DSL

DSL Prime News Weekly: The Inside Source —Continued

Other items that played a role: Customers don't care about price

If customers aren't price sensitive, why not charge more? But SBC's own numbers suggest they see a strong response to the price disincentive. They've dropped their forecasted demand about 500,000, with price the major change to explain it.

Protecting T-1 revenue from competition
Data revenues other than DSL are the key driver of telco growth (Verizon was up 30% this year). Much of that is T-1 connections, which really are HDSL and not substantially more expensive to deliver than SDSL lines. But prices are 2-8 times higher, especially out of metros where telcos have little competition. Any switch from T-1 to DSL costs the telco thousands of dollars in revenue foregone; with weaker competition, there is less incentive to cannibalize your own.

Response to court-ordered unbundling of DSL services
An appeals court decision looks to force SBC to unbundle and wholesale DSL services, probably at a 25% discount to competitors. Raising the price of the basic service above projected costs discourages competitors from entering the market, and provides SBC a windfall profit if they do jump in.

Underestimating deliberately to lower Wall Street expectations
SBC and Verizon have badly missed their numbers, a personal embarrassment to Ed Whitacre and Ivan Seidenberg. (It was perhaps even more embarrassing to the NY Times and WSJ, who both without checking picked up stories Verizon met their year end 2000 goals. Actually, the goal was not 500,000, but over 700,000, per a presentation by Seidenberg in the spring. When they fell far behind, they changed guidance later in the year.

But the 540,000 actually achieved, even if over-hyped, showed a strong Q4, implying Verizon is finally on track.) It might be a smart strategy to reduce expectations, so that they can have good news instead. Covad is probably doing the same, Microsoft is notorious for low projections, and we applaud companies that are conservative in their numbers.

Some ideas to reject:
SBC raised the price to reflect the value of the service to customers
Common economic fallacy—that price, in a competitive market, is related to use value. Actually, that's possible only when there is no strong competition visible. Supply and demand rule if there is competition—if there is an alternate supply.

DSL is a much better service than cable, and can charge more
DSL Prime wishes we could agree, but DSL as delivered in the US is not better than cable for most users. DSL Reports has just released a 13,000 user survey that found cable subscribers much happier than DSL users. No telco is even willing to estimate the speed they deliver to the internet, which is the product they sell. (Vague claims of "ultra high speed" are simply false advertising. What speed do you deliver over your network?) While cable speeds drop in peak periods, they are consistently averaging much higher than DSL speeds. DSL reliability, alas, has been poor, and customer service unresponsive. A good operator of either DSL or cable can do a great job; so far, DSL hasn't proven superiority for the mass market.

We choose SBC as company of the year 12 months ago, because last year they outlined plans to actually deliver the 1.5 meg data rates promised throughout their network. They are spending $1.5B upgrading their ATM backbone, to enable those speeds and quality for voice telephony. But they haven't come close to delivering yet, telling customers to be happy with 384K to the local DSLAM, a lower rate that is meaningless to boot. We urge SBC to compete not with press releases but instead by delivering the high quality service they were planning.

No one can make money at $40/month
Bell Canada believes they are profitable at $28 US; Verizon is confident at $40, expecting a price drop; Joe Nacchio of Qwest (who demands a 35% return on investment) sees enough profit in DSL to double subs in 2001. DSL has major economies of scale; profits won't come until millions of customers sign on, which has been planned from the beginning. We don't think they are all wrong.

SBC's operations are irretrievably fouled up
Everyone has been losing heavily on DSL to date, with delays, multiple service calls, high support costs etc. costing a bundle. "DSL Hell" has become a cliche, although perhaps less so in Pac Bell. The telcos' networks especially have been less reliable than expected, and the delivery network has not come up to speed. SBC's back office was so troubled they apparently lost 30,000 orders last summer. The costs, and resulting losses, were far higher than expected. As we write, Verizon's email servers are having problems.

But we believe SBC is well on the way to solving those problems, as the 250,000 lines installed Q4 demonstrate. We've met numerous highly capable technical managers at SBC—they know how to take care of the remaining difficulties. (Fred Chang, in particular, is one of the sharpest folks in the business.) Equipment shortages are now over—in fact, the world has excess capacity in DSLAMs, and Efficient and Westell had to hold in inventory modems manufactured for SBC. Strategic partner Cisco, supplier of much of the network, is delivering more capable switches, and the newer Redbacks have ten times the capabilities of the older models. They just signed Broadjump for a field-proven easy install program. We believe SBC clearly has the ability to grow as planned, if they choose to.

(A new article March 7th suggests I'm wrong about SBC's ability to offer service. Joyce Slaton of the SF Chronicle just wrote a devasting piece "Pac Bell Hell"

http://sfgate.com/cgi-bin/article.cgi?file=/technology/a/2001/03/08/pacbellhell.dtl&nl=top

We'll cover Slaton's report (first spotted by DSL Reports) more intensely after SBC has had time to reply. We again urge SBC to start releasing the basic facts about their DSL service: what's the speed to the internet , how often is it down, how many installs meet the telco standard of 3-5 days, what are the email problems, etc, and how those numbers change each quarter. The usual SBC response—we're doing a good job but won't give out the actual facts—leads any sensible person to conclude the negative reports are true.)

The consequences of the price move
Will price rises spread to cable companies and other telcos?
Everyone in the industry, and the cable competition, has been assuming prices will go down. The cost of delivering fast internet service is dropping dramatically. Chip and equipment costs will follow Moore's law down, dropping by half every 18 to 24 months for the foreseeable future, while the fiber-delivered bit becomes cheaper at an even faster pace. At least one other telco had plans to lower prices from $40 at a predictable time, and a cable broadband expert tells us they were headed down as well.

If Verizon or cable companies were to raise their price as well—contrary to all previous public plans—the conclusion will be obvious: deregulation is failing, and pricing is based on monopoly or cartel economics, not competition. Consumer movements are already arising, spearheaded by the "Broadband Bill of Rights", and further price rises would catalyze political opposition to the telcos.

What can Washington do?
We got no comment from SBC when we asked "How would Ed Whitacre respond if Mike Powell asked him to reverse the price move?", and we expect that Powell's response to the price move will be more subtle in any case. But with Earthlink joining in the price hike (although they dropped a $99 startup fee in partial compensation), and Verizon and Qwest considering doing the same, if DC doesn't move, Powell's goal of "bringing the high-speed Internet to all Americans" suffers a major setback.

Copyright 2001 Dave Burstein.
The DSL Prime Newsletter is reprinted with permission.

"The power of the printing press belongs solely to those who own the presses"
—A.J. Leibling

The Internet is the cheapest printing press ever invented.

1. News

2. DSL Prime News Briefs

3. SBC Justifies DSL Price Hike

4. Analyzing Perplexing Telco Financials

5. Consequences of the DSL Price Hike