BT Alex. Brown on ARTT
----------------------------------------------------------------------
ADVANCED RADIO TELECOM CORP. (ARTT) "BUY"
Initiating Research Coverage With A "Buy" Rating -Part 1/2
----------------------------------------------------------------------
Date: 09/28/1998 EPS: 1997A 1998E 1999E
Price: 4.0 1Q (0.79) (0.50) (0.70)
52-Wk Range: 18 - 3 2Q (0.65) (0.42) A (0.70)
Ann Dividend:0.0 3Q (0.66) (0.53) (0.73)
Ann Div Yld: 0.0 4Q (1.11) (0.61) (0.77)
Mkt Cap (mm):117 FY(Dec.) (3.22) (2.07) (2.90)
3-Yr Growth: FY P/EPS NM NM NM
CY EPS (3.22) (2.07) (2.90)
Est. Changed Yes CY P/EPS NM NM NM
----------------------------------------------------------------------------
---
HIGHLIGHTS:
-- We are initiating coverage of Advanced Radio Telecom (ART) with a
"2-buy" inv (often referred to, incorrectly in ART's case, as the
wireless competitive local exchange carrier, or CLEC) with a
nationwide footprint of licenses to offer high-speed data services,
primarily to business users.
-- The local business phone market is estimated at $35 billion
annually, which we believe will grow to $60 billion in 10 years.
Today, the incumbent local exchange carriers hold a virtual monopoly
on local phone service. We believe as competition heats up, CLECs can
ultimately win around 50% market share, up from essentially zero
today, and perhaps an even greater share of the data and private
network business.
-- We expect so-called point-to-multipoint (PMP) technology to be
introduced commercially later this year that will significantly
improve the returns on high-end subscribers and open up the lower-end
market for penetration. This event will likely mark the true
emergence of wireless as a viable local loop strategy.
-- ART began commercial operations in mid-September with the launch of
the Seattle market as a wireless Internet Service Provider.
-- Stock Price Performance: ARTT has fallen 57% YTD, versus a 22%
rise in our CLEC index and an 8% gain in the S&P 500.
-- Valuation: Based on our 10-year DCF, using a 25% equity discount
rate to reflect an imminent need for capital and a 10x terminating
multiple, our 12-month price objective for ARTT is $13/share. Despite
the attractive upside potential, we believe the need for incremental
capital is a significant source of risk, and hence offer a "2-buy"
rating.
DETAILS:
INDUSTRY POINTS
-- The wireless access industry is not a battle between the existing CLECs.
The real competitors are the incumbent local exchange carrier (ILECs)--in
most
cases the regional bell operating companies (RBOCs)--which currently control
nearly 90% of the total access lines in the U.S. As with most CLECs, the
wireless access companies' (WAC) initial focus is on the business market,
which
we estimate spends $35 billion annually on local phone service, of which a
small, but rapidly growing portion, is data.
-- We believe the wireline and wireless CLECs taken together could command
50%
market share of local phone revenue within 10 years, up from essentially 0%
today, and perhaps an even greater, if today unquantifiable, portion of data
and private network traffic.
-- The principal alternative to broadband wireless access is fiber optic
lines
(although that may change in the future). Of the estimated 750,000
commercial
office buildings in the U.S., only 3% are connected to fiber today (with
some
higher percentage actually "passed" by or touched by fiber). Wireless
economics allow for provisioning of smaller buildings/accounts that are not
close to the fiber ring.
-- Once a network hub, or node, is located in a market, the pool of
potential
customers includes all buildings with which a line-of-sight can be
established.
Therefore, the buildout costs for a market is directly related to the
demand
in that market. In other words, incremental capital expenditures are
success-based.
-- As a company's data needs expand from simple dial-up access to high-speed
data and multimedia and video conferencing, wireless access providers have
the
ability to scale the amount of bandwidth to the customer commensurately,
meaning it can capture incremental communications revenue at little or no
incremental cost. Adding additional capacity does not require a hardware
change.
-- Traditional CLECs are largely dependent on the existing ILEC
infrastructure,
while wireless access providers can completely bypass the copper facility.
-- By late 1998 or early 1999, point-to-multipoint technology (PMP) should
begin to complement the point-to-point (PP) technology that is available
today.
Basically, PMP cuts the capital cost of adding a subscriber roughly in half
by
eliminating one of the two radios necessary to complete a "link" (which
allows
the carrier to target lower-end subscribers). Wireless Access Companies
(WACs)
retain the benefit of increased spectral efficiency by allowing dynamic
bandwidth allocation, while an overlaid PP link efficiently serves
bandwidth-hungry customers.
--We believe these companies represent a consolidation play, as well. It is
likely just a matter of time before a major long-distance carrier seizes on
the
opportunity to completely bypass the Regional Bell Operating Companies
(RBOCs)
in offering its own local services, or before an international carrier sees
the
opportunity to gain a toe-hold in the $140 billion U.S. telecom industry.
-- We expect a fair amount of deviation from our models and even stated
objectives because these businesses are so new, but we expect the stocks to
surge, as the stories become "reality" over the next several quarters.
COMPANY DETAIL
ART owns licenses in the 38 GHz band in 90 of the top 100 markets covering
approximately 186 million POPs and approximately 50%-plus of the nations
business access lines. ART is not a CLEC, in the traditional sense of
competing for "access lines" and "voice traffic" against the ILEC. ART's
focus
is centered on the data market and provisioning high-speed Internet and
private
network services to business customers over a 100% IP-enabled ATM network.
In
that sense, ART is still competing for "access" to the customer, but from a
data perspective.
In early 1998, ART abandoned its original carrier's carrier strategy, in
which
it would provide wholesale capacity on its high-speed network as requested
by
resellers, in favor of the new data strategy and cut its buildout schedule
down
to three markets in 1998 and 9 more in 1999. ART's beta trial in Bellevue,
WA
became a commercial market in mid-September, with Phoenix and Portland to
follow by year-end. ART is not reselling its primary services to the end
user.
Unlike other wireless access carriers, ART will not offer voice service
until
voice-over-IP technology matures.
We believe ART's initial voice product will be LD-origination and could
follow
one of two paths. Either ART could aggregate traffic for a LD company and
provide access to the customers at a discount to the ILEC rates (remember,
ART
has what the IXC does not-direct access to the customer), or it could resell
cheap LD and pay the IXC for long haul and termination. ART has never been
a
"CLEC", though it still faces the same fight against the RBOCs as its CLEC
"cousins".
Recently, ART completed a 21-customer trial in Bellevue. The company
launched
commercial service as an ISP on 15-Sept with five buildings and the ART hub
with traffic backhauled from the hub to a router and internet POP in
Seattle.
For subscribers without a LAN, which accounts for about half of the beta
subs,
ART is selling a "LAN-in-a-box" solution. Access speeds on the LANs range
from
64 kbps to 10 Mbps in the trial. We fully expect this market to be
successful
in proving out the company's technology and business strategy. If ART can
raise the necessary capital, the return on that capital could be the highest
in
the group, given the extremely low cost of entering a market. We estimate a
data switch (router) costs less than 10% as much as a voice switch. From
there, most expenses are success-based.
Projections
We believe ART can achieve high-50% gross margins with a data-only strategy.
We have to acknowledge that voice-over-IP is likely just a year or two away,
but have not modeled any contribution from voice services at this time.
ART's
network architecture is well suited for the pay-by-the-bit model we believe
communications companies will eventually adhere to.
By 2007, we expect ART to be providing data access and private network
services
to almost 90,000 corporate subscribers at an average speed of 2.8 Mbps per
subscriber (the equivalent of 1.8 T1 lines per sub). We are modeling two
revenue streams for ART until voice-over-IP becomes more imminent: data
services and internet services such as web page-hosting. Based on ART's
efficient ATM network, we believe it will reach gross margins of 63% by
2007,
with a corresponding EBITDA margin of 33%.
Financial Position
We expect ART to reach operating cash flow breakeven in early-2002 after
incurring incremental losses beginning 3Q 1998 of approximately $1.0 billion
as
follows: $174 million EBITDA loss, $736 million CapEx requirements, and $74
million of cash interest payments.
The ART war chest currently amounts to $40 million in cash and short-term
investments and the bulk of the Lucent facility described below. Art has
one
bond issue outstanding, a $135 million Sr. Notes issue with a 14% coupon due
2007. As noted, ART is operating under interim Lucent financing while it
seeks
out new capital.
ART has selected Lucent as a complete systems integrator for its wireless
access network. Lucent is responsible for everything from network design
and
buildout, to the maintenance and upgrading of operating systems. In August
1998, the companies expanded an existing credit facility. Lucent now
provides
ART with a $10 million equipment facility for the purchase of network
equipment
and a $25 million facility for working capital needs. Upon completion by
ART
of another round of financing of at least $50 million, ART will pay back the
original $25 million and Lucent will extend a credit facility equal to
2-times
the amount of capital raised, up to $200 million, for equipment. At that
point, the $10 million facility would be rolled into the new facility.
Lucent received warrants for 3.3% of ART's equity with an exercise price of
$0.01 on the first 1.2% and market prices on the remainder, which are issued
in
proportion to the amount drawn on the $25 million facility. The Lucent
facility solves ART's short term financing needs and facilitates the
buildout
of the first three "demo" markets, but ART will still need financing in 1999
to
continue its roll out. We believe the "bridge" will enable ART to prove out
its technology and business plan and attract additional capital investment
In April 1998, WinStar purchased 3.3M shares of Advanced Radio Telecom from
private investors for shares of WinStar stock. The deal valued ART shares
at
$17.39 per share. WinStar issued 1.525 million shares in the deal, which
resulted in a 14.9% ownership of ART, at the time. After giving effect for
recent spectrum acquisitions for stock by ART, WinStar's ownership currently
stands at 12.9%.
Management
ART has shaken up the management group over the last 12 months. Harry
Hirsch
became chairman, president, and chief executive officer in November 1997 and
developed the data-centric strategy for ART. Prior to joining ART, Mr.
Hirsch
spent almost a decade at the Williams Companies, most recently as vice
chairman
and CEO of Williams Communications Group and as president and CEO of WilTel
before that.
On August 18, 1998, William Maxwell, who joined ART in December 1997 as
Executive Vice President for Strategic Planning, Marketing, and Sales,
became
president and chief operating officer. Before joining ART, Mr. Maxwell was
executive vice president of ICG Communications where he headed the CLEC
business at ICG Telecom Group.
ART is currently searching for a CFO to replace Tom Grina, who left the
company
in August 1998 to pursue other interests.
|
