[ strategy broadband ] prev next
Consumer DSL: Are US Carriers Crying Poor All the Way to the Bank?
(Please note: Originally published in October 2002, exhibits from this report can be seen in the downloadable .pdf copy)
October 2002
Just One Last Drink
It conjures the image of an alcoholic rummaging through the liquor cabinet in a burning house, hoping to simultaneously douse the flames and his sorrow with a cocktail
There is a growing and suspiciously broad consensus that regulatory tools (albeit with very different ends in mind) are what’s called for to put out the telecom fire. This month:
- Scott Kriens, CEO of Juniper Networks, an equipment vendor, suggests: “Were that [regulatory] freedom available to the carriers, we would see relief from what is an embarrassing situation.”
- AT&T is petitioning the FCC to reverse a two-year old decision to deregulate access charges, with CEO David Dorman drawing attention to the $5 billion in “excess” profit local carriers enjoy from special access services vs. allowing interconnection via functionally equivalent (but cheaper) UNE-P network elements.
- SBC, principal advocate of the failed Tauzin-Dingell bill, and ring leader among incumbents in the case against UNE-P (attributing many of their layoffs to this required wholesale arrangement), continues to press its case, now bolstered by the hiring of Dorothy Attwood, former chief of the Wireline Competition Bureau at the FCC
With such diverse constituents appealing to regulatory intervention, one might easily conclude that “DSL-killing” UNE-P and all manner of other plagues are choking off a business that is, without regulatory change, fundamentally uneconomic.
Beneath the Regulatory Addiction
DSL internet access has a bad name in the US. Consumer experience in the initial years was often poor, and a generation of early competitive carriers (the so-called DLECs, or Data Local Exchange Carriers) died. Both the negative consumer experience and DLEC failures were largely for the same reasons — very high cost, failure-prone provisioning processes
After a rocky start, DSL continues to enjoy double-digit growth rates (Exhibit 1) with costs per gross add (CPGA) cut in half since large-scale service roll-out in 2000, and continuing a sharp decline (Exhibit 4).
Penetration rates in markets where DSL service is available match the upper end of adoption rates experienced for new technologies, such as personal computers, and vastly exceed the first five-year adoption of new services like cable television.
Bad Signal-to-Noise Ratio
Incumbent carriers believe, or would have others believe, that DSL is a bad business based on four generalized complaints:
- Competitor freeloading: UNE-P resale allows competitors to profit from reselling an expensive investment, purchased at wholesale prices. Incumbent carriers thus perceive a disincentive to aggressive DSL roll-out constrained by a regulatory cost accounting framework which prevails for the majority of incumbent wholesale arrangements (e.g. TELRIC)
- MORE capex?!: if subscribers use “too much” bandwidth, carriers are faced with an expensive set of capacity upgrades, especially from the DSLAM to the core network
- Losing the (“unfair”) cable battle: cable carriers have an almost insurmountable 2:1 lead in market penetration (Exhibit 1) and historically enjoyed lower incremental costs, both at the network level, and at the operating level for provisioning, etc.
While telcos will not explicitly concede defeat, lest it contradict other arguments, this belief is reflected instead in vigorous lobbying to put cable MSOs on equal wholesaling terms requiring open access to ISPs as a means (regulatory, again) to throw obstacles in front of cable modem growth - Broadband the Cannibal: Incumbent telco core businesses are shrinking (Exhibit 2). Some of the blame is assigned to second-line disconnects driven by DSL (and wireless substitution), shifting customers to a lower net revenue, “unprofitable” product.
The noise of these arguments, in which there is an omnipresent theme of regulatory “relief”, obscures the fact that incumbent DSL is about at cash flow break-even and will contribute to net income next year (Exhibit 3). i2 Partners’ model of a nominal 500,000 subscriber, mid-size urban market served by 50 wire centers shows 7%+ net income contribution in 2003.
It is worth noting that no incumbent carrier provides segment disclosure for consumer broadband — only net subscriber additions are reported. Internal ILEC cost accounting practices are not good at clarifying the economics of businesses that are not structurally separated (e.g. wireless), or that share common infrastructure (e.g. central offices). The i2 Partners study we summarize here employed three complementary approaches to gain visibility into “true” broadband economics, net of myriad internal cost allocations in a multi-segment business:
- bottom-up development of an economic model including key cost trends in the infrastructure supply chain, costing of wire center, outside plant, and provisioning operations, as well as competitive pricing assumptions
- review of ILEC state filings to shed light on actual wholesale UNE-P price caps, as well as further interpolating cost data (despite largely “confidential” treatment accorded by state agencies)
- assessing the financials of more transparent (often non-US) broadband carriers.
Why DSL is a Good Business
DSL is a good business because, like broadband access in general, demand is high, is of the “right” kind, and costs to serve are rapidly declining. DSL may not be as good a business as cable broadband but, despite a two-year lag in mass deployment, DSL economics are nearly comparable and improving rapidly.
Demand and Usage
Overall consumer demand for broadband access remains high (Exhibit 1). As important, monthly time logged on the internet by broadband subscribers has shown only a modest increase over dial-up usage, and average bandwidth consumption per subscriber is at the low-end of what “mid-band” access services support
Broadband access offers consumers a clear, incremental benefit — basically, more convenience. Consumer satisfaction is quite high despite the failure to realize the pre-millennial visions of instant movie distribution and matter transportation. So debates over absolute bandwidth and fears of soon-to-be gridlock and network congestion from streaming video, etc., have been both misplaced and largely irrelevant
As significant increases in per-subscriber bandwidth consumption do materialize, tiered pricing has already emerged as a management tool, allowing subscribers to chose a price appropriate to their bandwidth consumption, and driving penetration in the (larger) more price-sensitive market segments.
Cost Trends and Capital
Upgrading network infrastructure will be done at increasingly low incremental cost (Exhibits 4,5), with correspondingly large decreases in IP transit costs. Much of this dramatic decrease in hardware costs is attributable to rapid progress on the experience curve from large overseas deployments where, unlike the US, DSL is the preferred choice for consumer broadband.
The majority of the non-CPE capital required for expansion will be “success-based”: expenditures contingent on actual increases in paid connections which trigger corresponding increases in network capacity.
Why DSL is an Essential Business
DSL is an essential business because without it large incumbent carriers have next to no long-term future. Despite the telecom meltdown, it remains true that many services, ultimately including significant portions of voice, will be delivered through broadband access networks. The key scarce asset, enjoyed only by ILECs and cable MSOs, is the local network — tens of millions of individual physical connections to end users. Whoever converts the most of those wires to broadband wires, wins. The endgame for carriers is to do for broadband what they did for the PSTN — extract maximum long-term value
Cable companies are extracting a great deal of value from their physical plant. Their core business — delivering video entertainment — is expanding. Average revenue per subscriber from the core business alone is enjoying growth from an ever expanding channel lineup, greater numbers of premium channels, pay-per-view, and now video-on-demand.
In addition, cable enjoys a still-expanding 2:1 advantage in broadband penetration and has begun launching cable telephony offers which, while expected to be adopted quite slowly, pose another fundamental threat to incumbent telcos.
By contrast, telco core businesses are eroding in what we believe to be a gradual, long-term secular trend, not simply a function of the economic downturn, as broadband and wireless drive residential second line (or even primary line) disconnects.
In the face of multiplying alternatives, telco customers must be given a reason to keep their telephone wire, use it more often, and for more services
DSL is the conduit not to just broadband, but everything else: retention of local voice, adding on ILEC-provided long-distance, and provisioning enhanced voice or hybrid PSTN/internet applications in the future.
The Distraction of “Value-Added”
During the “full service network” fad of the early ‘90s, ILECs combined to create Americast, a video-entertainment company. Really nothing more than a buyers’ club for programming, Americast was briefly the centerpiece of ILEC video “broadband” strategy. Each carrier prepared grand plans for both the infrastructure (usually VDSL) and content which would catapult them into head-to-head competition with cable’s own core business and put the holy grail of The Bundle (voice, video, data) in their grasp
The Americast episode has been rerun many times in smaller ways with different aspirations (internet portals, electronic directories, etc.) over the last ten years. The lure of “moving up the value chain” and “adding value” had been irresistible, at least to telco strategic planners, and telco CEOs at investment banking conferences
The argument appeared to be: sure, the wire is valuable, but only if we also charge for a lot more stuff, ideally exclusively-owned stuff, from businesses we are not in.
This line of thinking also drove such successes as Exite@Home, and in some measure AOL Time Warner.
Utility Is Not a Dirty Word
We have demonstrated that DSL today is a good business, about to get better. This despite being a distant second to cable, having no value-added content, and being subject to forcible wholesaling in today’s regulatory framework, at prices the incumbent cannot directly control.
The path for carrier growth lies in becoming (or remaining) the connectivity utility, where the carrier’s wires are used for a progressively broader array of services, not necessarily originated by the carrier, attracting higher spend per subscriber.
Being a connectivity utility in a competitive environment means focusing relentlessly on the following three drivers, in this order:
- Connections: broadband access is a proven, desirable consumer product whose ultimate penetration will be around 70% of households. Scale and profitability are directly correlated, thus it is essential to have as many connections as possible before market share “hardens”, acquisition costs increase, and competitors raise customer switching costs with more attractive services.
Currently, the key barrier to increased scale is price. The US has among the highest broadband access prices for a product with significant demand elasticity. Our analysis, drawing from our model and such relationships as Qwest/MSN, and AOL DSL resale shows substantial profitability of even wholesale customers, and assuming significant price declines (Exhibit 5).
There is plenty of room to buy share through lower or tiered prices, and points of wholesale distribution should be actively encouraged. - Usage: Increasing usage has two parts: getting people to want applications which use more connectivity (a long-term marketing activity analogous to Intel’s), and providing a tiered pricing structure which enables customers to upgrade their usage levels while receiving dramatically increased value. Usage-based pricing should encourage low-priced entry, and regularly increase (like flat-rate cellular) the included “minutes” for the same price point. Such price plans would be structured around peak/off-peak usage, throughput, or latency. While simplicity is essential, these parameters have not been even begun to be fully applied in the marketplace.
- Applications: It’s not that video is a bad idea per se, but video service largely developed and originated by a telco is. Applications developed “in-house” should have a clear linkage to connectivity and communication which are, in both the customers mind and in actual capabilities, core competencies of a communications provider. Applications generally can be sourced from many potential providers and profitably resold (see our report: “Transformation Interrupted: Can Telecom Be Saved From Itself?”) Some applications (high-bandwidth video) will benefit from “closed” network architecture in which performance is optimized at the edge of the carrier’s own network, as well as providing a measure of exclusivity or differentiated performance.
Regulatory Combat: A New Distraction?
DSL is a profitable business and a critical step on the path for carriers to become one of two mass-market connectivity players. Closing much of the gap with cable is eminently possible, but will require dedicated focus on the fundamentals of being a utility — not being distracted by unrelated services or complementary but separate bundling initiatives, nor being confused by one’s own political posturing.
Meanwhile, there’s a lot of money to be made.■
Download this article:: USConsumerDSL.pdf [421.5kb]
1 October 02
