spending where you can," he suggests, "by either finding ways to reuse the asset that you've already deployed or by looking for the most efficient way to service that particular piece of the business."
As a general rule, carriers keep their spending to a minimum in the non-leverageable category; they spend here only when they have no alternative. "This may be servicing an old piece of equipment that we just haven't been able to replace, or we've grandfathered some kind of capability in the network and we have to continue to support and maintain it," explains Smet. This category usually includes older analog services that may still be in production.
"We force the non-leverageable spend to 7% or lower; definitely less than 10% of our total investment," says Smets. "The target investment we want to maximize as high as we can possibly go, but we believe that a reasonable investment level is probably somewhere around 30% to 40%. So leverageable is the largest-chunk spend, but it's still a good investment when there really isn't any good alternatives that are ready to be deployed."
How carriers will spend money
According to Gross, the major trend, in terms of carrier spending, may come as a surprise. "There really isn't a major technology shift going on in their spending," he says. "In the metro, we hear a lot about metro DWDM-and they are certainly looking at it. There are some RFPs [requests for proposal] out there, but in terms of metro optical transport, they are still looking primarily at SONET." The traditional technologies, he says, should do very well this year.
The Big 7 carriers also continue to be interested in data services, claims Ashoka Valia, senior vice president of business development and corporate strategy at Nortel Networks (Ottawa, Ontario). They have good solid margins on voice, he explains, but voice revenue growth is relatively flat, ±3% or 4%. It's a bit of a conundrum, he admits. "You are in the low single digits of growth in terms of voice, but that's where all the margins are. The second part is the growth in data-IP and broadband and those capabilities. They experience high growth but low margins."
While Al-Chalabi acknowledges that carriers probably won't spend a significant amount on data services this year, he believes that it will continue to be a key growth area for them nonetheless. BellSouth, for example, has seen its data revenue jump from $1.5 billion five years ago to $4.5 billion in 2001, a compound annual growth rate of almost 32%. The carrier expects its data services revenue to grow 22-25% this year. Sprint, meanwhile, saw its IP revenues increase 9% in 2001 over the previous year. (For more on data services trends, see "Putting the value-add in data services," page 14.)
InterLATA or interstate services are also high on many carriers' wish lists, as is the ability to offer wavelength services. "We saw that a few months ago when SBC filed its MON [metropolitan-optical-network] tariff, and there was some talk that maybe the RBOCs would start to be more aggressive with wavelength services," explains CIR's Gross. "But given the persistent problems on the CLEC side, I don't think it's going to lead to significant DWDM spending."
The bottom line, he says, is that the major carriers will be more focused on traditional services as opposed to new ones. How will this affect the vendors? "We're looking at the same old vendors and maybe upgraded or higher-density versions of the same products that will continue to do well," Gross contends. "It's great if you're an Alcatel, Tellabs, or Fujitsu, but it's not the story everyone wants to hear."
Some companies look to put a twist in that plot. "There are a lot of vendors out there that want to change the way a carrier thinks about building a network or offering a new service," surmises Chad Dunn, co-founder and director of product management at Wavesmith Networks Inc. (Acton, MA). "And that's a tough sell in this kind of environment."
Today's vendors need to be more pragmatic, asserts Nortel's Valia. They have to deal with the reality of the carriers' embedded base and embedded operational support systems (OSSs). They must ensure that carriers will not lose any of the functionality or performance they have come to expect. But first and foremost, equipment vendors must help carriers save money.
Thus, the successful vendor will be able to articulate cost arguments, which may not be as easy as it sounds. "All of the vendors are saying, 'Our solution offers opex [operating-expediture] and capex savings,'" reports RHK's Swan. "The successful vendors need to get to the next level and actually demonstrate how the capex and opex can be achieved in the payback or an ROI [return-on-investment] analysis."
While it is important to reduce carriers' capex, vendors have to reduce carriers' opex as well, says Valia. "Networking operating expenditures are about a 2:1 ratio to capex," he estimates. "So it costs twice as much in opex expense-provisioning, maintenance, real estate, power-than it does in capex expense."
To further ensure survival, equipment vendors really have to target the Big 7. According to data recently released by Merrill Lynch, these carriers will account for close to 90% of all wireline capital spending in the United States this year and next year (see Figure 2). To do that, vendors must deliver a carrier-grade infrastructure-robust, reliable equipment that can scale to serve the carriers' millions of customers.
For startups, penetrating this market can be hampered by the expense and elongated process time of OSMINE, a series of Telcordia-run tests designed to assure carriers that a vendor's equipment will integrate with the carriers' existing OSSs. "If it's a technology that will replace some traditional level of service or it's the next generation of an existing technology that we're going to offer, we want to scale it," asserts BellSouth's Smet. "The OSMINE process helps us accelerate the delivery of new technology by making sure that technology will work with our systems."
Most vendors, it seems, regard OSMINE as a necessary evil and accept the high price and long process time as inevitable. The RBOCs require it, says Dunn, "and if you're not selling to them, who are you selling to?"
Greg Wortman, vice president of marketing at Coriolis Networks (Boxboro, MA), says vendors also need to be targeting the carriers' funded applications. Greenfield networks are-at least temporarily-a thing of the past; carriers are in build-to-demand mode. Funded applications tend to be smaller contracts, valued at $200,000 to $3 million to $4 million per project. While that may be pocket change to the Lucents, Fujitsus, and Nortels of the world, it could mean the difference between success and failure for a startup, states Wortman. "We don't need very many of those to make our revenue plans," he contends.
Light at the end of the tunnel?
Industry insiders agree that any semblance of a recovery will not appear this year. RHK does not expect significant orders of new gear until the first or second quarter of next year. "On the equipment side, we're still going through the worst of it," agrees CIR's Gross. "Anything that looks like a recovery-meaning revenue increases-is going to be 2003. We're just not finished with all the capex cuts."
"You have to be very judicious in your spending, there's no doubt about that," advises Steve Kemp, group manager in the global marketing department at Tellabs Inc. (Lisle, IL). "But I'd like to think that the last few years have taught us all a lesson, kind of slapped us upside the head a little bit and reminded us that in the end, you still need to pay attention to the business."