Telecommunications Equipment
Impact of Carrier Spending on Equipment Vendors -- Review & Outlook
September 9, SUMMARY
2002 * Despite a series of capex cuts in 1H02, we believe that
the reductions are not complete. Furthermore, we believe
Daryl that the weak trends reported by carriers in 1H02, will lead
Armstrong to '03 guidance of flat to down 10% on a Y-O-Y basis.
* Consequently, we believe that it is still too early to
consider increasing investment exposure to the traditional
Alex Henderson telecom equipment vendors.
* We believe the strong seasonal uptick in wireless spending
in 2Q will not sustain into 3Q, making it likely that we
Matthew will see a meaningful decline in revenues from Lucent and
Gavigan Nortel's wireless businesses on a Q-Q basis.
* Longer term. we believe that the order flow from VZ and
Sprint PCS's 1XRTT buildout might be more resilient than we
previously expected. The bad news is that this segment
represents over one-half of Lucent's business.
* Our Underweight sector rating is based on our view that a
rebound in carrier spending is unlikely, enterprise remains
sluggish, and limited investor interest.
OPINION
Our Continued Expectation of Additional Capital Spending Cuts With Capex
Probably Flat-to-Down in 2003 Keeps Us On The Sideline Relative To The
Traditional Telecom Equipment Names.
Similar to 1Q, 2Q held its share of capital spending guidance reductions.
Including all reductions, spending for the largest service providers now is
expected to decline by 53% on a year-over-year basis. Some have argued that
with such a steep decline is baked into investors' sentiment, and that now is
the time to re-visit the traditional equipment names. The logic is that
their core customers, the incumbents, will survive and that spending has
declined so much that trends, by definition, have to improve. We believe
this stance is mistaken for a number of reasons. First, we do not believe
that the spending reductions are done. While current spending expectations
suggest a capital intensity in the long-term historical range of 15%,
carriers continue to bring down revenue and earnings guidance. This reduces
their need to invest. Secondly, we believe that the tough conditions
experienced by the carriers in the 1H 2002 period, will likely result in 2003
spending guidance that is flattish at best. Realistically, if current demand
trends remain in place, we think that spending forecasts could be down as
much as 10%. We see this as being a negative near-term catalyst for the
group. We note that we are talking about forward-guidance not actual
spending. We believe that if the economy were to pick up sharply in 2003,
carriers would open their pursestrings in order to pursue revenue
opportunities. During our technology conference, Tellabs noted that it took
the carriers two quarters to cut off the spending spigot during this downturn
and that they probably could turn it back on in one quarter. We tend to
agree with this viewpoint.
A third bearish near-term concern we harbor revolves around wireless
spending. We believe the large national wireless carriers probably will not
show the strong spending performance experienced in 2Q. In fact, we believe
that the wireless equipment businesses at the traditional telecom vendors
like Lucent are probably under significant pressure in the September quarter.
For all these reasons, we think that its unlikely that the traditional
telecom equipment vendors outperform in the near term and would advise
investors to remain on the sideline.
Vendors Are Already Talking About Spending Declines In 2003. The most
interesting datapoint that we garnered from our recent Technology conference
is that vendors seem to have already resigned themselves to a decline in
capital spending guidance for 2003. Tellabs noted that they believed
spending could be flat to down 10% or more. Nortel noted their expectation
of declines probably in the single digit range. This sentiment was echoed by
the management of Advanced Fiber. Only Lucent seems to be managing their
business plan with the assumption of flat year-over-year growth. While we
agree with the viewpoint of declining capex, we still do not believe guidance
of spending declines next year are fully incorporated into market sentiment.
Consequently, we believe that as declining guidance figures for 2003 are
released by the service providers, the equipment names will come under
pressure.
WIRELINE -- 2Q REVIEW
Capital Spending In The Seasonally Strong Calendar 2Q Comes In Weaker Than
Expected. Following a sharp drop-off in 1Q spending and heading into a
seasonally strong 2Q, many of the bulls had argued that we had reached a
bottom and the traditional telecom equipment names were poised for a rebound.
However, the second quarter provided neither a cyclical nor much of a
seasonal rebound in capital spending. With the exception of AT&T and
Verizon, which reported sequential spending increases of 59% and 32%
sequentially, spending at most major carriers was roughly flat relative to
1Q. Sprint reported capital spending that was up 1% while BellSouth and
SBC's spending was down 1%. As noted previously, we think that part of the
spending from the carriers, particularly the access network oriented RBOCs,
was absorbed by a seasonal uptick in maintenance spending. Consequently,
equipment purchases almost certainly continued their decline on a sequential
basis. This seems to be borne out in the revenue numbers of the large
domestic telecom equipment vendors. We believe this trend is continuing into
the third calendar quarter, as the preannouncements from Nortel and Tellabs
seem to suggest.
More importantly, BellSouth, Qwest, SBC, Sprint and Verizon all announced
their intention to further reduce capital spending for 2002. As a result, we
are now forecasting CY02 capital spending to decline 53% Y-Y, which compares
with our prior forecast of a 46% decline.
FIGURE 1: CAPITAL SPENDING REDUCTIONS
2002 Target 2002 Target
(old) (new)
AT&T 3,800 -- 4,200 3,800 -- 4,200
BellSouth 4,200 -- 4,400 3,700 -- 3,900
Qwest 3,100 -- 3,300 3,000 -- 3,100
SBC < 9,200 < 8,000
Sprint 2,600 2,500
Verizon 14,000 -- 15,000 13,000 -- 13,500
TOTAL 36,900 -- 38,700 34,000 -- 35,200
Source: Company Guidance
From a capital intensity standpoint, with the new guidance cuts, we project
capital intensity to come in at roughly 15%. This is down from 28% in 2001
and 33% in 2000. As a point of reference, before the telecom bubble of the
late 1990s, capital intensity stood in the mid-teens.
WIRELESS -- 2Q REVIEW
Strong Spending Trends In Calendar 2Q, Sets Up Tough Sequential Comparisons
For Wireless Vendors in September Quarter. Unlike in the wireline industry,
we saw a strong surge in spending by the wireless carriers. For the national
publicly traded independent carriers, capital spending surged 43%
sequentially. This compares to revenue and EBITDA Q-Q growth of 8% and 16%,
respectively.
FIGURE 2: WIRELESS CAPITAL INVESTMENT
2000 2001 1Q01 2Q02
AT&T Wireless $3,751 $5,045 662 866
Nextel Communications 2,976 2,382 474 448
Sprint PCS 3,047 3,751 603 825
Cingular 2,250 3,422 455 906
Verizon 4,322 5,006 812 1,248
Totals $16,346 $19,606 $3,006 $4,293
Source: Company reports
Given where we are from a seasonal perspective and given that most of the
large carriers are engaged in some form of upgrade, its not surprising that
spending performance was strong in 2Q 2002. We are not factoring in similar
trends in the calendar 3Q period as demand seasonality works against the
vendors in 3Q. Thus, it would not surprise us if the wireless equipment
vendors like Lucent see wireless equipment revenues ramp down in the
September quarter. Nortel has already preannounced lower-than-expected
revenues; a shortfall that we believe incorporates some noticeable weakness
in wireless.
A Closer View At Lucent's 1XRTT Buildouts At Sprint and Verizon---It Will
Take Longer For Revenues From The Projects To Erode For Lucent, But The
Coming Mix Shift Will Hurt The Vendor's Margins. Two of the projects that we
have monitored over the past year have been the 1XRTT buildouts at both
Verizon Wireless and Sprint PCS. They are key initiatives given that both
are significant customers to Lucent, which now garners the majority of their
revenue stream from wireless-related equipment and services sales. As both
projects are slated to have their initial phases completed in calendar 2002,
we have been evaluating the likely impact will be on their equipment vendors
after this year.
For example, we believe that Sprint probably is investing $800 million -$1
billion on the project. As a point of comparison, this probably represents
less than one-third of the investment necessary for their 2G network.
Because Sprint is upgrading from a IS-95 2G network to 1XRTT, the upgrade
process is fairly straightforward. For the newer 2G base stations (deployed
within the 3-5 years), the carrier primarily has to perform a digital card
and software upgrade. We estimate that Sprint probably has roughly 10,000
base stations in their network. Of the total investment, we estimate that
roughly 85%-90% of the project budget is spent on equipment and software and
the residual allocated to cover labor expenses. In some instances, there is
also some incremental switching investment to support the additional traffic
that is likely to come onto the network. Given this process, the next
question revolves around what occurs when the initial buildout is complete.
Interestingly, when carriers perform these types of upgrades, they usually
only populate one-third of the base stations with 3G cards. Thus, even when
the initial phase is complete, there are still success-based orders for line
cards. We estimate that success-based spending can represent up to 75% of
the initial contracted amount in the first year after the upgrade occurs if
the carrier net addition statistics for 3G subs are strong. In the second
and third years, we believe it can fall to 50% and then 25%, respectively of
the original contracted amount.
We believe that that a similar upgrade process is occurring at Verizon
Wireless, except the buildout is probably larger. We estimate that the
carrier probably has almost twice as many base stations at Sprint.
Replacing the natural erosion of business from these carriers over the next
several years will be difficult for Lucent. Two potential incremental new
projects that Lucent could attempt to sell to these CDMA carriers include
intelligent antennae technology and 1xEV-DO. Intelligent antennae technology
allows the wireless carrier to expand the coverage of its wireless network.
Lucent's potential solution would allow an increase of the voice network by
70% and its data network by 100%. 1xEV-DO is an upgrade that allows the
carrier to significantly increase the throughput of their data offerings,
allowing burst rates of 600-1200 kbps. We see two issues with 1xEV-DO.
First, operators around the world are becoming increasingly more skeptical
about whether wireless data will be successful with customers. Consequently,
we believe they will need to see momentum from their current 2G offerings
before considering additional upgrades.
A second issue is that 1xEV-DO requires additional carrier bandwidth. For
cities where carriers have limited spectrum, they may not want to dedicate a
portion to data-related services when voice services still generate the bulk
of their profits.
The bottom line is that while we believe it will take longer for Lucent's
wireless revenues from Verizon and Sprint to tail off than we initially
expected, we still believe that it will be tougher to sell new projects.
Most important, however, is the expected mix shift of orders from these
carriers after the projects are completed. In the initial build, very-high
margin software dominates the mix of business coming to vendors. After the
initial mix is completed, there is a shift toward line cards which have good
margins but nearly as high as software.
Brief Review Of Carrier-Specific Commentary.....
AT&T
For the full-year 2002, management reiterated their spending guidance of $3.8
to $4.2 billion for the Core Communications Group that excludes the Broadband
segment. This compares to our SSB forecast of $3.8B. The company showed a
59% sequential improvement in capex for their core communication services,
from $613 million in 1Q to $976 million in 2Q. Capital intensity stood at
10%, up from the 6% level seen in 1Q02. This low corporate intensity is
partially due to the low capital requirements of the consumer long distance
segment. Within the Broadband segment, capital spending came in at $950, up
from the $739 million in the prior quarter. Of this total, roughly $408
million was directly related to network improvement and $320 million was
related to the growth and support of advanced services.
BELLSOUTH
BellSouth reduced their 2002 capital spending budget to come in at $3.7 to
$3.9 billion. This represented a $500 million, or 12% reduction from its
previous guidance. The new guidance implies a 37% year-over-year decline in
spending. During the quarter, the carrier invested roughly $1 billion in
capital spending. This is consistent with their level of investment in 1Q
2002. Their capital intensity came in at 17.8%, in line with 1Q levels.
During the quarter, they added over 500 SONET rings, bringing their total to
over 22,500. They also now have 830 broadband switches, up 10 sequentially
and around 180 DWDM systems, up 30 from 1Q levels. BellSouth announced an
access line count that was down roughly 1% sequentially. Within this total,
secondary lines for residential customers were down 4% sequentially. As some
of this decline was attributed to technological substitution as people move
up to DSL, this could actually be seen as somewhat constructive for access
and routing suppliers into their DSL network like Cisco.
Cingular's 2Q capex came in at $906 million, a sharp increase from the
roughly $450 spent in 1Q02. In terms of the GSM/GPRS upgrade, the carrier
announced that they had built out roughly 36% of their network. They also
reiterated their view on the buildout schedule for their 3G EDGE network.
They still expect the EDGE buildout to begin in the 4Q02 time and be
completed in 2004.
QWEST
Qwest reported 2Q capex came in at $618 million, down 48% sequentially but
within the range of our SSB forecast of $650 million. The carrier's spending
total represented a capital intensity of 14%. For the first six months of
2002, the carrier's capital intensity has come in at 21%. The carrier
reported that its DSL subscribers came in at 508K, up from 464K at the end of
1Q02. Access lines came in at 17.35 million, down from 17.6 million in
1Q2002. It also reported that they had no revenues associated with equipment
sales this quarter. This compares to $205 million last quarter.
Qwest reduced their 2002 capital spending budget down from a range of $3.1-
$3.3 billion to a range of $3-$3.1 billion. Based on the spending total in
1H2002, its guidance suggests that spending will be down around 30% at best.
When asked about the sustainable level of spending, management noted that it
believed that quarterly spending could remain at this level or below under
current conditions. Management commented that spending would be fairly
consistent on a quarterly basis going forward and was adamant that they would
not see a seasonal uptick in 4Qbudget flush. In terms of spending
allocation, the carrier expects roughly 75% of its spending total to be
focused on their local business leaving roughly $800 million for the long-
haul business. With at least one portion of this investment is for
maintenance capital spending, this does not leave much for its long-haul
equipment vendors.
SBC COMMUNICATIONS
SBC edged down their 2002 spending guidance to below $8 billion. This
compared to our SSB forecast of $8.25 billion. Even at the high-end of the
revised range, the new guidance suggests a year-over-year decline in the
neighborhood of 30%. SBC's capital spending came in at $1.73 billion,
essentially flat with the $1.77 billion invested in 1Q02. SBC's flat
spending probably reflects a sequential decline in equipment spending as a
portion of the 2Q investment flows into these external plant projects. SBC's
capital intensity in 2Q02 came in at 18%, consistent with the 18% intensity
. |
