SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : AUTOHOME, Inc -- Ignore unavailable to you. Want to Upgrade?


To: Moose who wrote (17128)11/18/1999 10:30:00 PM
From: thedewar  Respond to of 29970
 
From todays Barrons:

NOVEMBER 18, 1999

DSL Is Poised For a Big Comeback

By Carolyn Whelan

1999 was the year cable modems emerged as the high-speed access route of choice for Internet
users (see Weekday Trader, "Excite Deal Shows Cable Modem Is King," January 19).

Meanwhile the Regional Bell Operating Companies, or Baby Bells,
dragged their feet on a competing technology, Digital Subscriber
Line (DSL), which bulks up existing copper phone lines for
broadband usage. (DSL is between six and 125 times as fast as the
current standard 56 KB modem, according to the Yankee Group.)

"The whole sector was stillborn," says Joel Achramowicz, an analyst at Preferred Capital Markets

But that's changing: Cable infrastructure upgrades are taking longer than many thought, while
renewed commitment by phone companies to high-speed access is helping DSL turn things
around. Thursday's ruling by the Federal Communications Commission ordering local telephone
companies to share their networks with competitors will also accelerate the move to DSL.

"DSL is making a comeback ," says Karuna Uppal, an analyst at the Yankee Group. Very
aggressive rollouts by the Regional Bell Operating Companies were unheard of six months ago,
she adds.

SBC Communications, for example, is spending $6 billion to blanket 80% of its region with DSL
access, and Bell Atlantic hopes to have 100,000 residential DSL subscribers by the end of 1999.

"ADSL installations have really picked up the pace," says Steven Levy, an analyst at Lehman
Brothers. In fact, they could rise tenfold this year, to 500,000, from 50,000 in 1998, he says.

Cable modems definitely have a head start -- 80% of homes with high-speed Internet access have
them, vs. 20% with DSL lines -- but DSL is catching up.

"We're seeing a lot more DSL shipments than we expected," says Cathie Hackler, an analyst at
San Jose-based Dataquest, a unit of the Gartner Group.

More importantly, the higher revenue stream DSL promises from business users, few of whom
have cable access, means DSL may have the last laugh.

"The business segment is going to lead the way [to DSL]," says Chandan Sarkar, an analyst at the
Soundview Technology Group.

One promising application called Voice Over DSL allows users to bypass local phone companies
while getting both high-speed Internet access and plain-old voice telephone service through
competitive local exchange carriers (CLECs). That can help businesses cut their phone bills by
25% to 50%.

Today's FCC ruling will make the CLECs' efforts much easier and more affordable. By sharing
the Baby Bells' networks, they won't have to put in new lines, and could pass those savings on to
customers.

"This should speed up the whole process," says Achramowicz. "This pushes CLECs to be more
aggressive which should cause the RBOCs to be very aggressive as well."

That's good news for makers of DSL equipment, particularly those that make faster and cheaper
versions of DSL.

Those new, sexy flavors include G.Lite and HDSL2. G.Lite, which is used in homes, is
significant because installations can be done remotely, without the huge expense of sending a truck
to the home.

HDSL2, which is used in businesses, is twice as fast as the next-speediest form of DSL and as
much as 125x as fast as a 56K modem. Both will drive DSL adoption, says Michael Pericka, an
analyst at Olde Discount Corp.

"G-Lite could propel DSL to the mass market by bringing price points down, enabling access to
copper and the ability to compete on a price level with cable modems," he says.

Who will benefit the most? Pairgain Technologies has transitioned to providing HDSL2 from older
T1 lines, which are about as fast but are at least three times more costly than DSL.

Pericka, who rates the stock Accumulate with a
price target of 19 (it closed Thursday at 16 3/8),
believes the shares are undervalued. "A lot of the
upside in Pairgain lies with HDSL2," he says.

Pairgain, with a market capitalization of $1.1
billion, is nudging its 52-week high and has an
astronomical P/E of 149x estimated 2000 earnings
of 11 cents a share. (It won't make money this
year.) But investors like Sarkar expect Pairgain's
impending sale of some of its less profitable units
to unlock more value.

Levy also likes Orckit Communications (with a
market capitalization of $475 million), which he
rates a Buy, with a price target of 40 (it closed
Thursday at 26 13/16). "They're the third largest
vendor of DSL equipment in the world, behind
Alcatel and Cisco," he says. Orckit has 17% of the
global market, according to Dataquest, and its customers include Fujitsu, GTE and Deutsche
Telecom.

Orckit's revenues from DSL more than doubled to $15 million in the third quarter of 1999, from
$7 million in 1998.

"The market is just starting to take off," says Orckit president Izhak Tamir.

Orckit is 32% off its high of 38 7/16, and changes
hands at a P/E of 15x expected 2001 earnings of
$1.80, according to First Call. That's a nice
discount to its projected 25% annual earnings
growth rate.

Another favorite is Adtran, which has a market
capitalization of $1.6 billion. Adtran makes line
cards (which hardware carriers install at their
centers for high-speed access) for lower-speed
ADSL modems.

Levy considers Adtran undervalued. "The rest of
the group sells at ten times revenues, and [it] sells
at a discount," he says.

At 40 5/8, Adtran trades near its high and has a
high P/E of 32x expected 1999 earnings of $1.25
(though it's a discount to the group average of 70x
1999 earnings). But at 26x the $1.56 it is expected to earn in 2000, the stock is only at a slight
premium to its projected 20% long-term annual earnings growth rate, and sells at a nice discount to
its five-year average P/E of 41x.

On the down side, competition is fierce, and DSL
pure plays will be more exposed if other, sexier
technologies emerge. But cable modem suppliers
have to grapple with those issues, too.

"It's going to be an interesting horse race next
year," says Achramowicz. "I wouldn't be
surprised if [DSL] makes up a lot of ground
against the cable guys."

Investors are beginning to agree.

thedewar



To: Moose who wrote (17128)11/18/1999 11:18:00 PM
From: E. Davies  Respond to of 29970
 
What about the cost to fix either problem?
I would be curious about that as well, but I can take a guess or two that would indicate that the issues are pretty similar.

@home has said that they have already built into the network the capability to split nodes for future expansion since the major cost is the initial laying of the wires. The incremental cost to split nodes is apparently fairly small.

I suspect that once you get past "the last mile" what happens at a CO or a cable head end are pretty similar and have the same issues with building capacity. I would think that since @home runs their own national network they have an advantage in both economies of scale and performance when it comes to the backbone.

I still believe that the @home national network designed from the beginning for broadband will someday stand out as a major advantage over the majority of DSL providers. It all depends on whether the core of "the net" as is can handle broadband. I doubt it will.

Eric



To: Moose who wrote (17128)11/18/1999 11:26:00 PM
From: ahhaha  Respond to of 29970
 
You're creating a question which hasn't enough specificity to be answered. It is an interesting contention that cable enjoys a cost advantage at any scale. The cost differential may substantially widen as a function of scale. It's difficult to judge because there are intangibles like threshold throughput for added value service.

Consider 10 DSL CLECs sharing a CO with an ILEC. Who's going to buy the cooler? In fact, who's going to pay the rent on the bigger building and re-provision the premises? Who's going to pay for the terabit 'bone that has to be installed to support the star? What impact will this have on local circuit switching? These are just a few of the problems.

Eric did a fair job on your previous question, so I won't add quite a few other issues. If I thought any of this was an issue of X takes from Y, I'd go into detail. The two markets are evolving like I thought they would, but the RBOCs fooled me with their reticence. AOL too. Whatever they think they're doing, they're still fiddling. If they go to court against today's DSL sharing decision, Rome will be in ashes.