Snippets from the JPMorgan report, “Backbone!,” September 8, 2000:
Highlights:
· Explosive growth in IP data --- estimated to exceed 100% a year through 2004 --- will soon position IP as the dominant form of traffic on long-haul networks and drive aggregate long-haul traffic growth of over more than 60% annually versus 20% historically, in our view.
· Continued investment in new technologies by transport providers will be necessary to meet demand and lower costs, but we believe network scale and utilization will emerge as the most important determinants of cost competitiveness. As a result, carriers will have large incentives to cut price to compete for traffic on their networks.
· We believe key IP-based services are set to take significant share from legacy voice and data services and expect a disproportionate influence on pricing of these legacy services, resulting in slowing aggregate industry revenue growth.
· The industry will likely face considerable turmoil as challengers seek to take advantage of and accelerate this disruption and incumbents wrestle with cannibalization and restructuring issues.
· These underlying economic forces combined with too many industry players have the potential to drive insufficient aggregate returns on capital, in turn driving consolidation and shakeout.
>>>>
With the inevitable shakeout in the carrier space due to dropping revenues for traditional voice services and increasing network up-grade demands, the key to survival is the ability to offer advanced services at competitive prices.
The report continues:
<<<<
While the overall cost of transport continues to fall in aggregate, numerous debates rage as to the best network architecture and technology. Our own modeling indicates that carriers can reap significant cost savings as they transition to new architectures and technologies. . . . Over the next few years we believe the elimination of SONET and ATM in the core, coupled with more widespread adoption of mesh architectures and additional advances in fiber optics, will simplify networks and reduce both capital and operating costs. Yet at the same time it will be difficult for players to open up a sustained advantage based on technology alone for several reasons. First, networks are rarely pure breed --- almost everyone uses some mix of older and newer technologies. Second, equipment manufacturers have huge incentives to level the playing field between old and new carriers and make the technology widely available to all players. Finally, explosive growth in data volume forces carriers to continually invest in new technology that rapidly supercedes current investments. . . .
We believe that these underlying economic forces will drive consolidation and shakeout. The economics of scale and utilization will tempt players to price close to marginal cost as they first try to fill up their networks with the most accessible revenue pools (e.g., point-to-point, private line, wholesale). Ultimately the price competition will spread to other revenue streams. The capital demands of the industry will continue to be high as carriers continue to deploy the latest technology --- lest they fall too far behind. National networks will be expanded as players extend network reach into the metro, to second-and third-tier cities, and internationally. . . .
Long-haul networks are complex and execution is fraught with challenges. The underlying technology continues to change very rapidly. Overall volume demand continues to grow at breakneck pace. The systems and IT challenges are huge. The competition for skilled personnel is intense. The rate of applications and business innovation continues to grow, and the competitive landscape is constantly shifting as players seek sustainable positions. Against this rapidly changing backdrop, the quality of execution will be a monumental challenge. . . .
[Where the rubber meets the road:]
As in most rapidly changing markets, we believe winners will come from the most unanticipated directions and success is neither guaranteed nor unachievable for any of today’s major players. However, our basic conclusion of this report is that “network Nirvana” is not yet upon us. The hard reality of network economics, service pricing, margin dynamics, and execution excellence will take front and center stage from the capital markets euphoria of the last two to three years.
>>>>
The next segment of the report discusses the actual network.
>>>>
From the Top
Technology is revolutionizing the economics of long-haul transport. The effects of new technology have been most visible in the infrastructure and equipment industries. In essence, technology advances have simplified and will continue to simplify the layers of a typical network "stack," and at the same time deliver dramatic improvements within each layer. As Chart 15 shows, today's stack typically layers IP over ATM in the switching domain, with the physical layer composed of SONET over DWDM over fiber. But over the next couple of years, those five layers are expected to collapse into just three, with transformation at the switching and physical layers. Within switching, IP over ATM is expected to give way to IP alone, with MPLS essentially replacing ATM functionality. Hitherto, IP required ATM in order to provide quality of service via connection set-up, traffic control, and packet prioritization. But in MPLS, carriers have found a QoS protocol that in essence adds in ATM-like controls, thus allowing carriers to remain IP but without the current quality and service tradeoffs associated with a "best-efforts" network.
At the physical level, we expect SONET over DWDM to give way to what is often called the Intelligent Optical Network, or ION (not to be confused with Sprint's ION products). The ION allows forever-faster transport at ever-lower costs by essentially replacing an entire level of expensive SONET equipment with software. ION equipment provides the same functionality as older SONET equipment --- like reliability and service restoration --- but also adds new functionality like point-and-click provisioning and network performance management.
That's not All, Folks
Nor is the delayering the end of the story. As Chart 15 shows, there are dramatic improvements taking place within each network layer as well. Switching and routing speeds are now approaching speeds of 40Gbps versus speeds of 2.5Gbps, which seemed astonishing only four to five years ago. At the physical layer, DWDM's continual improvement cycle has taken colors per fiber path to 160 from 8 over the same period. This, in combination with improved fiber technologies that decrease signal attenuation and dispersion, has allowed fiber capacity to increase 80 times that of five years ago while simultaneously decreasing the need for costly regeneration.
[Think of SDL as controlling penicillin in the middle of a global typhoid epidemic. Carriers are experiencing a lot of pain and need DWDM efficiencies to compete.]
Demanding Customers
Meanwhile demand --- in particular demand for IP-based services --- is exploding. Industry participants routinely cite growth rates of 100% in public, with Internet traffic doubling every three to six months. While you can quibble with specific prognostications or about just what's being measured (demand, capacity, network depth. . .), the claim of hyper-growth is generally supported by carriers' recent and planned capacity expansions as well as by customer purchases of bandwidth. Clearly, though, not everything in the network is growing at the same speed. As Chart 16 makes clear, maintenance of current growth (which may be a stretch for the older networks in light of the IP migration) should drive staggering changes in traffic mix. Namely, voice will likely decline from its current 51% (down from about three-quarters two years ago) to under 10% --- perhaps less. And the balance will lie almost exclusively in IP, especially as much of what traverses the private line networks is IP anyway. . . .
[The report concludes:]
Death, Taxes, and Earnings
The enduring telecom legacy of the late 1990's lies in the cheap capital and favorable regulatory environment that spawned the great wave of challengers in the backbone space (and just about every other area of telecom). We believe the early 2000s will be shaped more by how those new carriers --- the like sof Level(3), Qwest, Williams, Enron, Global Crossing, Touch America, and others --- can shift their focus away from nearly completed network buildout and more toward filling their networks with sizeable and profitable traffic, especially as the need to meet investor expectations will only grow more intense from here. The incumbents have also entered a critical phase of decision making relative to capital structure and business strategy: invest or harvest, cannibalize or protect, realign management and investor interests through restructure or leverage integration capabilities and scale.
Within this context, strong execution will become increasingly critical for both sets of players. For incumbents, operating within a rapidly transforming environment will likely create enormous challenge, especially amid potential financial restructuring and cost-cutting initiatives. But challengers, too, will likely face execution risk. That is, it's one thing to build and light a network and to enter enter the wholesale pipe market. It's another still to build a large marketing and distribution force and to scale network provisioning and management systems --- all while maintaining high levels of service quality. Underlying nearly all of it are questions surrounding technology, how it will change, what it will enable, what advantage it can confer, and how it will creat or meet demand.
* * *
The truth is that many questions don't have a simple answer --- reality is more complicated. In this report, we've tried to take a stab at answering some questions about debates raging in the industry and at providing industry participants and investors with a roadmap for underlying change and its overall potential impact on the industry.
>>>>>
[All transcribing errors mine.]
Pat |
