June 18, 2023

Broadband Market Snapshot Archive

Can Cable Keep Its Broadband Lead?

So far, the big winner in the U.S. broadband arena is clearly the cable industry. Not only have cable operators tied up the lion's share of broadband customers with more than two-thirds of the overall base, but all those customers are essentially new revenue. Cablecos aren't giving up existing dial-up subscribers as they migrate to broadband.

Analysts have recently given cable more encouraging news. Several different studies claim that DSL growth isn't closing the gap with cable and predict that this state of affairs will persist at least into the mid-term future.

However, there are troubling signs on the horizon suggesting that the honeymoon between broadband users and cable operators is coming to an end. Still, in the absence of anything else to shake up the situation, the loss of cable's head start may not be any great news for DSL providers either.

With an occasional exception provided by startups with alternate delivery methods -- usually satellite or wireless -- a would-be residential broadband subscriber has two choices: DSL from his phone company or cable modem service from his cable company. Given the reputations cable and phone operators enjoy among their customers, this is a "lesser of two evils" decision from the beginning. But so far cable is clearly the choice of most customers.

Cable-based broadband is generally easier and quicker to install, and lacks the bizarre provisioning requirements of DSL. If cable passes your house, you can get broadband. Cable operators were wise to push for a retail-shelf, self-install model for service. But for the most part, their advantage seems to have come less from anything they did right than from how low phone companies have driven the competitive bar.

Whatever the cause, the results are clear. A newly released Yankee Group study predicts cable's dominance will continue for at least the next five years. Projections from Brokerage and Investment Bank Jefferies & Co. agree that cable will maintain its lead through 2005. In terms of raw numbers of subscribers added, Jefferies even projects dial-up to narrowly outpace DSL through 2005. New DSL subscribership figures from TeleChoice show steady growth, but the company also agrees that DSL is not closing cable's lead.

However, all those customers aren't just free revenue. Once they've bought the service, they're going to expect to use it, and in so doing they're going to mean lots of capacity demand. There are already signs that cable networks are getting overloaded. Early customers who grew accustomed to relatively uncontested bandwidth are now being told they must upgrade to maintain the speeds they've been getting. As broadband demand continues to grow cable's early success may very well turn on it.

Short of heavy capital spending to increase capacity, cable operators have two main tools for rationing shared bandwidth as demand for the limited supply grows. Both are now coming into more widespread use -- indicating that the problem is a real one -- and both are upsetting customers.

One is tiered pricing, capping users' upstream and downstream transmission speeds to create distinct classes of service. Typically this model creates an inexpensive service running perhaps at 128kbps and a higher-end faster service at a higher price. For example, Canadian cable operator Rogers Communications [RG] has split its Hi- Speed Internet offering into tiers. The new Hi-Speed Lite offers users 128k down and 64k up for $24.95 per month. The faster Hi-Speed product offers 1.5 Mbps down and 192k up, but now costs $44.95 a month, up from $39.95.

Tiered pricing is not strictly a bandwidth control mechanism. It has marketing applications as well and helps providers more accurately model the demographics of broadband customers. However, it also represents a kind of re-definition of cable broadband, a retreat from earlier hype about speed and costs. It also would seem to be connected to increasing user interest in "uncapping" cable modems to remove carrier-imposed speed limits on service.

The second mechanism at hand is more purely a bandwidth rationing tool, and one which carriers don't like to talk about: bit caps. Byte caps are a hard limit on the actual amount of data a customer can download or upload in a given month. The operator's terms of service could specify extra charges accruing on a measured basis if the customer goes over that limit, or simply declare it a violation punishable by service termination.

For the most part, carriers don't like to admit that such things exist, but there are claims that limits are buried in the fine print of service agreements. Stories circulate online from users reporting being cut off for various "outside the norm" activities and being told that they had overrun their data allotment.

Until recently, it seems, byte caps were there, but rarely enforced, on a discretionary basis, to deal with extreme situations. Now, though, byte caps are becoming more overt. AOL Time Warner [AOL] now appears to be waffling on earlier indications that it would impose a usage cap this summer. But others are moving forward. Once again, Canada's Rogers appears to be leading the charge although, oddly enough, it was encouraged by a cap put in place by its DSL-based competitor, Bell Canada.

On May 6, Bell Canada announced a three-tier split of its Sympatico DSL service with prices rising as speeds go up. But each of the three tiers will include a monthly cap, as low as 1 GB for the $29.95 "Basic" package. The $44.95 standard service caps at 5GB and the $69.95 "Ultra" service at 10GB. Customers will be charged 79 cents per 100 MB over their cap.

Rogers hasn't officially announced its response, but executives have said they plan to have byte caps in place by summer, and possibly a further stratification of their service modeled after Bell Canada's three-tier system.

Customers are already increasingly unhappy about unreliable and generally declining speeds. Cable infrastructure is increasingly being oversold and a hardwired usage cap seems to be a particularly hamhanded way of dealing with the situation.

Indeed for many users a capped broadband connection is less useful than an uncapped dial-up connection that costs less. If nothing else, intrusive Internet marketing practices will go from annoying to unacceptable. If every X10 popup ad or HTML-linked spam e-mail brings a user another 100K closer to a usage cap, carriers will start hearing about it.

Presumably carriers would try to set these caps high enough to avoid bothering casual surfers. Bell Canada, for example, claims that 92 percent of its users download an average of 1.5 GB per month, giving them plenty of room under a 5 GB cap. However this ignores the long-term idea behind broadband, as a delivery mechanism for advanced content. The best present example is streaming audio, but increasingly broadband connections are being used to watch online movie trailers and streaming video productions. Sharply limiting available bandwidth to provide minimal service to the greatest number of users will have the effect of strangling those high-bandwidth applications, effectively locking the Internet into the content situation it's in now. That's perhaps sustainable, but also stagnant.

If the goal is to promote overall Internet growth and develop new applications to drive uptake, then byte caps are an astonishingly bad idea. The seriousness with which they're being looked at by cable providers, combined with the growing discontent from crowded users, strongly suggests that the industry knows it is nearing crunch time. The supply of present bandwidth is rapidly being used up, and the carriers aren't about to spend the money to build more.

The result is going to be a lot of cable broadband subscribers looking for a better alternative, and soon. In theory this could be an excellent opportunity for the phone companies to close the gap and get back in the broadband race.

However, this new opportunity isn't the result of anything the phone companies have done themselves. It's not that DSL is getting better but that cable service is getting worse. As long as attitudes and strategies remain the same at the RBOCs, there's no reason to think they'll be able to rise to the occasion. Indeed note that it was Bell Canada that broke the ice on byte caps in its market, making it safe for cable operator Rogers to follow.

So far, DSL remains as much of a nightmare as it ever was. Availability is as much a matter of voodoo. Wait times, especially for line provisioning through a competitive provider, remain geologic. And even if customers can get the service, there's no reason to think it's going to be any better than what they left.

Thus the most likely result of the approaching cable crunch is not a surge in DSL growth -- though there may some narrowing of the overall gap. Instead, it is an even greater slowdown in broadband adoption as customers investigate the service only to find that all options fail to meet their expectations.

This article first appeared in PBI Media’s Broadband Networking News