Robbi Report on Broadband Communications and Telco Spending....3/4/98
If you look at the fundamentals, this is a good read. Sorry I don't have time to clean it up more.
The information contained herein is not a complete analysis of every material fact respecting any company, industry or security. Although opinions and estimates expressed herein reflect the
current judgment of the Firm, the information upon which such opinions and estimates are based is not necessarily updated on a regular basis; when they are, the date of the change in
estimate will be noted. In addition, opinions and estimates are subject to change without notice. This Report contains forward-looking statements, which involve risks and uncertainties. The
Company's actual results may differ significantly from the results described in the forward-looking statements. Factors that might cause such a difference include, but are not limited to, those
discussed in "Investment Risks." BancAmerica Robertson Stephens from time to time performs corporate finance services for some companies described herein and may occasionally
possess material, nonpublic information regarding such companies. This information is not used in the preparation of the opinions and estimates herein. Facts and other information
discussed have been obtained from sources considered reliable but are not guaranteed. BancAmerica Robertson Stephens, its managing directors, its affiliates, and/or its employees may
have an interest in the securities of the issue(s) described and may make purchases or sales while this report is in circulation. BA Robertson Stephens International Limited is regulated by
the Securities and Futures Authority in the United Kingdom. This publication is not meant for private customers.
The securities discussed herein are not FDIC insured, are not deposits or other obligations or guarantees of Bank of America NTSA, and are subject to investment risk, including possible
loss of any principal amount invested.
BancAmerica Robertson Stephens March 4, 1998
Broadband Communications Tim Savageaux 415-693-3543
Stephen Perkins 415-676-2905
BROADBAND COMMUNICATIONS
Telco Capital Spending Outlook Strong for 1998
Plenty of Opportunity for Well Positioned Vendors
Now that it appears that the quarterly panic regarding the outlook for capital spending by communications carriers
has subsided a bit, we believe it is appropriate to expand upon the spending outlook for 1998. Fueled initially by
ADC Telecom' s (ADCT $21) self-immolation and further fanned by vigorous arm waving by various of our
competitors, concerns over capital spending levels by public carriers reached a peak in front of a number of
vendors fourth quarter earnings reports in late January. As we anticipated, these reports by the likes of DSC
Communications (DIGI $19), Tellabs (TLAB $57-1/2), Nortel (NT $53) and Advanced Fibre (AFCI $29) yielded
both strong fourth quarter results and positive outlooks for 1998. In the wake of this concern, we are publishing this
brief to outline our views for telecom carrier capital spending in 1998. This serves as an update and expansion to
the information published in our January 14 piece "Fiber Optic Communications: SONET is Dead! Long Live
SONET!".
ú In total, we expect telco wireline capital spending of at least $42.5 billion, up 8% from 1997' s total of
$39.3 billion. These figures include spending by local, long distance and competitive local exchange carriers
in the US, and for 1998 represent an increase from our previous estimate of $38.6 billion. Key drivers include
new or more aggressive spending plans by both incumbent IXC' s [AT&T (T $62-1/16), Worldcom (WCOM $38-
15/16)] and new long haul carriers such Williams Corp, more aggressive ROBC spending as well as a more
refined estimate of CLEC spending from smaller carriers. We note that the $39.3 billion eclipses our previous
estimate of $37.2 billion in 1997 capital spending by a fair amount, principally as a result of higher than
expected fourth quarter RBOC spending as well as the CLEC adjustment. RBOC overspending has been a
recurring phenomenon, and building in the same expectation to 1998 spending yields an even more
robust growth expectation of 12% to $44 billion. Either way, our analysis suggests a great deal of
opportunity for well-positioned communications infrastructure vendors in 1998.
Table 1: US Carrier Capital Spending ($billions)
1994 1995 1996 1997 1998E
ILEC 18.7 19.6 21.4 23.6 24.3
IXC 7.3 7.1 8.9 10.4 11.6
CLEC 0.7 1.6 3.5 5.3 6.6
Total 26.6 28.3 33.8 39.3 42.5
ú We expect a continuing shift in spending towards access networks and towards data networking. New
high-speed access technologies such as next generation digital loop carriers, xDSL and other broadband
access platforms will receive increasing attention. We note that the "new Internet" will be constructed by
public network carriers to much higher quality, reliability and network management standards than the
original equipment vendors to the old Internet, namely Cisco (CSCO $64-1/2), Ascend (ASND $33-7/8)
and 3Com (COMS $35-1/4), have been used to providing. We believe this favors incumbent suppliers of
carrier class network equipment, which have the same advantages of installed base and account control that
Cisco has in the Enterprise, especially given the acquisition of substantially all of the ISPs by telecom carriers.
The "carrier class" equipment and technology acquisition activities currently pursued by the data networking
players are indicative of their need for additional expertise to meaningfully penetrate the carrier market.
ú Generalizations regarding the overall telco spending environment usually are not very helpful in
determining the results of any specific equipment supplier. This is certainly true for the mid to small
capitalization players we focus on, and even more so in the access arena, which is our primary area of
thematic focus. Lucent (LU $104-13/16) and Nortel may well be subject to some of these macro concerns,
while we remain alert for specific customer issues or execution concerns, we find the overall telco spending
environment to be quite positive. Specifically, MCI (MCIC $48) has been planning to moderate spending in
1998, which may be exacerbated by its merger with Worldcom. However, as with the Nynex-Bell Atlantic and
SBC-Pacific Telesis mergers, we expect the impact to be limited.
ú Our stock picks in the space for 1998 are centered on three of the five major system platforms
currently being deployed in public carrier networks. AFC and DSC are the market leaders in digital loop
carrier systems currently being deployed in telecom access networks. These players are also well positioned
for the migration of the local loop to broadband technologies, as is HDSL market leader, Pairgain
Technologies (PAIR $20-1/8). Tellabs and DSC are major players in the digital cross connect market, which
serves key bandwidth management functions in the core transport network. Positron Fiber Systems (PFSCF
$7-3/8) is a leader in sales of SONET add drop muliplexers to CLECs for access and transport applications,
and represents pure play exposure to this market unlike its larger competitors. Nortel has posted several
major contract wins to construct long haul fiber optic networks for the likes of Qwest (QWST $36), MCI,
Williams and GTE (GTE $54-3/4) based on its SONET OC-192 technology and DWDM systems. We have
excluded central office circuit switches where Lucent and Nortel are dominant, as we believe this represents
an area of decelerating growth in the networks. The final system platform, WAN cell/packet switches, is
beyond the scope of our coverage at present, although we note the increasing momentum towards carrier
deployment of ATM switches and recent significant orders at vendors such as Ascend Communications,
Newbridge Networks and Nortel.
BARS Buy Rated Telecom Equipment Stocks
Company Ticker Price Mkt
Cap
1998 Rev Growth 1998 EPS Growth
Advanced Fibre AFCI $29 $2.3 B $425 MM 59% $0.75 58%
DSC Communications DIGI $19 $2.2 $1,875 18% $1.40 65%
Pairgain Technologies PAIR $20 1/8 $1.5 $350 24% $0.80 27%
Positron Fiber PFSCF $7 3/8 $0.1 $33 104% $0.07 NM
Nortel NT $53 $27.6 $17.7 14% $1.91 24%
Tellabs TLAB $57 « $10.7 $1,560 30% $1.75 30%
Note: Other telecom equipment stocks under coverage at BARS include ADC Telecom ($21 rated MP) and Ciena (CIEN $54 rated LTA)
Methodology:
Below we have included both the reported capital expenditure levels of most major domestic telcos as well as their
forecast/plan for 1998. Because we are focusing upon the domestic network we have removed expenditures for
wireless operations and licenses, international projects, cable TV franchises, and telephone books. We have
consolidated these numbers in Table 1 above and will expand upon them later.
ú Competitive Local Exchange Carriers will provide the highest rate of spending growth in 1998. These telcos
are finding their budgets rather flush after raising several billions of dollars during 1997 and the demand to
rapidly expand footprints as well as press into existing markets. We expect CLEC spending growth to
approach 25% and total $6.6 billion in 1998, with an upward bias due to a large number of smaller, start-up
players, which are not captured, by our survey. Spending by most players small to medium sized players is
expected to increase much faster than the overall rate, which will be impacted by cutbacks in spending at MCI
Metro.
ú Incumbent Local Exchange Carriers are initially forecasting modest increases in spending off of a very
large base of over $23 billion in 1997. While spending growth is currently forecasted to moderate to the low
single digits from the 9-10% level seen in the past two years, we believe present projections are conservative
given the propensity several of the RBOCs have shown for exceeding their budgets rather sharply. This was
again the case in the fourth quarter of 1997, where actual spending exceeded budget by $1.3 billion, or 6% of
total year budget expectations. Applied to 1998, this yields a more likely spending estimate of $25.8 billion for
an increase of 9%.
ú InterExchange Carriers are expected to post double digit spending growth to nearly $12 billion in 1998 as
voice and data traffic continue to pressure network backbones. Growth in spending by emerging carriers like
Qwest and IXC as well as continued growth at AT&T and Worldcom are expected to offset planned reductions
at MCI. Recent aggressive spending plans by upstart carriers such as Williams and Level 3 Communications
further bolster the outlook for long haul backbone spending.
ú Internet Service Provider spending is now captured primarily as a part of Incumbent and Competitive Local
Exchange Carrier spending, with independent ISP spending of insufficient magnitude to warrant separate
analysis. Major components of this spending include UUNET, which was acquired by MFS and is now part of
Worldcom, and BBN, which was acquired by GTE and is now known as GTE Internetworking. Worldcom has
also purchased ANS, which serves as AOLs network infrastructure, as well as Compuserve' s network facilities.
COMPETITIVE LOCAL EXCHANGE CARRIERS
ú Fastest Growing Capital Budgets
The smallest, but fastest growing, segment of the domestic carrier market is the competitive local exchange area.
New entrants are competing against incumbents and each other to deliver high volume dial tone and data services
initially to business customers. Increasingly telcos are focused upon building the facilities to create their own
networks rather than depend upon incumbent local exchange carriers for resold services. We have aggregated
nearly a dozen of the public independent CLECs as well as monitored the local plans of long distance providers
MCI and AT&T.
This measure likely under-represents the totals spent upon entering the local exchange market as it fails to include
the efforts of the telecom services group of several major cable operators and utilities. For examples MediaOne
already has installed a half dozen switches across several of its markets, LightPath (Cablevision' s CLEC) serves
more than 21,000 access lines, and Cox Communication with 250,000 voice grade equivalent circuits in service.
We conservatively estimate that, in aggregate, the data we' ve collected may exclude 20% of total spending by the
other 100+ CLECs registered in the U.S.
ú Many new entrants; well connected to financial markets
Within the last several years these companies have already made a sizeable impact on the market with more than
one million access lines served. This has not passed unnoticed in the financial markets where the opportunity for
these operators has manifest itself through the roughly $16 billion of capital raised in the public markets. While
several of these companies, MFS, Teleport (TCGI $55-1/8), and Intermedia (ICIX $75-1/2) in particular, have
made a name for themselves outside telecom circles, most players must suffer anonymity. Many are actually parts
of larger entities more occupied with other businesses (ie. MSOs Cox and Cablevision). Regardless of their
source of capital or size CLECs are increasing spending faster than any other group of operators.
Table 2: Major CLEC Capital Spending ($millions)
1994 1995 1996 1997 1998E
MFS (WCOM) 363 507 888 950 1,000
Teleport (T) 143 282 375 500 1,000
Brooks (WCOM) 42 42 217 400 450
ICG 55 88 175 270 400
Intermedia 14 32 144 250 275
GST 0 34 148 250 220
McLeod 0 15 70 150 190
NextLink 0 18 52 140 180
ACSI 0 0 104 142 135
SubTotal 623 1,017 2,173 3,052 4,200
Gross Up 650 1,100 2,414 3,590 5,250
AT&T 0 250 650 900 600
MCI 32 265 390 800 700
Total 682 1,615 3,454 5,290 6,550
INCUMBENT LOCAL EXCHANGE CARRIERS
ú Largest capital budgets
The largest budgets for telecom equipment are found among the incumbent local exchange carriers. The mega-RBOCs
of Bell Atlantic-NYNEX and SBC-PacTel are expected to spend in excess of $5 billion apiece in 1998, with
BEL spending nearly as much as the entire CLEC industry. Spending by the RBOCs and the major independents
upon landline equipment has increased every year since the start of the decade. After looking at the modest
increases most telcos are planning for 1998 we don' t see any reason to panic, rather note that this is the historic
rate at which the largest LECs set their budgets at the beginning of the year. Of the major independents we have
included only Sprint and GTE, while noting the abundance of small to medium sized independent telcos in the US.
Table 3: Incumbent LEC Capital Spending ($ billions)
1994 1995 1996 1997 1998E
Ameritech $1.6 $1.6 $1.9 $2.1 $2.3
BellSouth 3.0 3.2 3.2 3.4 3.7
SBC Comm 3.3 3.7 4.6 5.0 5.0
Bell Atlantic 4.6 4.9 4.9 5.5 6.0
GTE 2.8 2.6 2.7 3.6 3.4
US West 2.4 2.7 2.8 2.6 2.5
Sprint 0.9 1.0 1.2 1.4 1.5
Total 18.7 19.6 21.4 23.6 24.3
Ameritech (AIT $ 43-7/8) continues to increase its expenditures, as competition within its region becomes more
intense. With heavy GTE presence throughout and several CLECs targeted at the Chicago marketplace we expect
Ameritech to remain one of the most progressive in its planning process. This is likely to be driven further by cable
modem rollouts in Illinois and Ohio. SBC Communications (SBC $79-9/16) is still in the process of developing their
1998 plans while it digests its Pac Bell acquisition. We expect this is likely to continue to be lumpy should the
acquisition of SNET, announced the start of this year, go through. Overall, however, we believe this consolidation
is likely to drive increased spending over time as SBC brings the networks of Pac Bell and SNET in line with their
own architecture. Bell Atlantic (BEL $93-1/8) has resolved more of their capital plans for 1998 after its acquisition
of Nynex. A meaningful portion of their increased budget has already been awarded to DSC Communications for
its Litespan DLC and both narrowband and broadband digital cross connects. Tellabs has benefited from these
increases as well with a recent wideband digital cross connect award. GTE is maintaining its budget for 1998 at
$3.4 billion despite a $400 million overrun to the 1997 plan of $3.2 billion. While we are rather uncertain about this
low number, we are more comfortable allowing it to serve as upside than speculating as to a more realistic
number. We note that GTE' s number also does not include an estimated $600-700 million in spending at GTE
Internetworking, up from around $500 million in 1997. We look for a continued steady increase from Bellsouth
(BLS $60-11/16), and view US WEST' s (USW $51-7/8) plan as conservative given their aggressive plans for high
speed data and fiber to the curb access rollouts in region.
ú Continue to increase plans; often exceed annual targets
It is important to remember that these budgets can, and historically do, change over the course of the year. Most
often these increases are driven as a response to either customer demand or increased competition. Most of the
capex increases in 1998 driven by customer demands for increased service whether high speed data links for
business or second lines for residential customers. These expenditures, which deliver immediate revenue, are
easily justified by the telcos. We expect these customer driven expenditures to continue, competition to create
additional pressure for upgrades, and, specific to 1998, local number portability deadlines to drive continued
budget overruns.
Table 4: LEC Spending vs Budget
Original
Est
Actual
1997 OverRun
Ameritech 2,100 2,100 NM
BellSouth 3,400 3,429 NM
SBC Comm 4,700 4,961 6%
Bell Atlantic 5,000 5,520 10%
GTE 3,200 3,607 13%
US West 2,500 2,643 6%
ú Change in focus of spending toward access
While the aggregate total of over $42 billion in telco spending is impressive, particularly for a company with sales
measured in the millions, we believe it is important to recognize that much of this spending is upon purchases not
really relevant to the company' s we follow, for example purchases of cable and wire accounted for 31% of 1996
expenditures. While we believe investment in transmission equipment, which along with switching represents our
definition of "telecom equipment", will continue to increase not only in absolute dollars, but as a percentage of total
levels as well. Transmission has scaled from 24% of LEC capital spending in 1993 to 28% in 1996, or
approximately $6.3 billion and, we believe, is likely to reach approximately one third of local exchange carrier
capital investments by the year 2000.
Table 5: LEC Capital Spending Priorities
1993
General
18%
CO Switching
25%
CO Trans
24%
Cable & Wire
33%
1996
General
15%
CO Switching
25%
CO Trans
29%
Cable & Wire
31%
Source: FCC
INTEREXCHANGE CARRIERS
ú Continuing Pressure on Incumbent Network Capacity
Spending levels at the IXCs continues to climb as each responds to the rapid growth of data traffic, with our
forecast calling for an increase in spending of 11% to $11.4 billion. Often cited to grow from its approximately 50%
share of network capacity to dominating levels - John Sidgmore of Worldcom last week speculated voice could be
just 1% of network traffic in 2003 - bandwidth supply is becoming constrained. To meet demand backbone carriers
have more than doubled capital expenditures since 1993. Most carriers are also expecting to increase spending
during 1998. The one exception to this is MCI, the number two long distance carrier which attributes its spending
declines to its decision to increase focus spending upon the local market and recover from a robust spending spree
as the network was dragged from one of the least to one of the most reliable national backbone. In particular the
company may slash the program of heavy spending upon disaster recovery and rerouting capabilities that it had
embarked on several years ago when MCI' s network was the least reliable of the IXCs.
Table 6: IXC Capital Spending ($ billions)
1994 1995 1996 1997 1998E
AT&T $3.4 $2.8 $3.9 $4.2 $4.8
MCI 2.9 2.9 2.9 3.0 2.3
Sprint 0.8 0.9 1.0 1.2 1.4
WorldCom 0.2 0.4 0.7 0.7 0.8
LCI 0.1 0.1 0.2 0.3 0.6
Qwest 0 0 0.1 0.5 0.6
IXC 0 0 0.1 0.4 0.5
Williams 0 0 0 0.3 0.8
Total 7.3 7.1 8.9 10.4 11.6
ú Several new carriers spend aggressively
We expect that over the next several years we will see the number of cross country networks grow beyond the four
that presently exist. Long haul carriers which substantially did not exist four years ago are expected to
spend over $2 billion on network infrastructure in 1998. With new carriers like Qwest and Williams as well as
startups like Level3 constructing networks that will overlay many of the larger telcos networks we expect the
incumbents to push for advanced technologies to leverage their existing infrastructure. This has already begun to
happen as TDM line rates are pushed to OC-192 and WDM migrates beyond eight channels. We believe that the
companies who supply this technology are likely beneficiaries of their increased use. These new networks are also
more focused on the transport of data and multimedia information than voice, suggesting a potential greater role in
the public network for suppliers with data communications expertise. We note that several deals have been struck
in recent months by new carriers with a wide variety of customers, including carriers, ISPs and corporations, for
the wholesaling of network bandwidth.
INVESTMENT IMPACT: Given the central role to be played by both traditional and emerging telecommunications
carriers in transporting voice, video and data across public networks, and the resultant demand for carrier
communications technology and infrastructure, we believe it is important to understand where the exposure to
there players lies. The table below outlines the vendors with maximum exposure to carrier infrastructure builds
both in North America. Lucent and Nortel offer the greatest absolute exposure to the North American carrier
spending trends we have discussed, while DSC, Tellabs, Newbridge and ADC offer still sizeable absolute exposure
which is more pure play in nature. Complete pure plays of some substance include AFC, CIENA, ADTRAN and
Pairgain Technologies.
Table 7: Carrier Equipment Revenues by Vendor
Vendor Ticker 97 Carrier
Revs
Pct Total
Revs
Product Focus
Lucent Tech LU $11.6 Bn 44% Circuit Switch; Fiber Optic Transport; Cross Connects;
Access
Nortel NT $7.4 48% ATM/Frame and Circuit Switches; Fiber Optic Transport
Cisco Systems CSCO $2.0 27% ATM Switch (Stratacom), Routers, Dial Access
DSC Comm DIGI $1.3 80% Access, Circuit Switch,Cross Connects, Fiber Optic Transport
Newbridge NN $1.0 85% ATM Switch/TDM Mux
Tellabs TLAB $1.0 80% Digital Cross Connect; Digital Transport
ADC Telecom ADCT $0.9 75% Access
Ascend Comm ASND $0.7 60% ATM/Frame Switch (Cascade); Dial Access
CIENA CIEN $0.4 100% Fiber Optic Transport
Advanced Fibre AFCI $0.3 100% Access
Pairgain Tech. PAIR $0.3 100% Access
ADTRAN ADTN $0.3 100% Access
Note: Excludes Wireless and Enterprise (PBX, LAN) sales; Carriers includes LECs. IXCs and CLECs
Of the data networking vendors, Cisco appears to be the best positioned to become a meaningful supplier
to the carriers. With the acquisition of Stratacom and several smaller technology acquisitions in the xDSL,
SONET, Signaling System 7 and ISDN areas, plus the development of the Cisco 12000 carrier class router, the
company' s strategy appears solid. Of an estimated $2 billion of carrier sales in 1997, over half are estimated to
come from router sales with the Stratacom business representing the core of the carrier-class business. Cisco has
announced major deals with carriers such as US WEST and GTE for data networking infrastructure. Ascend also
has significant carrier revenues, both from the acquisition of Cascade Communications as well as from sales of
remote access equipment to ISPs which have been purchased by carriers such as UUNET, which is now part of
Worldcom. Assuming that half of the company' s dial up access sales belong in this category, we estimate that as
much as 60% of Ascend' s sales are derived from carriers.
Given the acquisition by public carriers of substantially of all of the major internet service providers, we
believe the integration of carrier voice, data and potentially video networks will be driven by the carriers,
not the ISPs, and this may favor the carriers historical suppliers. We view the "carrier class" reliability
requirements of carrier networks as much harder for data vendors to achieve than the converse. On the other side
of the coin, traditional carrier suppliers have acquired data networking technologies to add to their product
platforms. Importantly, these players already have (a) a large installed base of equipment (b) account control of
their carrier customers and (c) knowledge of public network reliability and management requirements. If current
switching, transport and access platforms can be made to seamlessly migrate to a more data-centric public
network, meaningful market entry by new suppliers may be challenging. Interestingly, many of these same
factors, most notably a technology-agnostic, customer-centric strategy, are cited in explaining Cisco' s hammerlock
on the enterprise networking market. |
