Copyright 1999 Federal News Service, Inc.
Federal News Service
JULY 14, 1999, WEDNESDAY
SECTION: IN THE NEWS
LENGTH:
2335 words
HEADLINE: PREPARED TESTIMONY OF
ANNA-MARIA KOVACS
FIRST VICE PRESIDENT
JANNEY MONTGOMERY SCOTT
BEFORE THE SENATE JUDICIARY COMMITTEE
ANTITRUST,
BUSINESS RIGHTS AND COMPETITION SUBCOMMITTEE JULY 14, 1999
BODY:
Thank you for giving me the
opportunity to appear before you today to discuss consumer choice in the
broadband Internet-access market. My name is Anna-Maria Kovacs and I am the
telecommunications and broadband services analyst for Janney Montgomery Scott, a
brokerage and investment banking firm. My job is to make judgments about the
strategies, business plans and financials of firms within the telecommunications
and cable industries and to gauge their chances for success in the face of their
competitors' strategies and plans, so that I can help our investors pick stocks
within this industry.
One of my observations is that the Telecommunications
Act is on the verge of bearing the kind of fruit that was hoped for when the Act
passed. Meaningful competition in various segments of the telecommunications
market has finally emerged, and where it existed before the Act it is taking
increasingly strong hold.
Facilities-based competition against the incumbent
telcos is becoming a reality in the consumer as well as business markets.
Competition in the business markets was real even before the Act passed. Cable-
telephony has been deployed in large enough volume to assure us that competition
in residential markets is becoming real, as well. While mass deployment may take
a couple of years and another generation of technology, there can be no doubt
that it is coming. Video competition has become a reality, with satellite-based
services providing real competition to cable. Competition for high-speed
Internet access is becoming a reality as well. Cable modems are proliferating
and telcos have begun to deploy DSL, with both technologies
able to run data at speeds 20 or more times those we have been accustomed to. In
the backbone, long-haul segment, several new competitors are creating networks
each of which provide more potential capacity than the total current traffic
requires. Wireless is beginning to replace wireline.
The major players in
each of these segments are trying to play in all segments, as they prepare for a
world in which they expect a large part of the market to require bundled
services. Thus, they are moving from their traditional areas of strength into
new areas, concerned that they will not be able to defend their original
position unless they are equally competitive in the other segments.
Cable
companies have upgraded their video capabilities and moved into data carriage
and have begun to move into voice carriage because they see threats to their
traditional markets from satellite-based video and need the new sources of
growth that Internet access and telephony provide them. Telcos have accelerated
their deployment of DSL in response to the threat posed by
cable modems, which could decimate the telcos' second-line growth unless the
telcos can offer a product that offers competitive speed. The long-distance
carriers have moved both into Internet-based value-added services and into local
markets, AT&T most notably with its enormous investments in the cable
industry. In each case, the presence of a real, facilities-based competitor has
spurred the incumbent to move more rapidly to provide new technologies, products
and services.
That is an important lesson, in my view, to keep in mind as we
look at the Internet access market and concern ourselves with ways to insure
that the Internet continues to flourish and that there is unimpeded access to it
by both consumers and content providers.
Today, there are two primary ways
to access the Internet. The vast majority of users do so over the telcos'
networks. Some do so at high speed, most often off corporate networks. Millions
of consumers do so at relatively low speed, generally at or below 56 kilobits,
though as many as 100,000 consumers may be gaining high-speed access via
DSL. The telco network these customers use provides
point-to-point connections to any of thousands of ISPs, who in turn provide
access to a plethora of websites that hold the actual content the customers want
to reach. Slightly under a million consumers reach the Internet over cable
networks, at speeds that may reach a megabit or more. They generally have direct
access to one Internet Service Provider, @Home or Road Runner, through whom they
may reach other ISPs and the content of the Web. From the consumer's standpoint,
today's choices can be roughly described as low-speed and the ISP of my choice
on my telco or high-speed and a single ISP on my cable, for a more or less
comparable total price of about $40 for the connection and ISP. Speed vs. ISP of
my choice.
That clear-cut choice, however, is blurring. It is becoming
possible in more and more locations to get high-speed on the telco via
DSL, and @Home has made it possible to access other ISPs
through it, albeit at an extra charge. In other words, there are real
technologies deployed in the field that make it realistic to expect that within
a year or two, most consumers will be able to get high-speed access to the
Internet via at least two media, cable and telco-DSL.
The
ability of cable to offer high-speed is spurring telcos' deployment of
comparable speed even though it is not necessarily economic at this early stage
in DSL's learning curve. I believe that the deployment of
DSL, in turn, will spur the cable industry to insure that it
offers consumers a choice in content, content providers, and gateways that is
comparable to what the telcos can offer.
In other words, I believe that
consumers, given a choice of two media which offer equally high speed at
comparable prices will select the provider that gives them the content and ISP
of their choice. The best guarantee that consumers will enjoy the benefits of
broadband and the content of their choice, and that content providers will have
access to all consumers, is to do everything possible to encourage both sides to
deploy as vigorously as technology, human resources, and capital allow.
Both
sides face some barriers on each of those fronts. DSL is a
difficult technology to deploy. It is sensitive to distance from the central
office as well as to the quality of the loop, and current versions of it are not
compatible with the digital loop carrier that the best modernized telcos have
deployed. All of these make it expensive to deploy and account for the slowness
with which it has reached the field. Once it is deployed, however, it provides a
secure, point-to-point connection whose speed is predictable and controllable.
Some of these problems will disappear as new generations of
DSL come to market over the next year or two, thus increasing
the market that can be physically targeted and lowering the cost of deployment.
A factor that will lower deployment cost for both DSL and
cable-modems is the appearance of PCs that are DSL-and/or
cable-ready. Those have begun to come to market and will help to further lower
deployment cost and alleviate the human-resource problem -- the shortage of
competent technicians who today have to go out and install either cable-modems
or DSL directly into the PC. Thus, it is reasonable to foresee
that at some point during 2000, DSL deployment will kick into
high gear, which I would define as passing the million customer mark that
cable-modems are already approaching. By that point, cable companies will have
to face the fact that telcos can provide a product that is equally attractive in
terms of speed and price to cable-modems.
Consumers will no longer face the
current choice of speed vs. my favorite ISP, but will be able to get both over
DSL.
At that point the pressure will be on cable to open
access to its network, a task that faces some real technology barriers. Cable
networks are shared pipes.
Because they are shared, it becomes difficult to
control the actual speed any user will enjoy when multiple users are on-line.
@Home and Road Runner are able to some extent to control bandwidth allocation,
to ensure that a few customers do not hog the entire pipe and exclude all
others. There is today no network management system that can do that
bandwidth-allocation job when many ISPs are providing service over the cable
network directly to the end user during periods when the network is carrying a
full load. It is likely that such an operating system could be developed for
cable networks, but it is not here today. Hence, the cable industry's insistence
that other ISPs use @Home or Road Runner as their gateway to the customer.
There are many who insist that the cable industry is motivated to limit or
control access to its network not only by technical difficulties but by
anti-competitive motives. The potential for that certainly exists given the
vertical integration in this industry and the small number of horizontal players
providing local access. It does not take much imagination to envision the
potential for a player like AT&T that controls access to the majority of
cable homes in the U.S. through its own properties or its affiliates, which is a
part-owner of @Home and will be of Road Runner, and which has a variety of
content properties, finding ways to advantage its own content and sites on its
own network. But it also does not take much knowledge of history to understand
that in a competitive market that is likely to be a highly self-destructive
strategy. Consumers who, at comparable prices and speeds, can get unlimited
choice of content over the telcos vs. limited choice over their cable network
are not likely to opt for the cable network. Beta vs. VHS and Apple vs.
Microsoft both tell us that customers primarily care about content and
applications and will flock to the vendor that gives them the best and widest
selection of each. Thus, if AT&T were inclined to try to limit the number of
ISPs and the content on its network, it would be punished severely by the market
place, assuming there is another choice in that marketplace. Most Internet
access would happen over the telcos' DSL pipes. Given the
enormity of AT&T's investment in cable systems and its inability to earn
adequately over those systems without a hefty penetration of cable-modems and
telephony, its stock would suffer severely if it maintained a closed-access
strategy once DSL is readily available in the market-place.
The key, then, to ensuring that the cable industry, and especially AT&T
which has invested so heavily in its cable networks, do not act in ways that
anti-competitive against ISPs and content providers is to ensure that it has a
real competitor at the network level. That is, the key is to ensure that
DSL can be deployed as efficiently, economically, and rapidly
as possible. That will put pressure on the cable industry to open its network.
Ultimately that means creating new cable-network operating systems that allow
network capacity control to be distributed among multiple ISPs. In the short
run, it may mean reaching agreements with ISPs that enable them to look to the
consumer like the primary ISP even when @Home is actually providing the network
control Regulators can also have some impact on the speed of deployment on each
side. On the DSL side, rules that are likely to discourage
deployment by the telcos themselves include the requirement that telcos to have
separate data subsidiaries, that they provide competitors with a portion of the
spectrum on the line on an unbundled basis, that they provide collocation for
DSLAMs in already-crowded field-vaults. Each of these makes it operations more
difficult and expensive for the telco. On the other hand, each of these
facilitates deployment by Data LECs who ride on the telco's network. If the
primary need is to encourage as much DSL deployment as possible
to put pressure on cable operators to open their networks, then the key question
in considering such regulations has to be whether more DSL will
be deployed by the telcos themselves, if they are left free of regulation, or by
the DLECs, if they are helped by such regulations.
Similarly, regulators can
have some impact on cable deployment. It is unlikely that cable will refuse to
upgrade its networks in the face of regulation. AT&T, in particular, has
already spent so much on buying TCI and will spend so much more on MediaOne,
that it has no choice but to upgrade its networks to make as much money as it
can on Internet access and telephony. However, regulations that do not take into
account actual technological realities could slow deployment.
Forcing kluged
solutions to allow multiple ISPs direct access to customers before an effective
operating system is ready would be one such possibility, because it could
increase expense and might degrade service and therefore the marketability of
cable Internet access.
How Wall Street allocates capital within this
industry, or more simply how stock prices will move, will depend on the
development of these various technologies, on the strategies chosen by the
various players, and on regulation as well. To focus most specifically on the
latter with some examples, minimizing regulations on the telcos is likely to
help their stocks, but is likely to hurt the Covads and other Data LECs who
provide DSL over the telco networks. Immediate open-access
rules are likely to help the stocks of ISPs other than @Home, and to hurt
@Home's as well as to some extent cable stocks. That means that regulators need
to be very clear on what their over-riding goals are, and to balance short-term
vs. long-term goals. Is it more important to pit telcos vs. cable to ensure that
each side is as aggressive as possible right now or is it more important to
promote the health of the Data LECs? Is it critical to ensure open-access on
cable today via regulations that might impose extra expenses on cable companies
and will probably damage the financial health of @Home, or is it possible to
wait and see whether DSL-based competition takes care of the
problem? How regulators answer these questions will help determine which
companies and industry segments receive support from investors.
END
LOAD-DATE: August 25, 1999
