Sie sind auf Seite 1von 316

North America Equity Research

13 January 2011

U.S. Telecom Services & Towers


Initiating Coverage with Favorable View as Data
Trends Drive Demand
We are initiating coverage on 14 US telecom and wireless tower stocks with a
favorable view on investment in the industry due to strong cash flow characteristics,
stable business trends and potential improvements as the economy rebounds, and the
potential for data demand (wired and wireless) to drive traffic and revenue higher in
the next few years. While some carriers will continue to see heavy competition, we see
little irrational competition in either wireless or wireline today.
Wireless fundamentals shift to data from voice. We expect wireless voice adds to
decline again in 2011, and forecast only 2.8 million postpaid adds among the Big 4
carriers, down from 3.7 million in 2010. Instead, data should be the main driver of
value creation from here, as consumers upgrade to smartphone plans with higher
monthly bills and add 3G-enabled tablets and MiFi devices to their everyday lives.
An improving economy could benefit all carriers, but particularly those that target
lower demographics like Leap Wireless and MetroPCS. The move to data is also
driving an upgrade cycle of hardware at the cell site, which should continue to
support strong growth for tower companies in 2011 and 2012.
Wireline trends are stabilizing, led by data and video. While only a few years ago
the wireline business looked terminal, today access line trends are slowly improving
and data/video businesses have become big enough to drive growth in some
companies consumer wireline revenue. Despite constant legacy revenue pressure,
carriers have managed to cut costs and maintain margins to support their large
dividends, which we believe are sustainable. As consolidation continues, we expect
merger synergies to allow companies like Frontier to expand margins.
2H10 stock performance leads to some caution near term. Many telecom stocks
outperformed the S&P 500 in the second half of 2010 as investors looked for
defensiveness in high-dividend-paying stocks and worries about a dividend tax hike
faded. As we enter 2011 we expect these stocks to lag the market somewhat (as has
happened in the last few years), as investors view them as a source of funds, but still
see significant upside for investors willing to hold through any volatility.
In telecom we favor stocks with highest cash flow yields (AT&T, Frontier) and
potential turnaround stories (Sprint, Leap Wireless). Among the largest telecom
players, we prefer AT&T over Verizon given its better cash flow profile and low
multiples, which we believe offset the risks created by the loss of iPhone exclusivity.
Not only should AT&Ts EPS grow faster in 2011, but the multiple should also
expand once iPhone risk has been quantified. Among the wireless names, we like the
risk/reward on Sprint as it continues to turn around and recommend Leap Wireless
and MetroPCS into the companies traditionally strong first quarters, though we
prefer Leap Wireless given its earlier-stage turnaround. We would avoid US Cellular
given its high multiple, weak cash flow, and controlled status.

Telecom Services & Towers


AC

Philip Cusick, CFA

(1-212) 622-1444
philip.cusick@jpmorgan.com

Richard Choe
(1-212) 622-6708
richard.choe@jpmorgan.com

Manish Jain
(1-212) 622-8692
manish.x.jain@jpmorgan.com

Derya Erdemli
(1-212) 622-8529
derya.erdemli@jpmorgan.com
J.P. Morgan Securities LLC

Coverage Universe
Ticker
T
VZ
S
CLWR
LEAP
PCS
NTLS
USM
TDS
FTR
WIN
AMT
CCI
SBAC

Price
(1/12/11)
$28.04
$35.47
$4.41
$5.70
$13.19
$12.97
$19.68
$50.30
$36.28
$9.43
$13.37
$50.70
$42.09
$40.19

JPM
Rating
OW
N
OW
N
OW
OW
N
UW
N
OW
N
OW
OW
OW

YE11
Target
$33
$38
$7
$6
$16
$16
$22
$45
$42
$11
$14
$60
$55
$53

Source: J.P. Morgan.


Note: J.P. Morgan ratings: OW Overweight;
N Neutral; UW Underweight.

Tower group remains attractive. Though there are small differences in strategy,
the wireless tower business is fundamentally homogenous and exposed to strong
secular tailwinds. We recommend all three companies, American Tower, Crown
Castle, and SBA Communications, with Overweight ratings, and among them prefer
SBA Communications given its higher growth potential and flexibility.

See page 307 for analyst certification and important disclosures.


J.P. Morgan does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may
have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their
investment decision.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Equity Ratings and Price Targets


Company
AT&T Inc.
Verizon Communications
Sprint Nextel
Clearwire
Leap Wireless International
MetroPCS
NTelos Holdings Corporation
US Cellular
Telephone & Data Systems
Frontier Communications
Windstream Corp
American Tower
Crown Castle International
SBA Communications

Symbol
T
VZ
S
CLWR
LEAP
PCS
NTLS
USM
TDS
FTR
WIN
AMT
CCI
SBAC

Mkt Cap
($ mn)
166,501.52
100,380.10
13,185.90
5,619.13
1,001.98
4,591.43
821.60
4,347.33
3,820.21
9,326.27
6,413.59
20,455.17
12,042.75
4,611.72

Price($)
28.04
35.47
4.41
5.70
13.19
12.97
19.68
50.30
36.28
9.43
13.37
50.70
42.09
40.19

Sourc e: Company data, Bloomberg, J.P. Morgan es timates. n/c = no change. All pric es as of 12 Jan 11.

Rating
Cur
OW
N
OW
N
OW
OW
N
UW
N
OW
N
OW
OW
OW

Prev

Price Target
Cur
Prev
33.00

38.00

7.00

6.00

16.00

16.00

22.00

45.00

42.00

11.00

14.00

60.00

55.00

53.00

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table of Contents
Investment Conclusion ............................................................5
Wireless Industry....................................................................13
Wireless Key Investment Points ................................................................................15
Wireless Stock Recommendations.............................................................................15
Wireless Investment Risks.........................................................................................17
Wireless Industry Overview ......................................................................................18
Handsets Smartphone Share Is Growing.................................................................27
Emerging Devices to Augment Growth.....................................................................43
Competition and New Entrants ..................................................................................45
Spectrum Auctions Will Come Eventually ................................................................47
2011 Is the Year of 4G Transitions............................................................................49
Cash Flow Solid Across Wireless Space ...................................................................52

Wireline Industry ....................................................................53


Wireline Key Investment Points ................................................................................55
Wireline Stock Recommendations.............................................................................56
Wireline Investment Risks.........................................................................................57
Wireline Industry Overview ......................................................................................57
Consumer Buoyed by Data, Video ............................................................................62
Business, Regulatory 60%+ of Wired Revs...............................................................67
Consolidation to Insulate Cash Flow .........................................................................71

Tower Industry........................................................................75
Tower Key Investment Points....................................................................................77
Tower Stock Recommendations ................................................................................78
Tower Investment Risks ............................................................................................79
Tower Industry Overview ..........................................................................................80
Wireless Tower Overview .........................................................................................98

Large Cap Integrated Telecom Operators ..........................109


AT&T Inc. ...............................................................................................................111
Verizon Communications ........................................................................................123
Sprint Nextel ............................................................................................................139
Clearwire..................................................................................................................151
Leap Wireless International .....................................................................................161
MetroPCS ................................................................................................................171
NTELOS Holdings Corp..........................................................................................179
US Cellular Corp .....................................................................................................185
Telephone & Data Systems......................................................................................191

Wireline Operators ...............................................................197


Frontier Communications Corp ...............................................................................199
Windstream Communications..................................................................................209

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless Tower Operators ...................................................217


American Tower ......................................................................................................219
Crown Castle International ......................................................................................225
SBA Communications .............................................................................................231

Company Models..................................................................237
AT&T Model ...........................................................................................................239
Verizon Model .........................................................................................................245
Sprint Nextel Model.................................................................................................251
Clearwire Model ......................................................................................................257
Leap Wireless Model ...............................................................................................261
MetroPCS Model .....................................................................................................265
NTELOS Model.......................................................................................................269
US Cellular Model ...................................................................................................275
TDS Model ..............................................................................................................279
Frontier Model .........................................................................................................283
Windstream Model ..................................................................................................287
American Tower Model...........................................................................................291
Crown Castle Model ................................................................................................295
SBA Communications Model ..................................................................................303

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Investment Conclusion
We believe that worries around slowing growth in the wireless space and destructive
competition in wired data and video are overblown. In our view, additional data
services (including video) and devices (like tablets, smartphones, and machinemachine) represent a tremendous opportunity for the telecom industry and will drive
continued revenue and subscriber growth. In regards to competition, wireless data
capacity is barely keeping up with demand, so we see no reason for meaningful price
reductions in 2011, and on the wired side the telecom and cable companies seem
very rational and recently have been raising, rather than reducing, prices.
Fundamentals look favorable in the telecom industry
In general we are favorable on investment in the telecom industry due to strong cash
flow characteristics, stabilized business trends and potential improvements as the
economy rebounds, and the potential for data demand (wired and wireless) to drive
traffic and revenue higher in the next few years. While some carriers (Leap Wireless
and MetroPCS in particular) likely will continue to see heavy competition, we see
little irrational competition in either wireless or wireline for the time being.
Wireless fundamentals shift to data from voice
We expect wireless voice adds to decline again in 2011, and forecast only 2.8 million
postpaid adds among the Big 4 carriers, down from 3.7 million in 2010. Instead, data
should be the big driver of value creation from here, as consumers trade up to
smartphone plans with higher monthly bills and add 3G-enabled tablets and MiFi
devices to their everyday usage list. An improving economy could also benefit
carriers that target lower demographics like Leap Wireless and MetroPCS. The move
to data is also driving an upgrade cycle of hardware at the cell site, which should
continue to support growth for tower companies in 2011 and 2012.
Wireline trends are stabilizing, led by data and video
While only a few years ago the wireline business looked terminal, today access line
trends are slowly improving and data/video businesses have started to actually drive
growth in some companies consumer wireline revenue. Despite constant revenue
pressure, carriers have managed to cut costs and maintain margins to support their
large dividends, which we believe are sustainable. As consolidation continues, we
expect merger synergies to allow companies like Frontier to expand margins.
Recent stock performance leads to some caution near term
Many telecom stocks outperformed the S&P 500 in the last six months, as investors
looked for defensiveness in high-dividend-paying stocks like AT&T, Verizon,
Frontier, and Windstream and worries about a dividend tax hike faded. As we get
into 2011 we expect some of these stocks to lag the market somewhat (as has
happened in the last few years), as investors view them as a source of funds into the
new year, but still see significant upside for investors willing to hold through any
volatility.
In addition, we believe that expectations of a Verizon iPhone announcement have
driven that stocks outperformance lately, to levels that we find unattractive.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 1: AT&T and Verizon Forward P/E Multiple vs. S&P 500 P/E Multiple
17.0

16.0

15.0

14.0

13.0

12.0

11.0

VZ PER

T PER

12/31/2010

12/17/2010

12/3/2010

11/19/2010

11/5/2010

10/22/2010

10/8/2010

9/24/2010

9/10/2010

8/27/2010

8/13/2010

7/30/2010

7/16/2010

7/2/2010

6/18/2010

6/4/2010

5/21/2010

5/7/2010

4/23/2010

4/9/2010

3/26/2010

3/12/2010

2/26/2010

2/12/2010

1/29/2010

1/15/2010

1/1/2010

10.0

SPX PER

Source: Bloomberg.

Telecommunications Stocks
Among telecom companies, we favor stocks with the highest cash flow yields (AT&T,
Frontier) and potential turnaround stories (Sprint), as well as seasonally strong
players like MetroPCS and Leap Wireless. We would avoid Verizon given its high
EPS multiple and iPhone excitement, and US Cellular given its high multiple and
controlled status.
AT&T Initiating coverage with Overweight rating and $33 YE11 price target
AT&T is well positioned on the wireless side to capture the coming wave of data
devices and currently supports both Apples iPad as well as all of the eReaders that
are sold wirelessly enabled in the US. We expect postpaid wireless subscribers to
grow more slowly than the industry in 2011 due to AT&T losing exclusivity on the
iPhone, but still expect wireless top-line growth of 5.5% and EBITDA growth of
10% as a result of additional margin expansion. In wireline, AT&T has seen slowing
consumer revenue losses as data and video become more important, and stabilization
in enterprise trends which should improve with the economy.
Among the largest telecom players, we prefer AT&T over Verizon given its better
cash flow characteristics and far lower EPS multiple, which we believe offset the
risks created by the changes in iPhone exclusivity. Not only should AT&Ts EPS
grow faster in 2011 but the multiple should also expand once the iPhone risk has
been quantified. We model $2.50 in EPS for AT&T in 2011, which represents 9%
growth from estimated 2010 EPS of $2.29.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Verizon Initiating with Neutral rating and $38 YE11 price target
Verizon is also in a strong position in the wireless space, and postpaid subscriber
growth should exceed that of the industry in 2011 due to both its overall network
quality and the addition of the iPhone to its handset lineup. Margins though are likely
to compress somewhat with higher smartphone subsidy costs, so we expect no
wireless EBITDA growth vs. top-line growth of 7%. In wireline, we also see
stabilizing revenue trends as video and data growth should outweigh voice losses,
and we expect some rebound in margins.
While we like Verizons business, the recent move in the stock price to 16.1x 2011E
EPS forces us to recommend avoiding the stock. We would consider coming back to
the stock once the excitement about the iPhone has receded further and if the stock
retreats to more attractive multiples of EPS and free cash flow. We model $2.23 in
2011 EPS, or 4% growth from estimated 2010 EPS of $2.14 (ex SpinCo).
Sprint Initiating with Overweight rating and $7 YE11 price target
We see Sprint as the stock in our universe with the largest potential upside due to its
continued operational turnaround and opportunity to expand margins from the
current industry-low level. While we expect that margin expansion to happen in
2012-13 rather than 2011, the company should be able to stabilize margins here
given a stable top line. We also expect cash flow generation of $1.2 billion in 2011
(9% yield) which should go to paying down debt or possibly further investment in
Clearwire, once the companies iron out their differences. We model EBITDA of
$5.55 billion in 2010 and $5.42 billion in 2011.
MetroPCS Initiating with Overweight rating and $16 YE11 price target
MetroPCS has executed well in the last 12 months despite broad secular concerns
regarding competition. The company has managed to improve its churn and ARPU
profile, while growth has slowed but remained fairly robust. From a profitability and
cash flow perspective, Metro is in an attractive position as pricing should remain
stable, supporting EBITDA growth. Free cash flow, however, is likely to ramp more
meaningfully as capital expenditures decline and move closer to maintenance levels.
Despite outperforming its closest peer, Leap Wireless, and the wireless group in
2010, MetroPCS has additional upside in 2011, in our view. We model $1.18 billion
in adjusted EBITDA in 2010 and $1.35 billion in 2011, and $394 million in free cash
flow in 2011, or an 8.5% yield at current prices.
Leap Wireless Initiating with Overweight rating and $16 YE11 price target
Leap executed poorly through most of 2010 but changes started in 3Q and completed
in 4Q put the company back on a growth path in 4Q10, which should accelerate in
1Q11. From a profitability and cash flow perspective, Leaps margins have been
under pressure through the transition, but as churn comes down and ARPU rebounds,
it appears poised for margin expansion. Free cash flow is likely to ramp meaningfully
in 2011 as well, as capital expenditures decline and move closer to maintenance
levels. We model $533 million in adjusted EBITDA in 2010 and $608 million in
2011.
We recommend owning shares of Leap Wireless and MetroPCS into the companies
traditionally strong first quarters, though we prefer Leap Wireless given its earlierstage turnaround.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Clearwire Initiating with Neutral rating and $6 YE11 price target


In our view, Clearwire has a tremendous competitive advantage in its deep wireless
spectrum portfolio and wholesale relationships with Sprint, Comcast, and Time
Warner Cable, but these are being diluted by strategic confusion and an overly
aggressive retail effort. In addition, we forecast the company will need another
$7 billion in capital to complete a buildout of 225 million pops and reach positive
free cash flow. Until Sprint and Clearwire get their build efforts on the same page,
we see significant risks to Clearwire from a fundamental standpoint and recommend
staying on the sidelines.
nTelos Initiating with Neutral rating and $22 YE11 price target
Improved execution in wireless combined with accretive wireline acquisitions should
drive better financial trends in 2011. In addition, management recently announced
plans to separate its wireless and wireline businesses, both of which we believe are
attractive strategic assets. However, at current prices, much of the fundamental
upside potential to nTelos is already priced in, we believe, and with transaction
restrictions after the planned spinoff, we recommend staying on the sidelines and
looking for a better entry point.
US Cellular Initiating with Underweight rating and $45 YE11 price target
We expect US Cellulars wireless business to struggle to grow; while recent moves
to improve subscriber and margin performance should help to some degree, we do
not expect a meaningful change given heavy competition. In addition, shares trade at
a premium valuation of 6.0x 2011E EBITDA, free cash flow is weak, and there is no
dividend. Finally, the Carlson family controls US Cellular through TDS, a structure
we believe is unlikely to change.
Telephone & Data Systems Initiating with Neutral rating and $42 YE11 target
For Telephone & Data Systems (TDS), the companys 83% stake in US Cellular
represents 94% of its enterprise value, leaving the implied wireline business trading
at only 1.0x 2010E EBITDA. This represents a big discount to the group and an
attractive level, despite the family control, so we rate TDS shares Neutral.
Frontier Initiating with Overweight rating and $11 YE11 price target
We prefer Frontier within our RLEC coverage as we find the 7.8% dividend yield,
the highest within our telecom coverage universe and the highest in the S&P 500,
attractive and believe it is sustainable. The company is undergoing a major transition
to integrate the acquired Verizon assets, which increases its risk profile, though early
signs from initial integration milestones have been comforting. Nevertheless, we
expect Frontier to put in the necessary capital investment as well as implement its
local engagement strategies to begin turning around these markets as early as 1H11.
The roadmap to an operational turnaround, likely upward revisions on synergy
guidance, as well as an attractive dividend make Frontier our top pick within RLECs.
Windstream Initiating with Neutral rating and $14 YE11 price target
Windstream has a rural footprint with the lowest access line density within our
RLEC coverage, which has enabled the company to deliver better operating metrics
than its peers. Moreover, management has been shifting its revenue stream away
from the declining consumer voice business and towards business and data segments,
which we think positions the company well for top-line stabilization when macro
trends improve. While the current dividend yield of 7.4% remains one of the highest
within telecom, the companys higher leverage and EV/EBITDA multiple compared
to those of wireline peers, as well as ongoing acquisition risk, keep us sidelined.
8

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless Tower Stocks


Though there are small differences in strategy, fundamentally the wireless tower
business remains more or less homogenous and exposed to strong secular tailwinds.
We recommend all three companies, American Tower, Crown Castle, and
SBA Communications, with Overweight ratings, and among them prefer
SBA Communications given its higher growth potential and flexibility.
American Tower Initiating with Overweight rating and $60 YE11 price target
American Tower, with its scale, conservative financial position, and focus on
profitability, has been the most stable tower company and offers an attractive
risk/reward at this level, in our view. International opportunities could drive higher
returns in comparison to those in domestic markets and the company has diversified
into eight countries already, with no country but the US accounting for more than
10% of revenue. The firms relatively low leverage at 4x EBITDA provides us with
additional comfort and gives the company the ability to consider a larger tower deal
or buy back stock. Finally, we see REIT conversion as the most likely case for
American Tower and include it in our target valuation of $60 per share; however, if
this does not occur, our estimated YE11 valuation for the company would be only
$45, implying 11% downside from the current level.
Crown Castle International Initiating with Overweight rating $55 YE11 target
Crown Castle has substantial scale and an attractive tower portfolio in the top 100
markets. We are confident Crown Castle will be able to turn the positive business
trends into solid revenue, EBITDA, and cash flow growth and view shares of the
company as attractive at current levels. In addition, leverage at Crown Castle is
relatively modest at about 5x, which provides us with additional comfort and gives
the company the ability to execute a larger tower deal or buy back stock. Finally, we
see REIT conversion as the most likely case for Crown Castle given NOLs are
forecast to run out in 2016 and include it in our target valuation of $55 per share. We
see little downside risk to our valuation if a conversion is delayed due to minimal net
impact of tax changes.
SBA Communications Initiating with Overweight rating and $53 YE11 target
SBA has been a superior operating and financial performer; as a result, we believe
SBA warrants a premium valuation. SBA has decided to not sign a master lease
agreement (MLA) with AT&T, and so should report faster growth in 2H11 and 2012
than its peers as AT&T signs LTE amendments. SBA has started to expand
internationally with a focus on markets that are fairly mature but still offer higher
growth and better return prospects than the US. While we see REIT conversion as a
possibility when the companys NOLs run out, we dont think this will happen in our
forecasting time frame (before 2020).

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Telecom and Tower Sector Risks


High competition in wireless and wireline telecom spaces
The telecom sector remains one of high and aggressive competition. In wireless
especially, there are an average of 5-6 players per market and Clearwire and
Lightsquared represent two potential new entrants that may have a lower cost
structure than incumbent players. In the fixed broadband/video universe, cable and
telecom companies aggressively target each others customers and temporary price
discounts are the norm.
Slowing subscriber growth
The pay TV, fixed broadband, and wireless voice markets are essentially saturated,
and we expect share shifts to prove more important than industry growth as a driver
of carrier success. As the pie of new subscribers shrinks, there is risk that weaker
players could get more aggressive on price and drive down margins across the
industry.
High capital intensity with steady upgrade cycles
The telecom space generally requires large fixed costs in terms of spectrum
acquisition and network build, so a struggling entity will often find it better to cut
prices to a marginally positive (or potentially negative) cash return than exit the
business. Given the ongoing shift to 4G technology in wireless, we expect capital
intensity to increase at most carriers in the next two years. Finally, while Verizon has
taken the hit on its wireline fiber upgrade and can expect to see lower wireline capital
intensity going forward, AT&T has taken a more gradual approach and we expect its
capex levels to remain elevated. Wireless carriers may also require additional
spectrum to support the rapidly increasing demand for data services.
Consolidation potential
The telecom space remains fragmented, and we expect to see additional wireless
consolidation as well as regional wireline roll-up in the next five years. We would
view another major merger as a long-term negative for the tower companies. In
addition, with the US market saturated, international acquisition risk is likely higher,
as domestic growth slows and cash flows are strong. The major independent tower
operators either own towers already or could consider acquiring towers in
international markets. Building up a bigger presence in any emerging market, in
which the wireless industry is still developing and country-specific risks are very
different from those in the United States, could negatively impact a companys risk
profile and multiple.
iDEN shutdown could be a 3%-plus headwind for the tower industry
We expect Sprint Nextel to shut down its 20,000 iDEN cell sites in the 2012-2016
time frame, with the bulk coming in 2013-2015. We model iDEN site shutdowns
weighing on our base-case revenue scenario of 68,000 sites by 4% in 2012, and we
expect an incremental ~600 bps in dilution annually until 2015. Peak dilution of 26%
is expected in 2016, a substantial negative impact if CDMA augmentations do not
partially offset the revenue loss. We believe the wireless tower industry could lose
about 3% or more in site revenue from Sprint Nextel once its network is fully
optimized.

10

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Tower debt refinancing could be at risk if credit markets close or tighten


The tower companies have substantial leverage that ranges from 4 times to 7 times
with debt maturities in the 2014-20 time frame. The companies need to be able to
refinance debt as it matures. If the credit markets were closed or refinanced debt bore
substantially higher interest rates, the tower companies would be negatively affected.
High multiple valuations could come under pressure in a volatile environment
The tower companies have high forward EV/EBITDA and EV/revenue multiples of
about 14-17x and 9-11x, respectively, substantially higher than those for the
S&P 500. Disruptions in the financial markets or negative earnings results could
dramatically reduce the multiples investors are willing to pay for tower company
stocks. Any operational missteps could also lead to a higher risk profile for the tower
companies, especially given their reputation for stability.

11

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 1: Telecom Valuation Comparison


($ in millions, except per share data)

Company

Rating/PT

Price
1/12/11

OW - $33.00

$28.85

Fundamentals
Upside
52-Week
to Target
Range

2010
Return

YTD
Return

Current
Sh Out

Market
Cap

Valuation and Leverage


2010E
Adj. Ent. Net Debt/
Value
EBITDA
Net Debt

Dividend
Net Debt/
Capital

Dividend

Yield

Payout Ratio
2011E
2012E

2010E

EPS
2011E

2012E

Earnings Multiples
P/E Ratio
2010E
2011E

Operational Metrics
2012E

Relative P/E Multiple


2011E
2012E

2010E

2011E

2012E

Diversified Telecom Carriers


AT&T (T)

30-24

4.8%

-1.8%

5,938

171,311

66,073

227,754

1.5x

29%

1.72

6.0%

70%

65%

2.29

2.50

2.64

12.6x

11.5x

10.9x

0.77x

0.84x

Verizon Communications (VZ) - ex Vodafone

N - $38.00

35.93

6%

38-25

15.6%

0.4%

2,830

101,682

40,316

137,019

1.7x

29%

1.95

5.4%

90%

128%

2.27

2.23

2.45

15.8x

16.1x

14.7x

1.08x

1.12x

Ntelos (NTLS)

N - $22.00

19.51

14%
13%

20-16

7.0%

2.4%

42

815

693

1,507

2.7x

46%

1.12

5.7%

62%

54%

1.07

1.70

2.06

18.3x

11.4x

9.5x

0.77x

0.72x

Telephone & Data Systems (TDS)

N - $42.00

35.94

17%

38-29

7.8%

-1.7%

106

3,798

693

4,491

0.6x

15%

0.45

1.3%

29%

24%

1.44

1.66

1.92

25.0x

21.7x

18.8x

1.45x

1.43x

Wireless Carriers
Clearwire (CLWR)

Wireless Subscribers (m)


N - $6.00

5.61

7%

9-5

-23.8%

8.9%

986

5,530

2,853

8,384

-2.3x

34%

(1.79)

(2.74)

(2.35)

NM

NM

NM

NM

NM

Leap Wireless (LEAP)

OW - $16.00

13.72

17%

19-10

-30.1%

11.9%

76

1,042

2,256

3,298

3.7x

68%

(9.15)

(2.87)

(1.36)

NM

NM

NM

NM

NM

5.5

5.9

MetroPCS (PCS)

OW - $16.00

12.95

24%

14-6

65.5%

2.5%

356

4,616

3,289

7,905

2.4x

42%

0.34

0.86

1.10

38.5x

15.0x

11.8x

1.00x

0.90x

8.2

4.3

9.0

9.7

Sprint Nextel (S)

OW - $7.00

4.68

50%

5-3

15.6%

10.6%

2,988

13,983

15,443

24,898

2.8x

62%

(1.14)

(0.45)

(0.08)

NM

NM

NM

NM

NM

49.6

51.4

53.2

US Cellular (USM)

UW - $45.00

49.52

-9%

51-34

17.8%

-0.8%

87

4,291

820

5,111

1.0x

16%

1.69

1.93

2.34

29.4x

25.7x

21.1x

1.72x

1.61x

6.1

6.1

6.2

Rural Carriers
Frontier Communications Corp. (FTR)
Windstream Corp. (WIN)

8.9

13.5
6.3

Access Lines (m)


OW - $11.00

9.59

15%

10-7

24.6%

-1.4%

989

9,485

7,930

17,414

3.0x

46%

0.75

7.8%

70%

70%

0.36

0.38

0.42

26.7x

25.0x

22.7x

1.68x

1.73x

5.1

4.9

4.7

N - $14.00

13.51

4%

14-6

26.8%

-3.1%

480

6,481

6,490

12,971

3.3x

50%

1.00

7.4%

70%

72%

0.83

0.84

0.85

16.2x

16.2x

15.9x

1.08x

1.21x

3.3

3.2

3.1

Wireless Towers

Towers (k)

American Tower (AMT)

OW - $60.00

50.50

19%

54-38

403

20,374

4,299

24,674

26.6x

3.18x

2.03x

34.0

34.7

35.3

Crown Castle (CCI)

OW - $55.00

42.60

29%

44-34

12.3%

-2.8%

286

12,189

6,007

18,195

4.8x

33%

(1.22)

0.33

0.61

NM

NM

NM

NM

NM

23.9

24.0

24.2

SBA Communications (SBAC)

OW - $53.00

39.66

34%

41-30

19.8%

-3.1%

115

4,551

2,709

7,260

6.3x

37%

(1.60)

(0.80)

(0.61)

NM

NM

NM

NM

NM

8.8

9.1

9.3

1,287-1,011

12.8%

1.1%

61.77

85.30

97.18

20.6x

14.9x

13.1x

1.00x

1.00x

S&P 500

1,271.50

19.5%

-2.2%

2.8x

17%

1.9%

0.95

1.07

1.90

NM

47.4x

Valuation Ratios

Financial Highlights and Profitability


Revenue
Company

Revenue Growth

2010E

2011E

2012E

124,125

124,628

125,696

75,136

77,570

78,439

3.2%

546

636

656

16.6%

4,990

5,032

5,100

0.8%

EBITDA

EBITDA Growth

11E/10E

12E/11E

2010E

2011E

2012E

0.4%

0.9%

42,543

44,234

45,253

1.1%

23,606

23,080

24,471

3.2%

218

257

279

17.6%

1.4%

1,101

1,136

1,198

3.1%

Free Cash Flow

FCF Growth

11E/10E

12E/11E

2010E

2011E

2012E

11E/10E

4.0%

2.3%

13,890

14,567

15,642

-2.2%

6.0%

9,057

6,151

4,323

8.6%

58

75

86

30.5%

5.4%

153

165

200

8.0%

Free Cash Flow Yield

EV/EBITDA

EV/Revenue

Operational Valuation

Price/FCF

12E/11E

2011E

2012E

2010E

2011E

2012E

2010E

2011E

2012E

2011E

2012E

4.9%

7.4%

8.5%

-32.1%

-29.7%

6.0%

4.3%

1.8x

1.8x

1.7x

5.8x

5.9x

5.6x

14.0%

9.3%

10.6%

9.1%

1.8x
2.8x

1.8x
2.4x

1.8x
2.3x

5.4x
6.9x

5.1x
5.9x

5.0x
5.4x

11.8x
10.8x

11.0x

16.5x

9.5x

20.9%

4.4%

5.3%

0.9x

0.9x

0.9x

4.1x

4.0x

3.8x

23.0x

23.5x
19.0x

2011E

2012E

Diversified Telecom Carriers


AT&T (T)
Verizon Communications (VZ) - ex Vodafone
Ntelos (NTLS)
Telephone & Data Systems (TDS)
Wireless Carriers
Clearwire (CLWR)

EV/Subscriber
560

1,352

2,305

141.6%

70.4%

(1,264)

(1,268)

(1,034)

NM

NM

(3,579)

(2,772)

(2,808)

NM

NM

NM

NM

15.0x

6.2x

3.6x

NM

NM

NM

NM

NM

Leap Wireless (LEAP)

2,521

2,847

3,066

12.9%

7.7%

533

608

716

14.1%

17.8%

(13)

14

18

NM

NM

1.4%

1.7%

1.3x

1.2x

1.1x

6.2x

5.4x

4.6x

73.3x

59.3x

557

522

MetroPCS (PCS)

4,060

4,548

4,909

12.0%

7.9%

1,177

1,349

1,515

14.7%

12.3%

(12)

394

585

NM

NM

8.5%

12.7%

1.9x

1.7x

1.6x

6.7x

5.9x

5.2x

11.7x

7.9x

883

816

Sprint Nextel (S)

32,354

32,405

32,794

0.2%

1.2%

5,551

5,420

5,670

-2.4%

4.6%

1,799

1,276

1,219

-29.1%

-4.5%

9.1%

8.7%

0.8x

0.8x

0.8x

4.5x

4.6x

4.4x

11.0x

11.5x

484

468

4,183

4,229

4,300

1.1%

1.7%

825

858

919

4.0%

7.2%

56

51

92

-8.8%

81.4%

1.2%

2.1%

1.2x

1.2x

1.2x

6.2x

6.0x

5.6x

84.5x

46.6x

831

829

US Cellular (USM)
Rural Carriers

946

621

EV/Access Line

Frontier Communications Corp. (FTR)

5,685

5,383

5,192

-5.3%

-3.6%

2,703

2,646

2,624

-2.1%

-0.8%

836

1,064

1,067

27.3%

0.3%

11.2%

11.2%

3.1x

3.2x

3.4x

6.4x

6.6x

6.6x

8.9x

8.9x

3,581

3,713

Windstream Corp. (WIN)

3,864

3,827

3,800

-1.0%

-0.7%

1,945

1,946

1,929

0.1%

-0.9%

600

686

668

14.4%

-2.6%

10.6%

10.3%

3.4x

3.4x

3.4x

6.7x

6.7x

6.7x

9.4x

9.7x

4,108

4,229

Wireless Towers

EV/Tower

American Tower (AMT)

1,976

2,277

2,422

6.4%

1,344

1,562

1,667

16.3x

710,676

698,603

Crown Castle (CCI)

1,871

2,003

2,119

7.1%

5.8%

1,165

1,262

1,359

8.3%

7.7%

634

727

830

14.8%

14.2%

6.0%

6.8%

9.7x

9.1x

8.6x

15.6x

14.4x

13.4x

16.8x

14.7x

757,038

750,791

624

675

727

8.1%

7.7%

385

429

471

11.3%

9.9%

220

261

301

18.3%

15.4%

5.7%

6.6%

11.6x

10.8x

10.0x

18.9x

16.9x

15.4x

17.5x

15.1x

802,165

781,442

SBA Communications (SBAC)

(1) Adjusted EV excludes a value for investments that do not contribute to revenue and EBITDA.
(2) Free cash flow before dividends. Free cash flow for towers is recurring free cash flow, excluding impact of discretionary capital expenditures.
(3) EPS from operations for T, VZ, NTLS, TDS, LEAP, PCS, S, USM
(4) Share count is last reported; net debt is 2010E

Source: Company reports and J.P. Morgan estimates.

12

15.2%

16.2%

6.8%

1,027

1,139

1,253

11.0%

10.0%

5.6%

6.2%

12.5x

10.8x

10.2x

18.4x

15.8x

14.8x

17.9x

North America Equity Research


13 January 2011

Wireless Industry

Wireless Industry

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

13

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

14

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless Key Investment Points


Wireless fundamentals shift to data from voice
We expect wireless voice adds to decline again in 2011, and forecast only 2.8 million
postpaid adds among the Big 4 carriers, down from 3.7 million in 2010. Instead, data
should be the big driver of value creation, as consumers trade up to smartphone plans
with higher monthly bills and add 3G-enabled tablets and MiFi devices to their
everyday usage list. An improving economy could also benefit carriers that target
lower demographics like Leap Wireless and MetroPCS.
iPhone exclusivity will slow AT&T growth, but expect impact to be moderate
Verizon will begin selling the iPhone on February 10, and we model AT&T losing
100k postpaid subs in 1Q and 300k in 2Q before slowly turning back to positive
adds. We model AT&T Mobility margins expanding by 160 basis points in 2011 on
slower growth and upgrade expense, while we expect Verizon Wirelesss to contract
by 265 basis points as faster upgrades and smartphone mix raise subsidy expense.
We expect that AT&Ts aggressive move to get customers on contract, high
enterprise and family plan mix, and superior iPhone capabilities will keep customer
churn to a reasonable level.
Sprint turnaround to continue; network transition impacts minimal in 2011
We expect Sprint to add postpaid subscribers in 4Q10 for the first time in over three
years on slowly moderating churn and improved CDMA gross add traffic. While we
expect capex to expand to $2.8 billion in 2011, the company should still generate
$1.2 billion in FCF on stable EBITDA, or an 8.8% yield. Given the continued
turnaround, we look at Sprint as a great secular holding in the telecom space, with
the greatest potential for stock price appreciation over the next two years.
We favor stocks with highest cash flow yields (AT&T) and potential turnaround
stories (Sprint, Leap Wireless)
Among the largest telecom players, we prefer AT&T over Verizon given its better
cash flow characteristics and low EPS multiple, which we believe offset the risks
created by the loss of iPhone exclusivity. Not only should AT&Ts EPS grow faster
in 2011 but the multiple should also expand once the iPhone risk has been quantified.
Among the wireless names, we like the risk/reward on Sprint as the company
continues its turnaround, and also recommend owning Leap Wireless and MetroPCS
into the companies traditionally strong first quarters, though we prefer Leap
Wireless given its earlier-stage turnaround.

Wireless Stock Recommendations


AT&T Initiating coverage with Overweight rating and $33 YE11 price target
AT&T is well positioned on the wireless side to capture the coming wave of data
devices and currently supports both Apples iPad as well as all of the eReaders that
are sold wirelessly enabled in the US. We expect postpaid wireless subscribers to
grow more slowly than the industry in 2011 due to AT&T losing exclusivity on the
iPhone, but still expect wireless top-line growth of 5.5% and EBITDA growth of
10% as a result of additional margin expansion. In wireline, AT&T has seen slowing
consumer revenue losses as data and video become more important, and stabilization
in enterprise trends which should improve with the economy.
15

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Among the largest telecom players, we prefer AT&T over Verizon given its better
cash flow characteristics and far lower EPS multiple, which we believe offset the
risks created by the changes in iPhone exclusivity. Not only should AT&Ts EPS
grow faster in 2011 but the multiple should also expand once the iPhone risk has
been quantified. We model $2.50 in EPS for AT&T in 2011, which represents 9%
growth from estimated 2010 EPS of $2.29.
Verizon Initiating with Neutral rating and $38 YE11 price target
Verizon is also in a strong position in the wireless space, and postpaid subscriber
growth should exceed that of the industry in 2011 due to both its overall network
quality and the addition of the iPhone to its handset lineup. Margins though are likely
to compress somewhat with higher smartphone subsidy costs, so we expect no
wireless EBITDA growth vs. top-line growth of 7%. In wireline, we also see
stabilizing revenue trends as video and data growth should outweigh voice losses,
and we expect some rebound in margins.
While we like Verizons business, the recent move in the stock price to 16.1x 2011E
EPS forces us to recommend avoiding the stock. We would consider coming back to
the stock once the excitement about the iPhone has receded further and the stock
retreats to more attractive multiples of EPS and free cash flow. We model $2.23 in
2011 EPS, or 4% growth from estimated 2010 EPS of $2.14 (ex SpinCo).
Sprint Initiating with Overweight rating and $7 YE11 price target
We see Sprint as the stock in our universe with the largest potential upside due to its
continued operational turnaround and opportunity to expand margins from the
current industry-low level. While we expect that margin expansion to happen in
2012-13 rather than 2011, the company should be able to stabilize margins here
given a stable top line. We also expect cash flow generation of $1.2 billion in 2011
(9% yield) which should go to paying down debt or possibly further investment in
Clearwire, once the companies iron out their differences. We model EBITDA of
$5.55 billion in 2010 and $5.42 billion in 2011.
MetroPCS Initiating with Overweight rating and $16 YE11 price target
MetroPCS has executed well in the last 12 months despite broad secular concerns
regarding competition. The company has managed to improve its churn and ARPU
profile, while growth has slowed but remained fairly robust. From a profitability and
cash flow perspective, Metro is in an attractive position as pricing should remain
stable, supporting EBITDA growth. Free cash flow, however, is likely to ramp more
meaningfully as capital expenditures decline and move closer to maintenance levels.
Despite outperforming its closest peer, Leap Wireless, and the wireless group in
2010, MetroPCS has additional upside in 2011, in our view. We model $1.18 billion
in adjusted EBITDA in 2010 and $1.35 billion in 2011, and $394 million in free cash
flow in 2011, or an 8.5% yield at current prices.
Leap Wireless Initiating with Overweight rating and $16 YE11 price target
Leap executed poorly through most of 2010 but changes started in 3Q and completed
in 4Q put the company back on a growth path in 4Q10, which should accelerate in
1Q11. From a profitability and cash flow perspective, Leaps margins have been
under pressure through the transition, but as churn comes down and ARPU rebounds,
it appears poised for margin expansion. Free cash flow is likely to ramp meaningfully
in 2011 as well, as capital expenditures decline and move closer to maintenance

16

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

levels. We model $533 million in adjusted EBITDA in 2010 and $608 million in
2011.
We recommend owning Leap Wireless and MetroPCS into the companies
traditionally strong first quarters, though we prefer Leap Wireless given its earlierstage turnaround.
Clearwire Initiating with Neutral rating and $6 YE11 price target
In our view, Clearwire has a tremendous competitive advantage in its deep wireless
spectrum portfolio and wholesale relationships with Sprint, Comcast, and Time
Warner Cable, but these are being diluted by the companys governance distractions
and an overly aggressive retail effort. In addition, we forecast the company will need
another $7 billion in capital to complete a buildout of 225 million pops and reach
positive free cash flow. Until Sprint and Clearwire get their build efforts on the same
page, we see significant risks to Clearwire from a fundamental standpoint and
recommend staying on the sidelines.
nTelos Initiating with Neutral rating and $22 YE11 price target
Improved execution in wireless combined with accretive wireline acquisitions should
drive better financial trends in 2011. In addition, management recently announced
plans to separate its wireless and wireline businesses, both of which we believe are
attractive strategic assets. However, at current prices, much of the fundamental
upside potential to nTelos is already priced in, we believe, and with transaction
restrictions after the planned spinoff, we recommend staying on the sidelines and
looking for a better entry point.
US Cellular Initiating with Underweight rating and $45 YE11 price target
We expect US Cellulars wireless business to struggle to grow; while recent moves
to improve subscriber and margin performance should help to some degree, we do
not expect a meaningful change given the heavy competition. In addition, shares
trade at a premium valuation of 6.0x 2011E EBITDA, free cash flow is weak, and
there is no dividend. Finally, the Carlson family controls US Cellular through TDS, a
structure we believe is unlikely to change.

Wireless Investment Risks


High competition
The telecom sector remains one of high and aggressive competition. In wireless
especially, there are an average of 5-6 players per market and Clearwire and
Lightsquared represent two potential new entrants that may have a lower cost
structure than incumbent players. In addition, the rise of MVNO wireless companies
has lowered barriers to entry for specialized retailers like Wal-Mart and Best Buy to
enter the market, potentially trimming off some of the largest sales channels for
prepaid carriers. In the fixed broadband/video universe, cable and telecom companies
aggressively target each others customers and temporary price discounts are the
norm.
Slowing subscriber growth
The pay TV, fixed broadband, and wireless voice markets are essentially saturated,
and we expect share shifts to prove more important than industry growth as a driver
17

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

of carrier success. As the pie of new subscribers shrinks, there is risk that weaker
players could get more aggressive on price and drive down margins across the
industry.
High capital intensity with steady upgrade cycles
The telecom space generally requires large fixed costs in terms of spectrum
acquisition and network build, so a struggling entity will often find it better to cut
prices to a marginally positive (or potentially negative) cash return than exit the
business. Given the ongoing shift to 4G technology in wireless, we expect capital
intensity to increase at most carriers in the next two years. Finally, while Verizon has
taken the hit on its wireline fiber upgrade and can expect to see lower wireline capital
intensity going forward, AT&T has taken a more gradual approach and we expect its
capex levels to remain elevated.
Additional spectrum spending may be necessary
Wireless carriers may require additional spectrum to support the rapidly increasing
demand for data services. While new 4G networks help, we expect additional
spectrum to come to market in the next five years which could cost the carriers in the
tens of billions of dollars. If existing carriers choose not to buy such spectrum, there
is further risk that its availability could spark another new entrant to the space.
Consolidation potential
The telecom space remains fragmented, and we expect to see additional wireless
consolidation as well as regional wireline roll-up in the next five years. In addition,
with the US market saturated, international acquisition risk is likely higher, as
domestic growth slows and cash flows are strong.

Wireless Industry Overview


We expect that total wireless industry additions increased in 2010 to 16.4 million
from 15.3 million in 2009, but look for a decline in 2011 to 12.7 million as postpaid
subscriptions decline faster and prepaid unlimited growth falls off. Within postpaid,
we look for 2.8 million additions in 2011, down from 3.7 million in 2010 and
4.6 million in 2009. There could be downside risk to our postpaid estimate if the
economy weakens further, but we think higher total customer additions are more
likely, as second devices, such as iPad, Kindle, and 4G MiFi devices, start to gain
mass-market adoption and add to prepaid and wholesale subscriber counts.
In our view, carriers with the greatest downside risk in postpaid are AT&T (due to
iPhone exclusivity loss) and T-Mobile USA (which is struggling for relevance),
while our estimates for Sprint dont assume much further improvement in churn or
gross add traction, and Verizon could do better on the iPhone launch and drive
incremental growth in data devices with its 4G efforts.
In the following sections, we look at the current state of the wireless industry and
how the industry has evolved, sources of future growth including both subscribers
and revenue per user, and some important potential drivers of market share shift such
as the end of iPhone exclusivity, the growth of emerging devices, and the shift to 4G.

18

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 2: We Expect Wireless Industry Reported Penetration Levels to Exceed 100% by 2012
Penetration
100%

Net Adds
30,000
25,000

80%

20,000
60%
15,000
40%
10,000
20%

5,000

0%
20
11
20
12

20
08
20
09
20
10

20
04
20
05
20
06
20
07

20
01
20
02
20
03

19
97
19
98
19
99
20
00

19
94
19
95
19
96

19
90
19
91
19
92
19
93

19
87
19
88
19
89

19
85
19
86

Source: Company reports, CTIA, and J.P. Morgan estimates.

How Did We Get to 97% Penetration? Where Can It Go?


We estimate that at the end of 2010 there were 302 million wireless devices in the
United States, out of a total population of 310 million. While this reported
penetration is 97% of the US population, we believe that double-counting, in the
form of people carrying a regular voice phone or smartphone in addition to a
Blackberry, 3G tablet, or laptop card, as well as machine-to-machine
communications, accounts for 15-20% of reported subscribers. In addition, the overreporting of subs due to inactive customers counted in the base, which have not yet
been churned off, likely accounts for another ~3-4% of subscribers, indicating actual
penetration of around 75-80%.
Many Asian and European countries report penetration of well over 100% as
customers carry multiple SIM cards to arbitrage price depending on who theyre
calling. We dont expect that to ever be a significant issue in the US.
Figure 3: Postpaid Represents ~83% of Wireless Market; Substantial Unlimited Growth Since 08
100%
95%
90%

Unlimited

85%

Prepaid

80%

Postpaid

75%
70%
2004

2005

2006

2007

2008

2009

2010E

Source: Company reports and J.P. Morgan.

Population growth, wireline substitution, data devices dominate wireless growth


Our thesis from here is that wireless penetration of the population will increase only
slowly, and that most of the incremental postpaid voice growth will come from
population growth. On the data side, 2009 and 2010 saw the introduction of eReaders
and tablet devices but we expect a major broadening of offers in 2011 and beyond,
and look for people carrying a tablet or MiFi device as well as a cell phone to drive
subscriber growth higher.

19

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Across other categories of wireless, there can be incremental voice penetration in


prepaid, we believe, as more customers migrate to wireless from landlines for
unlimited offers or USF-subsidized lines such as Assurance from Sprint or Safe Link
from Tracfone. In addition, data devices that come through the prepaid or wholesale
subscriber lines, such as iPad, Kindle, and GPS devices, can continue to ramp, and
we see no reason that total penetration has to stop growing.
In the table below, we break down our estimates of current real penetration and how
that translates across age groups, and our expectations for 2016.
Table 2: Estimated Wireless Penetration by Age Group
Age Group
Under 5 years
5 to 9 years
10 to 14 years
15 to 19 years
20 to 24 years
25 to 29 years
30 to 34 years
35 to 39 years
40 to 44 years
45 to 49 years
50 to 54 years
55 to 59 years
60 to 64 years
65 to 69 years
70 to 74 years
75 to 79 years
80 to 84 years
85 to 89 years
90 to 94 years
95 to 99 years
100 years and over
Total
Double Counting
Total

% of Population by
Age Group (2008) Population (2010)
6.8%
21,045
6.9%
21,402
7.2%
22,421
7.3%
22,549
6.9%
21,429
6.4%
19,842
6.8%
21,039
7.5%
23,235
8.1%
25,042
7.5%
23,324
6.6%
20,566
5.4%
16,709
4.2%
13,019
3.5%
10,755
3.0%
9,450
2.6%
8,193
1.9%
5,781
1.0%
3,178
0.5%
1,433
0.1%
447
0.0%
90
100.0%
310,950

Est. 2010
Wireless
Pen.
0%
50%
80%
92%
92%
92%
92%
92%
92%
92%
92%
92%
92%
92%
60%
60%
60%
60%
60%
50%
50%
79%
18%
97%

2010E
Wireless
Subs.
10,701
17,937
20,746
19,714
18,255
19,356
21,376
23,039
21,458
18,920
15,372
11,978
9,895
5,670
4,916
3,469
1,907
860
224
45
245,836
55,971
301,807

Population
(2016)
22,208
22,584
23,659
23,795
22,612
20,938
22,201
24,518
26,425
24,613
21,701
17,631
13,738
11,349
9,972
8,645
6,101
3,354
1,512
472
95
328,124

Est. 2016
Wireless
Pen.
0%
55%
85%
95%
95%
95%
95%
95%
95%
95%
95%
95%
95%
95%
70%
70%
70%
65%
65%
55%
55%
83%
26%
109%

2016E
Wireless
Subs.
12,421
20,110
22,605
21,482
19,891
21,091
23,292
25,104
23,382
20,616
16,750
13,051
10,782
6,980
6,052
4,270
2,180
983
260
52
271,355
86,264
357,618

Source: J.P. Morgan estimates.

Postpaid Subscriber Growth Moves from Voice to Data


We calculate there are currently more than 210 million postpaid customers in the US,
the vast majority of which represent voice-enabled devices. Our postpaid thesis is
that wireless voice penetration of the population will increase only slowly, and that
most of the incremental postpaid voice growth will come from population expansion
which, at the current rate of ~0.9% with 210 million postpaid subs currently, implies
postpaid voice adds of 2 million in 2011. From there, it is not hard to envision that
growth in the number of people who carry a tablet or MiFi device as well as a cell
phone can translate to our postpaid growth estimate of 2.8 million, and we believe
there could be upside to our estimate.
The reported growth of postpaid subscribers will depend very much on providers
focus and willingness to get involved in the data business. Verizon, for example,
already has the biggest datacard/MiFi business in the US with 5.5 million
subscribers, and we expect that to grow as the company introduces LTE devices and
an aggressive 4G marketing campaign. AT&T, however, is focused on enhancing its
network rather than aggressively going after datacard business.

20

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Sprint has an interesting situation as well. Its datacard business has ramped nicely in
the last year as it has pushed 3G/4G laptop cards and the Overdrive MiFi device,
and recently started to sell tablet devices such as the Samsung Galaxy Tab. The data
usage on these, however, often goes to the Clearwire network, of which Sprint owns
more than 50%. This relationship, however, has been bumpy and we wonder if Sprint
may start to rethink how aggressively it drives customers to Clearwires network.
Table 3: Postpaid Net Adds Across Big 4
1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10E

1Q11E

2Q11E

3Q11E

4Q11E

Net Adds
AT&T Mobility
Verizon Wireless
Sprint Nextel
T-Mobile USA

897
936
(1,250)
160

1,128
1,044
(991)
56

1,333
901
(801)
(140)

841
1,106
(504)
(117)

512
412
(578)
(119)

496
661
(228)
106

745
584
(107)
(60)

525
700
25
30

(100)
650
(25)
25

(300)
750
100
13

200
500
50
0

300
400
100
63

Net Add Share


AT&T Mobility
Verizon Wireless
Sprint Nextel
T-Mobile USA

121%
126%
-168%
22%

91%
84%
-80%
5%

103%
70%
-62%
-11%

63%
83%
-38%
-9%

226%
181%
-255%
-52%

48%
64%
-22%
10%

64%
50%
-9%
-5%

41%
55%
2%
2%

-18%
118%
-5%
5%

-53%
133%
18%
2%

27%
67%
7%
0%

35%
46%
12%
7%

Source: Company reports and J.P. Morgan estimates.

Postpaid market share has been stable at the top, shifting at the bottom
In addition to overall industry growth, the shifting share of churn and gross additions
among the postpaid players has become increasingly important. For example, from
2006 to 2008 Sprint went from adding 279,000 net customers in 2006 to losing over
4 million postpaid customers in 2008. At the same time, the industry went from
adding 14 million postpaid subs in 2006 to only 8.5 million in 2008, but because of
Sprints deterioration the other carriers growth was impacted only minimally. Now,
however, Sprint is on track to lose only ~900k subscribers in 2010 and add subs in
2011, as industry postpaid growth dropped to only ~3.7 million in 2010. Thus, the
other carriers went from adding 12 million postpaid subs in 2008 to only 4.6 million
in 2010 and an estimated 2.5 million in 2011.
This market share shift toward Sprint since 2008 has been driven by improvements in
the companys network, handset selection, and customer service, as well as steady
reduction in price. On the gross add side, as Sprints share rebounded from 12% in
3Q08 to over 18% in 3Q10, the loss was split evenly between AT&T and Verizon
while T-Mobiles share remained steady at 19%. Sprints share of disconnects fell a
similar 600 basis points, from 28% to 22% in the same period, again with AT&T and
Verizon making up the majority of the share shift.
Unfortunately for Sprint, the companys churn is still nearly double that for the
postpaid bases of AT&T and Verizon. If it cannot start to reduce churn from the
recent 1.9% level to at least the mid-1% range, the potential for the company to ever
grow subscribers or hold margins is tentative, in our view.

21

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 4: Sprints Postpaid Gross Add Share Has Rebounded

Figure 5: And Churn Has Come Down Significantly

100%

3.0%

80%

2.5%
2.0%

60%

1.5%

40%

1.0%

20%

0.5%

0%
2006

2007
T

2008
VZ

2009

2010E

0.0%
1Q06 3Q06 1Q07 3Q07 1Q08 3Q08 1Q09 3Q09 1Q10 3Q10

2011E

TM-USA

Source: Company reports and J.P. Morgan estimates.

VZ

TM-USA

Source: Company reports and J.P. Morgan estimates.

Figure 6: Resulting in a Slowdown in Net Sub Losses


2,500
2,000

Postpaid Net Adds

1,500
1,000
500
0
(500)
(1,000)
(1,500)
1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08
T

VZ

2Q08
S

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

TM-USA

Source: Company reports and J.P. Morgan estimates.

Prepaid Industry Increasingly Reliant on Data


Prepaid and pay in advance monthly plans today represent 21% of wireless
industry subscribers and in the last 12 months contributed 45% of industry growth.
These customers are broken up into three main categories the highest-spending
tend to be on pay-in-advance monthly plans in the $30-60 range, the middle tier tend
to buy minutes upfront and spend $15-30 per month, and the lowest are given service
for free through federal and state lifeline programs from which the carrier gets ~$10
per month.

22

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 4: Prepaid Net Adds Across the Industry


Thousands
1Q08

2Q08

3Q08

4Q08

2008

1Q09

2Q09

3Q09

4Q09

2009

1Q10

2Q10

3Q10

4Q10E

2010E

Prepaid net adds


AT&T Mobility
% of total CNG net adds

118
9%

12
1%

(36)
-2%

(23)
-1%

71
1%

(155)
-13%

(412)
-30%

(176)
-9%

(58)
-2%

(801)
-11%

24
1%

300
19%

321
12%

450
17%

1,095
13%

Sprint-Nextel
% of total S net adds

(200)
16%

(138)
17%

(329)
23%

(314)
19%

(981)
19%

674
-152%

777
1407%

666
-214%

435
-630%

2,552
-278%

348
-151%

173
-315%

471
129%

450
95%

1,442
260%

T-Mobile
% of total T-Mobile net adds

306
31%

143
21%

377
56%

355
57%

1,181
40%

255
61%

268
83%

63
-82%

488
132%

1,074
104%

41
-53%

(199)
214%

197
144%

270
90%

309
116%

Verizon
% of total VZW net adds

352
20%

127
7%

85
5%

171
10%

735
10%

268
22%

48
4%

15
1%

51
2%

382
7%

(146)
-10%

(207)
-15%

(137)
-14%

(80)
-6%

(570)
-11%

US Cellular
% of total USM net adds

14
19%

1
5%

(15)
83%

(8)
-40%

(8)
-9%

3
6%

(27)
31%

(14)
0%

13
130%

(25)
45%

33
236%

29
-264%

0
0%

20
100%

82
-456%

Leap Wireless
% of total LEAP net adds

230
100%

171
100%

263
100%

385
100%

1,049
100%

493
100%

203
100%

116
100%

298
100%

1,110
100%

446
100%

(112)
100%

(200)
100%

50
100%

184
100%

MetroPCS
% of total PCS net adds

452
100%

184
100%

249
100%

520
100%

1,404
100%

684
100%

206
100%

66
100%

317
100%

1,272
100%

692
100%

303
100%

223
100%

350
100%

1,568
100%

381
100%
1,671
43%
-29%
43,131
17%
23%

213
100%
601
18%
-50%
43,774
17%
21%

341
100%
932
24%
-10%
44,741
17%
20%

743
100%
2,045
44%
-25%
46,794
18%
17%

1,678
100%
5,253
33%
-28%
46,794
18%
17%

567
100%
2,655
65%
59%
49,318
19%
14%

730
100%
1,523
42%
153%
50,680
19%
16%

712
100%
1,624
42%
74%
51,961
19%
16%

1,199
100%
2,743
48%
34%
54,465
20%
16%

3,208
100%
8,696
47%
66%
54,639
20%
17%

1,052
100%
2,489
59%
-6%
57,130
21%
16%

460
100%
747
25%
-51%
57,915
20%
14%

745
100%
1,620
39%
0%
59,504
21%
15%

750
100%
2,260
44%
-18%
61,764
21%
13%

3,007
100%
7,117
43%
-18%
61,937
21%
13%

Tracfone
% of total TF net adds
PREPAID NET ADDS
% of total
% change, y-t-y
Total prepaid ending subs
% of total
% change, y-t-y

Source: Company reports and J.P. Morgan estimates.

Voice and text commoditized as barriers to entry disappear


At the low end of the wireless market, voice and text offerings in the $30-40 range
have become heavily commoditized with virtual operators like Tracfone and WalMart pressuring pricing at the facilities-based operators like Leap Wireless,
MetroPCS, and Sprints Boost and Virgin offerings. Whereas two years ago
consumers had to spend $40+ for unlimited voice and text service, this can mostly be
had today for as little as $30 per month. This has driven the facilities-based operators
to offer a higher handset mix and data options to differentiate from the MVNOs.
Prepaid Android devices are
flying off the shelves

Better handsets, data plans could reflate ARPU


Until a year ago the prepaid and pay in advance markets were strictly focused on
voice and text, offering a low-end handset and service at low prices. As weve seen
in the past though, the last year has demonstrated that the low-end consumer wants
access to many of the same services (internet, music, etc.) and handset options
(QWERTY, touch, etc.) that attract the postpaid customer. Carriers like MetroPCS
and Leap Wireless that offer an attractive handset like an Android touch device have
seen far faster uptake than they expected, and a nice lift in ARPU from those
customers.
While MetroPCS does not forecast any increase in ARPU going forward, it does
expect to be able to hold the ~$40 level, and Leap Wireless has said that it expects
ARPU uplift from 3Q10 levels as data customers at the $50-60 level replace churn in
the$30-40 range.

23

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 7: Cricket Partial Handset Lineup

Samsung Stunt
$40/Free After Rebate

Samsung Messager
Touch
$160/$80

Huawei Ascend
(Android)
$160/$100

Blackberry Curve
8530
$300/$180

Sanyo ZIO by Kyocera


(Android) $250/$130

Source: Company reports (used with permission).

Figure 8: MetroPCS Partial Handset Lineup

Huawei M228
$10 After Instant
Savings

LG Select

$59

LG Baanter Touch
$149

Samsung Craft $299


After Instant Savings

Source: Company reports (used with permission).

How long the facilities-based carriers have this data business to themselves is a big
question. Tracfones Straight Talk brand already offers a number of low-end
QWERTY and touch devices, and at some point we expect carriers like T-Mobile
and Verizon to enable their MVNOs to offer postpaid-equivalent data applications.
AT&T and Verizon put little effort
into prepaid voice

AT&T and Verizon prepaid voice offers continue to drift


Verizon recently lowered its branded prepaid price from $4 to $2 per day for
unlimited voice calling to make it comparable with AT&Ts, but both are still far
more expensive than most prepaid offerings. We see these offers as targeted to
prepaid-credit-quality customers who care about being on the high-quality Verizon
and AT&T networks, but dont expect them to return to growth any time soon.
AT&T prepaid growth totally iPad-driven
In addition to the prepaid voice offers that have typically made up this category at
AT&T and Verizon, AT&T in 2010 started to offer a prepaid data plan on Apple
iPads sold with a 3G modem. We estimate that AT&T added about 400,000 3G iPad
customers in each of the second and third quarters of 2010, and likely substantially
more in 4Q10. These typically come with a $25 ARPU, which is slightly dilutive to
AT&Ts traditional ARPU for prepaid in the high $20s.

ARPU Trends Mixed Data Drives Postpaid, but Additional


Devices Dilute Headline Numbers
Average revenue for all users has fallen steadily by 1-2% in each of the last few
years, and we expect this to continue as industry growth is increasingly driven by
prepaid voice subs at the low end and second data devices with lower-than-average
prices at the high end. Offsetting this is the trend toward increasing postpaid ARPU
(especially at AT&T and Verizon) as customers increasingly trade up to smartphones
with $15-30 data plans, driving increases of ~2% annually.

24

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 9: Total ARPU Has Been on a Steady Decline


$65
$60
$55

$54.30

$54.73

$54.63

$52.24

$51.88

$51.19

$51.53

$50

$50.86 $49.86
$48.73

VZ

Sprint Nex tel

TM-USA

20
10

20
09

20
08

20
07

20
06

20
05

20
04

20
03

20
02

20
01

$45

National carrier av erage

Source: Company reports and J.P. Morgan estimates.

Drop in voice ARPU driven more


by dilution than price drop

Voice ARPU continues down due to family plan dilution, high-end price cuts
Wireless voice ARPU continues to decline, driven by price declines at the high end,
family plan expansion, and dilution from data-only devices like laptop cards and
MiFi. We estimate that postpaid voice ARPU is $37 currently, down from $50+
levels five years ago.
For individuals, while standard postpaid price levels at the low end have stayed
relatively flat at $35-60 as more and more minutes have been bundled into the plan,
at the high end unlimited voice plan prices from postpaid carriers have tumbled from
$200-300 three years ago to $60-100 today. Unlimited voice can be had from
postpaid players for as little as $60 at T-Mobile or $70 at Verizon Wireless.
In addition, most incremental postpaid voice subscribers in the last few years have
come in on family plans, on which the family can share a single bucket of minutes
for an additional $0-10 per line. Our checks indicate that family plan customers (both
the core and the add-on lines) now make up 50-70% of postpaid customers at each of
the Big 4 carriers, with the highest penetration at AT&T and Verizon Wireless.
Fortunately, most of these incremental family customers also sign up for an SMS or
data package, and we believe total incremental ARPU from a family plan customer is
roughly $30.
Figure 10: Voice Accounts for ~65% of Total Postpaid ARPU
90%
85%
80%
75%
70%
65%
60%
55%
2006

2007
T

2008
VZ

2009
TM-USA

2010E

2011E

AVERAGE

Source: Company reports and J.P. Morgan estimates.

25

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Data-only connections are dilutive to voice ARPU


CTIA estimates there are 13.0 million data-only laptop cards and MiFi devices in use
in the US, up 20% from 10.8 million two years ago.
Verizon has 5.5 million laptop
customers to AT&Ts 1.5 million

Verizon internet device growth diluted voice ARPU by 1% in last 12 months.


Verizon Wireless has roughly 5.5 million internet-only devices on its network (data
cards, MiFi, and netbooks), up from 4.5 million a year ago. These additional devices
(~1 million) count as part of the companys postpaid net additions in the last year,
and account for 15% of the companys reduction of $2.50 in voice ARPU during the
same period.
Connected devices diluted AT&T consolidated ARPU by 5% in 2010. At AT&T,
the company has roughly 1.5 million laptop card-like devices, a level that has
remained fairly steady through the last couple of years, and we estimate that postpaid
subscriber growth and ARPU have been generally unaffected. At the total subscriber
and ARPU level, however, the company added 4.5 million connected devices in
the last 12 months, all at an ARPU of only $3-4 per device. We estimate that the
majority of these were Kindle or similar eReader devices (AT&T also supports the
other big eReaders including Barnes & Nobles Nook and Sonys eReader). Thus,
these connected devices made up 52% of total subscriber growth at AT&T in the
last 12 months and we estimate diluted total ARPU by 500 basis points. In other
words, if we counted only the retail business at AT&T, ARPU would have been up
1.8% from 3Q09 to 3Q10 instead of down 2.3% as reported.
but data is now 32% of total ARPU for the Big 4
Spending on data, including SMS, smartphones, data cards, etc., represented 32% of
industry service revenue in the third quarter of 2010, up from 23.2% in 3Q09. This
has been a function both of more customers signing up for SMS and data plans and
$25-30 smartphone data plans penetrating the mass market of wireless subscribers as
well as the early adopters. With new smartphones and data devices coming to market
almost every week, we expect data penetration and spending to continue to increase
in 2011. We delve more into the drivers of this phenomenon in the smartphone and
emerging devices sections, below.
Figure 11: Growth in Data ARPU Supports Wireless ARPU Growth

Postpaid data ARPU ($)

$30.00
$25.00
$20.00
$15.00
$10.00
$5.00
2006

2007
T

Source: Company reports and J.P. Morgan estimates.

26

2008
VZ

2009
TM-USA

2010E
AVERAGE

2011E

North America Equity Research


13 January 2011

Figure 12: Data Is Driving Wireless Revenue Growth


50,000
$ (m)

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

40,000
30,000

Data rev
Voice rev

20,000
10,000
1Q06

3Q06

1Q07

3Q07

1Q08

3Q08

1Q09

3Q09

1Q10

3Q10 1Q11E 3Q11E

Source: Company reports and J.P. Morgan estimates.

Handsets Smartphone Share Is Growing


Carriers Are Close to Smartphone Parity
Leaving the iPhone aside, the US wireless carriers have all improved their handset
offerings in the past year, to the point where customers looking for a Blackberry or
Android device do not need to exclude any carrier. While the Big 4 have gotten
better, the biggest changes have come at the regional level where carriers like Leap,
MetroPCS, US Cellular, and nTelos have improved dramatically.

27

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 13: Smartphone Lineup Across Carriers

AT&T
Smartphone
/Tablet
Samples

iPhone 4 16/32 GB $199-$299

HTC Aria $129

Blackberry Torch $100

iPad 16/32/64 GB $629/$729/$829

Droid X by Motorola $199

Blackberry Bold 9650 $100

LG Vortex $40

Samsung Galaxy Tab $600

HTC EVO 4G $199

Samsung Epic 4G

Blackberry Bold 9650 $199

Samsung Galaxy Tab $400

T-Mobile myTouch 4G $200

Blackberry Bold 9780 $129

Samsung Vibrant free

Dell Inspiron Mini 10 4G $230

Verizon
Smartphone
/Tablet
Samples

Sprint
Smartphone
/Tablet
Samples

T-Mobile
Smartphone
/Tablet
Samples

Source: Company reports (used with permission).

Smartphones 43% of 3Q Sales at VZW, ~70% at AT&T


As the quality of smartphones has improved, customers are finding more utility in
them, and the share of smartphones as a percentage of handset sales has expanded
dramatically in the last year.
Verizon smartphones ramp with Android. At Verizon in particular, smartphones
represented 43% of postpaid handset sales in the third quarter of 2010, up from just
23% in 3Q09. A year ago that 23% was dominated by Blackberry devices like the
Curve and Storm, but starting with the Motorola Droid which launched in 4Q09
Android-based devices now dominate smartphone sales at Verizon, we believe. We
28

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

forecast that by 4Q11 smartphones will account for 75% of postpaid handset sales,
including 30% of sales for iPhones.
Table 5: Handset Sales Split at Verizon
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet
Feature phone
Postpaid handset sales

1Q09
0.0%
26.0%
3.0%
5.0%
66.0%
100.0%

2Q09
0.0%
21.0%
17.0%
6.0%
56.0%
100.0%

3Q09
0.0%
23.0%
30.0%
6.0%
41.0%
100.0%

4Q09
0.0%
30.0%
34.0%
6.0%
30.0%
100.0%

1Q10
0.0%
36.0%
24.0%
7.0%
33.0%
100.0%

2Q10
0.0%
40.0%
17.0%
6.0%
37.0%
100.0%

3Q10
0.0%
43.0%
13.0%
7.0%
37.0%
100.0%

4Q10E
0.0%
52.5%
15.3%
8.6%
23.5%
99.8%

1Q11E
20.0%
36.1%
0.3%
8.6%
33.7%
100.0%

2Q11E
30.0%
35.9%
3.4%
8.2%
21.2%
100.0%

3Q11E
30.0%
38.1%
-1.7%
9.0%
23.1%
100.0%

4Q11E
30.0%
45.4%
-4.4%
9.5%
17.9%
100.0%

Source: Company reports and J.P. Morgan estimates.

iPhone drove AT&T share far higher. At AT&T, the companys iPhone franchise
drove smartphone share high earlier than at Verizon. In 3Q10 we estimate that 80%
of postpaid handset sales were 3G integrated devices, which include smartphones
as well as dumb touch and QWERTY devices, and we estimate that true
smartphones made up close to 70% of sales, including iPhone sales which
represented 48% of total postpaid sales. This is up from 68% integrated in 3Q09
when true smart phones represented roughly 50% of postpaid handset sales.
Table 6: Handset Sales Split at AT&T
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Postpaid handset sales

1Q09
20.6%
36.4%
5.9%
1.4%
35.7%
100.0%

2Q09
27.8%
32.2%
0.0%
2.5%
37.5%
100.0%

3Q09
33.2%
34.8%
1.4%
1.8%
28.8%
100.0%

4Q09
34.0%
36.0%
0.0%
1.6%
28.4%
100.0%

1Q10
34.6%
35.4%
0.0%
1.9%
28.1%
100.0%

2Q10
38.5%
33.5%
0.0%
1.5%
26.5%
100.0%

3Q10
48.1%
31.9%
0.0%
1.2%
18.8%
100.0%

4Q10E
42.6%
38.4%
0.0%
1.4%
17.6%
100.0%

1Q11E
36.4%
46.6%
0.0%
1.7%
15.3%
100.0%

2Q11E
38.3%
44.7%
0.0%
1.5%
15.5%
100.0%

3Q11E
41.1%
43.9%
0.0%
1.3%
13.7%
100.0%

4Q11E
36.9%
48.1%
0.0%
1.4%
13.6%
100.0%

Source: Company reports and J.P. Morgan estimates.

Average Subsidies Should Increase with Smart Share


Smartphones have driven up handset subsidies in the last three years, and we expect
this trend to continue as the mix skews more to smartphones. We have also seen
carriers save some money as feature phone prices have dropped and carriers use
fewer of their subsidy dollars to upgrade customers who are not willing to step up
into a big data plan.
The iPhone from Apple is possibly the most heavily subsidized handset in the US
certainly on a gross dollar basis if not per device. We estimate that AT&T pays
Apple $425-650 per handset depending on the version a subsidy of approximately
$325-350 compared to the average postpaid subsidy which we estimate to be
~$200-250. AT&T sees this as a good trade though, given these customers generate
~1.7x average postpaid ARPU with lower churn.
Table 7: Subsidy Analysis at AT&T
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Total postpaid subsidy

1Q09
$350
$250
$100
$100
$57
$191

2Q09
$350
$180
$100
$100
$65
$182

3Q09
$325
$175
$100
$100
$68
$192

4Q09
$325
$125
$100
$100
$54
$173

1Q10
$325
$135
$100
$100
$57
$178

2Q10
$325
$160
$100
$100
$61
$196

3Q10
$350
$250
$100
$100
$52
$259

4Q10E
$325
$225
$100
$100
$54
$236

1Q11E
$325
$200
$100
$100
$143
$235

2Q11E
$325
$200
$100
$100
$53
$224

3Q11E
$325
$200
$100
$100
$61
$231

4Q11E
$325
$200
$100
$100
$81
$228

Aggregate handset subsidy expense


% of service revenue

1,439
12.4%

1,507
12.6%

1,738
14.0%

1,494
11.9%

1,307
10.2%

1,502
11.4%

2,550
18.7%

2,038
14.8%

1,803
13.0%

1,787
12.7%

2,125
14.9%

1,952
13.8%

Source: Company reports and J.P. Morgan estimates.

29

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

At Verizon, we estimate that postpaid subsidy will rise from ~$150 in 3Q10 to more
than $200 once it has sold the iPhone for a full quarter.
Table 8: Subsidy Analysis at Verizon
iPhone subsidy
Smartphone subsidy (non-iPhone)
3G Multimedia subsidy
Internet subsidy
Feature phone subsidy
Total postpaid subsidy

1Q09
$350
$250
$100
$100
$147
$170

2Q09
$350
$250
$100
$100
$194
$184

3Q09
$350
$250
$100
$100
$200
$176

4Q09
$350
$250
$100
$100
$242
$187

1Q10
$350
$250
$100
$100
$140
$167

2Q10
$350
$250
$100
$100
$131
$171

3Q10
$350
$250
$125
$125
$69
$158

4Q10E
$350
$230
$110
$125
$54
$161

1Q11E
$350
$225
$100
$120
$44
$177

2Q11E
$350
$200
$100
$120
$51
$201

3Q11E
$350
$200
$100
$120
$56
$203

4Q11E
$350
$180
$100
$100
$51
$201

Aggregate handset subsidy expense


% of service revenue

1,647
12.9%

1,696
13.1%

1,802
13.7%

1,908
14.5%

1,654
12.3%

1,756
12.7%

1,740
12.3%

1,690
11.9%

2,017
14.1%

2,485
17.0%

2,335
15.6%

2,215
14.8%

Source: Company reports and J.P. Morgan estimates.

Upgrades, Not Gross Adds, Drive Majority of Handset Sales


As churn has declined and customers have spent more time with a carrier in the last
five years, weve not seen a corresponding increase in the life of a handset. Instead,
carriers have become more willing to offer an upgraded handset to existing
customers, and this trend has accelerated as smartphones from Android and Apple hit
mass-market prices and consumers are willing to step up to a higher-priced data plan
to offset the carriers cost.
Upgrades were >60% of 2010
handset sales

We estimate that currently ~61% of handset sales by the Big 4 carriers are actually
upgrades rather than phones for new customers coming in the door, and that the
average handset life in the US is 18-24 months for a handset, and at the low end of
this range for a smartphone.

Upgrades % of handset sales

Figure 14: Upgrades Account for an Increasing Proportion of Handset Sales


75%
65%
55%
45%
35%
25%
2003

2004

2005

Big 4 av erage

2006
T

2007
VZ

2008
S

2009

2010

2011

TM-USA

Source: Company reports and J.P. Morgan estimates.

Unlocked Handsets Still DOA


For a little while in 2008 and 2009 carriers were encouraging customers to bring in
their own handsets in exchange for an activation discount or contract flexibility, but
this never really went anywhere.
Americans are well trained to not
pay full price for a handset

30

In 2008, MetroPCS rolled out MetroFlash, which allowed customers who brought in
old CDMA handsets to pay only US$30 for activation and the first month of service,
compared to what is typically a US$100 handset cost for new customers. This
business, however, essentially died when MetroPCS moved to eliminate free-firstmonth pricing and switched to discount handsets instead. Verizon Wireless also

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

started allowing customers to bring in their own phones to get postpaid-level pricing
on service on a month-to-month basis, with the option to drop off the service
whenever they like, but the company rarely mentions the program. Finally, Google
launched the Nexus One in January 2010, selling the phone exclusively through its
website with no attached subsidy. While the Nexus One generated significant
excitement, the high price ($599) and limited carrier usefulness resulted in few actual
sales.
The bottom line is that as long as carriers will subsidize handsets in the US by $50400 every 12-24 months, consumers have little reason to pay full price for a device.
We dont expect this to change any time soon, and even as wireless technologies
converge, the different spectrum bands for US carriers likely will keep handsets
mostly tied to a single carrier.

iPhone Exclusivity to End in February


In 2007 AT&T was the carrier launch partner for Apple (AAPL, rated Overweight by
J.P. Morgan IT Hardware analyst, Mark Moskowitz) to bring the iPhone to market in
the US. That deal was exclusive to AT&T in the US for the last few years, even as
once-exclusive countries in Western Europe went non-exclusive. Now, AT&Ts
exclusivity is ending, as Verizon begins offering the phone in early February, and
possibly other carriers later in the year.
AT&T ended 2010 with nearly
20 million iPhone customers

How big is iPhone at AT&T today?


AT&T today has roughly 20 million iPhone customers in its base, who we estimate
spend ~$90 per month, churn at about two-thirds of the postpaid average, and use an
average of around 350 MB per month on the 3G network vs. 200 MB on average. We
estimate that, after selling roughly 14 million iPhones in 2010 and another 10 million
in 2009, AT&T will have nearly 20 million iPhone users at the end of 2010, of which
more than 90% are on contract.

Table 9: AT&T iPhone Sales and Subscribers Analysis


Apple reported iPhone sales
% sold through AT&T

1Q08
1,703
53.7%

2Q08
717
118.5%

3Q08
6,892
34.8%

4Q08
4,363
43.5%

2008
13,675
44.4%

1Q09
3,793
42.2%

2Q09
5,208
46.1%

3Q09
7,367
43.4%

4Q09
8,737
35.5%

2009
25,105
41.0%

1Q10
8,752
30.9%

2Q10
8,398
38.1%

3Q10
14,102
36.9%

4Q10E
9,403
42.5%

2010E
40,655
37.1%

1Q11E
8,656
35.8%

2Q11E
8,783
38.7%

3Q11E
9,970
42.1%

4Q11E
10,748
32.6%

iPhone handset sales


% of AT&T handset sales
% new to AT&T
iPhone as % of gross adds

915
10.5%
40.0%
13.2%

850
9.7%
40.0%
12.6%

2,400
22.4%
40.0%
25.8%

1,900
18.7%
40.0%
22.6%

6,065
15.8%
40.0%
19.3%

1,600
18.2%
40.0%
21.5%

2,400
26.0%
33.0%
25.6%

3,200
32.1%
40.0%
37.0%
37%

3,100
31.8%
33.0%
33.6%

10,300
27.3%
36.3%
29.7%

2,700
32.1%
35.0%
36.4%

3,200
35.0%
27.0%
34.6%

4,000
39.3%
25.0%
34.9%

3,400
38.1%
25.0%
33.3%

4,200
39.5%
25.0%
38.2%

3,500
34.1%
25.0%
31.8%

265
848
6.4%

265
1,060
7.7%

15,100
38.7%
26.9%
36.9%
4,057.00
265
4,002
7.5%

3,100
34.4%
25.0%
30.3%

265
716
5.6%

5,200
46.2%
24.0%
40.9%
36.5%
265
1,378
10.1%

265
822
5.9%

265
901
6.4%

265
1,113
7.8%

265
928
6.5%

iPhone subsidy per device


Total iPhone subsidies
Margin impact from iPhone subsidies
0.25%
(16)
1.27%

0.25%
(22)
1.14%

0.25%
(27)
1.30%

0.25%
(43)
1.30%

0.2%
(108)
1.25%

0.70%
(148)
1.22%

0.70%
(165)
1.14%

0.70%
(197)
1.24%

0.70%
(242)
1.27%

0.6%
(752)
1.22%

0.70%
(281)
1.18%

0.70%
(298)
1.11%

0.70%
(327)
1.31%

0.70%
(386)
1.34%

0.6%
(1,291)
1.24%

1.05%
(623)
1.40%

1.40%
(835)
1.40%

1.05%
(632)
1.35%

1.00%
(628)
1.29%

iPhone replacement rate


iPhone replacements

1.0%
(62)

1.0%
(87)

3.0%
(329)

3.0%
(513)

1.8%
(991)

3.0%
(634)

3.0%
(707)

3.0%
(845)

3.0%
(1,039)

2.6%
(3,225)

4.0%
(1,604)

3.5%
(1,489)

4.5%
(2,105)

4.0%
(2,203)

3.7%
(7,401)

4.0%
(2,373)

4.0%
(2,385)

4.5%
(2,708)

4.5%
(2,824)

iPhone subs
iPhone additions
% of postpaid on iPhone
Average iPhone subs

2,912
837
5.2%
2,493

3,653
741
6.4%
3,282

5,696
2,044
9.8%
4,674

7,041
1,345
11.8%
6,369

7,041
4,966
11.8%
4,558

7,859
818
13.0%
7,450

9,387
1,528
15.2%
8,623

11,545
2,158
18.3%
10,466

13,364
1,818
20.7%
12,454

13,364
6,323
20.7%
10,202

14,179
816
21.8%
13,771

15,593
1,413
23.3%
14,886

18,360
2,768
27.1%
16,976

19,771
1,411
29.0%
19,066

19,771
6,408
29.0%
16,568

19,876
105
29.2%
19,824

20,056
180
29.6%
19,966

20,917
861
30.8%
20,486

20,966
49
30.7%
20,941

0%

0%

40%
2,279

70%
4,929

80%
6,288

85%
7,979

86%
9,929

88%
11,760

11,760

90%
12,761

92%
14,345

95%
17,442

96%
18,981

18,981

97%
19,280

98%
19,655

100%
20,917

100%
20,966

$90.00
1,262
11.2%
$56.56

$90.00
1,720
14.9%
$56.31

$90.00
2,012
17.3%
$55.24

$90.00
2,328
19.5%
$55.72

$90.00
2,826
22.8%
$55.88

$90.00
3,363
26.7%
$54.62

iPhone churn rate


iPhone disconnects
non-iPhone postpaid churn rat

iPhone 3G mix
3G+ iPhones
iPhone ARPU
iPhone service revenue
% of total service revenue
non-iPhone postpaid ARPU

$90.00
673
6.3%
$56.76

$90.00
886
8.1%
$57.23

$90.00
4,922
11.1%
$56.46

$90.00
11,018
22.7%
$54.83

Source: Company reports and J.P. Morgan estimates.

31

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 15: AT&T Is Also Important to Apple


16,000

140%

14,000

120%

12,000

100%

10,000

80%

8,000

60%

6,000
4,000

40%

2,000

20%
0%

1Q
08
2Q
08
3Q
08
4Q
08
1Q
09
2Q
09
3Q
09
4Q
09
1Q
10
2Q
10
3Q
10
4Q
10
E
1Q
11
E
2Q
11
E
3Q
11
E
4Q
11
E

Apple iPhone Sales

AT&T iPhone Sales

AT&T share of total

Source: Company reports and J.P. Morgan estimates.

iPhones generate highest revenue at AT&T, and heaviest data users


At a roughly $90 ARPU and low churn, these 20 million customers (~29% of
postpaid customers) contribute 37% of AT&Ts wireless service revenue, by our
estimates.
We also estimate that the gross margin is as high or higher on iPhone customers as
on typical non-iPhone postpaid subscribers. While the extra ~150 MB of data is a
burden, we dont estimate it costs anywhere near the $30 ARPU premium for the
network to support, and think our iPhone margin numbers below look conservative.
By comparison, AT&T is willing to sell laptop usage to customers at $60 per month
for an estimated average of 1 GB of usage ($0.06/MB) or a maximum of 5 GB
($0.012/MB).
Table 10: AT&T Subscriber Metrics
Avg. iPhone Sub-Exclusive
$90
$550
$36
60%

Avg. Postpaid Sub


$60
$300
$24
60%

Non-iPhone Postpaid Sub


$55
$258
$22
60%

Churn
Lifetime (months)
Lifetime (years)

0.55%
181.8
15.2

1.10%
90.9
7.6

1.19%
83.8
7.0

Recurring subsidy
Upgrades (18 months)
Upgrade cost per month

$300
9.1
$15.02

$100
4.1
$4.46

$66
3.7
$2.88

$250

$200

$192

10.0%
$3,185

10.0%
$1,734

10.0%
$1,575

ARPU
CPGA
CCPU
Pre-marketing margin

Marketing cost per gross add


Discount rate
Customer NPV

Source: Company reports and J.P. Morgan estimates.

iPhones generate ~45% of data usage on AT&Ts wireless network


Our checks indicate the typical iPhone customer uses ~350 MB of data each month
on average, roughly 1.6x that of a typical non-iPhone smartphone user, and the
iPhone base contributes roughly 47% of total wireless network data. While other
carriers have not had to deal with this level of usage or growth, AT&T also doesnt
have as large a number of heavy laptop users as Verizon.

32

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 11: Data Usage at AT&T


Postpaid data usage (MB)
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Average postpaid sub (MB)
Total monthly subscriber data demand (GB)
% change y/y
iPad analysis (in prepaid business)
iPad device net adds
iPad embedded base
iPad data usage (MB)
Total monthly iPad data usage on network
Total network data usage/month (GB)
% change y/y
iPhone as % of data usage

1Q09
300
150
50
1,000
20
89
5,411

5,411

2Q09
300
175
50
1,000
20
104
6,401

3Q09
300
200
50
1,000
20
122
7,690

4Q09
325
200
50
1,000
20
139
8,997

6,401

7,690

8,997

37%

38%

43%

2Q09
150
100
50
1,600
20
125
9,799

3Q09
200
100
50
1,700
20
130
10,310

4Q09
250
150
50
1,800
20
150
12,049

1Q10
325
200
50
1,000
20
149
9,695
79%

2Q10
325
225
50
1,000
20
161
10,796
69%

3Q10
350
225
50
1,000
20
181
12,249
59%

4Q10E
350
275
50
1,000
20
203
13,868
54%

1Q11E
350
300
50
1,000
20
215
14,674
51%

2Q11E
375
325
50
1,000
20
238
16,148
50%

3Q11E
400
350
50
1,000
20
261
17,735
45%

4Q11E
425
375
50
1,000
20
285
19,460
40%

9,695
79%
45%

400
400
600
240
11,036
72%
43%

400
800
600
480
12,729
66%
46%

600
1400
600
840
14,708
63%
45%

350
1750
600
1050
15,724
62%
44%

300
2050
600
1230
17,378
57%
43%

250
2300
600
1380
19,115
50%
42%

300
2600
600
1560
21,020
43%
42%

1Q10
300
175
50
1,900
20
170
13,748
51%

2Q10
325
200
50
2,000
20
189
15,434
57%

3Q10
350
200
50
2,000
20
199
16,334
58%

4Q10E
350
225
50
2,100
20
223
18,463
53%

1Q11E
350
225
50
2,200
20
244
19,617
43%

2Q11E
400
250
50
2,300
20
282
21,438
39%

3Q11E
400
275
50
2,400
20
315
23,259
42%

4Q11E
425
300
50
2,500
20
351
25,283
37%

1,900

2,000

2,000

3,500
2,078

3,700
2,128

4,000
2,137

4,200
2,074

4,200
1,915

16,334
58%

40.0%
126
18,337
52%

50.0%
363
19,254
40%

55.0%
733
20,705
34%

60.0%
1,515
21,744
33%

65.0%
2,540
22,744
24%

Source: Company reports and J.P. Morgan estimates.

Table 12: Data Usage at Verizon


Data usage (MB)
3G iPhones
3G Smartphones (ex-iPhone)
3G Multimedia
Internet*
Feature
Average postpaid sub
Total subscriber data demand (GB)
% change y/y

1Q09
100
100
50
1,500
20
118
9,112

LTE internet device MB/mo (on all networks)


3G internet device MB/mo
% of data use on LTE network for LTE devices
LTE Monthly network data usage (GB)
3G monthly network data usage (GB)
% change y/y

9,112

9,799

10,310

12,049

13,748
51%

15,434
57%

Source: Company reports and J.P. Morgan estimates.

Verizon iPhone is here


On January 11, Verizon and Apple announced the availability of a CDMA iPhone.
Eligible existing Verizon customers can pre-order the device on February 3, and
devices will be available and officially activated for new and existing customers on
February 10. The device is essentially the same iPhone 4 available on AT&T today,
yet there is only CDMA functionality. Given the extra size necessary for additional
antenna and power amplifier in the CDMA/WCDMA Bold, Apple may have chosen
small size over global roaming ability for the first iteration. However, Verizon will
be including mobile hotspot capabilities. Data pricing and hotspot pricing were not
announced, but we would not expect major differences with current plan pricing on
other smartphones.
Apple share of handset sales
Apple has received 30-40% of AT&Ts postpaid handsets in the last year, and this
spiked to 48% in 3Q10 on the iPhone 4 as AT&T tried to lock down its base into the
likely loss of exclusivity in 1Q. We expect Apples share at AT&T to retreat to 40%
in 4Q10 and dip back to the mid-30% range in 1H11. Beyond that we expect Apples
share at AT&T to continue in the mid-30% range as AT&T pushes Android and
WP7 devices somewhat harder than in the past. Overall, we expect an 8% reduction
in iPhone sales at AT&T in 2011, and flat total handset sales due to more non-iPhone
integrated devices.
33

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 13: Handset Sales Split at AT&T


Handset sales split
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Postpaid handset sales

1Q09
1,552
2,744
446
104
2,691
7,537

2Q09
2,304
2,665
208
3,105
8,282

3Q09
3,008
3,160
128
159
2,616
9,071

4Q09
2,945
3,114
136
2,460
8,656

1Q10
2,538
2,593
141
2,058
7,330

2Q10
2,944
2,563
118
2,023
7,648

3Q10
4,732
3,146
119
1,851
9,848

4Q10E
3,680
3,323
119
1,524
8,645

1Q11E
2,790
3,577
129
1,175
7,671

2Q11E
3,060
3,573
120
1,239
7,991

3Q11E
3,780
4,039
120
1,260
9,198

4Q11E
3,150
4,116
120
1,162
8,548

iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Postpaid handset sales

1Q09
20.6%
36.4%
5.9%
1.4%
35.7%
100.0%

2Q09
27.8%
32.2%
0.0%
2.5%
37.5%
100.0%

3Q09
33.2%
34.8%
1.4%
1.8%
28.8%
100.0%

4Q09
34.0%
36.0%
0.0%
1.6%
28.4%
100.0%

1Q10
34.6%
35.4%
0.0%
1.9%
28.1%
100.0%

2Q10
38.5%
33.5%
0.0%
1.5%
26.5%
100.0%

3Q10
48.1%
31.9%
0.0%
1.2%
18.8%
100.0%

4Q10E
42.6%
38.4%
0.0%
1.4%
17.6%
100.0%

1Q11E
36.4%
46.6%
0.0%
1.7%
15.3%
100.0%

2Q11E
38.3%
44.7%
0.0%
1.5%
15.5%
100.0%

3Q11E
41.1%
43.9%
0.0%
1.3%
13.7%
100.0%

4Q11E
36.9%
48.1%
0.0%
1.4%
13.6%
100.0%

36.1%

46.4%

48.8%

48.6%

49.5%

53.5%

60.1%

52.6%

43.8%

46.1%

48.3%

43.4%

iPhone as % of integrated sales

Source: Company reports and J.P. Morgan estimates.

Verizon has a more developed


non-iPhone smartphone lineup
than AT&T

We expect Verizon to sell 12.8 million iPhones in 2011. We expect Apples share
at Verizon to jump to 20% in 1Q10 (even assuming sales in only ~1/2 of the quarter)
and peak at 30% in 2Q10 assuming supplies are full. This would put Apple at 36% of
estimated 1Q smartphone sales at Verizon and 45% in the 2Q-3Q time frame before
backing off. While not as high a share as Apple gets at AT&T, Verizon has a more
established non-Apple offering in the Droid franchise and wed expect Blackberry to
keep a high-single-digit percentage of sales as well. The companys smartphone
share of sales has been ramping and was 43% in 3Q10 we expect it to pop to 66%
once the iPhone is available. (We estimate that >70% of AT&Ts handsets sold in
3Q10 were integrated devices.)

Table 14: Handset Sales Split at Verizon


Handset sales split
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet
Feature phone
Postpaid handset sales

1Q09
2,520
291
485
6,398
9,693

2Q09
1,936
1,568
553
5,164
9,221

3Q09
2,359
3,077
615
4,205
10,257

4Q09
3,054
3,461
611
3,054
10,178

1Q10
3,562
2,375
693
3,265
9,895

2Q10
4,096
1,741
614
3,789
10,239

3Q10
4,731
1,430
770
4,071
11,002

Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet
Feature phone
Postpaid handset sales

0.0%
26.0%
3.0%
5.0%
66.0%
100.0%

0.0%
21.0%
17.0%
6.0%
56.0%
100.0%

0.0%
23.0%
30.0%
6.0%
41.0%
100.0%

0.0%
30.0%
34.0%
6.0%
30.0%
100.0%

0.0%
36.0%
24.0%
7.0%
33.0%
100.0%

0.0%
40.0%
17.0%
6.0%
37.0%
100.0%

0.0%
43.0%
13.0%
7.0%
37.0%
100.0%

4Q10E
5,513
1,607
899
2,469
10,507

0.0%
52.5%
15.3%
8.6%
23.5%
99.8%

1Q11E
2,283
4,122
31
980
3,850
11,414

2Q11E
3,710
4,443
424
1,019
2,626
12,368

3Q11E
3,446
4,381
(197)
1,030
2,656
11,486

4Q11E
3,306
5,003
(486)
1,051
1,976
11,020

20.0%
36.1%
0.3%
8.6%
33.7%
100.0%

30.0%
35.9%
3.4%
8.2%
21.2%
100.0%

30.0%
38.1%
-1.7%
9.0%
23.1%
100.0%

30.0%
45.4%
-4.4%
9.5%
17.9%
100.0%

Source: Company reports and J.P. Morgan estimates.

Potential for iPhone to be offered at other carriers?


In the past, once a device has moved beyond exclusivity at one carrier in a country,
most carriers in a market have typically also offered the device within a year. There
have been numerous industry reports speculating that the iPhone could be offered by
Sprint and other CDMA carriers or T-Mobile. Apple indicated that it has a multiyear, non-exclusive deal in place with Verizon, enabling other carrier to offer the
device.
From a technological perspective, Sprint could use the exact same iPhone as Verizon
for 2011. However, at T-Mobile USA the device would need an AWS-band radio to
access its 3G network. While Verizon may have a few months before other CDMA
34

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

carriers get the device, we would consider this to be a function of natural migration
of a handset. In our view, it would make sense for Apple to be sure it can fill
inventory at Verizon and satisfy demand before releasing the device to other carriers.
Finally, J.P. Morgans handset supply chain analysts in Asia estimate Apple could
build as many as 20 million CDMA iPhones in 2011 vs. the 12-13 million that we
estimate Verizon could sell and the additional one million it might need to hold in
inventory. If accurate, build volumes this high could indicate CDMA distribution
could potentially be wider than just Verizon.
Gross add share likely to shift from AT&T
Since launching the iPhone in 2007, AT&Ts share of industry postpaid gross
additions has risen from the mid-20% range to the low 30s and AT&T and Verizon
have separated themselves from the pack as the largest carriers and the destination
for the highest-quality customers. With iPhone exclusivity ending, however, the big
question is then how much share of gross additions AT&T could lose, and how much
its churn could rise.
Figure 16: Postpaid Gross Add Share Among Big 4

Figure 17: Postpaid Net Add Share Among Big 4

50%
40%

150%

30%

50%

20%

-50%

10%

-150%

0%
1Q07

4Q07

3Q08
T

2Q09
S

Source: Company reports and J.P. Morgan estimates.

1Q10

4Q10E

TM-USA

3Q11E
VZ

-250%
1Q07

4Q07
T

3Q08

2Q09
S

1Q10

4Q10E

TM-USA

3Q11E
VZ

Source: Company reports and J.P. Morgan estimates.

We calculate that in the last 12 months the iPhone accounted for 36.5% of AT&Ts
gross additions, representing 11.2% share of industry postpaid additions. In our base
case we estimate that roughly one in four of those iPhone customers new to AT&T
came exclusively for the device, and will take AT&Ts share of industry gross adds
to 27.4% in 2Q11 (from 30.7% in 3Q10) the first full quarter of sales at Verizon.
This should rebound somewhat to over 29% in 3Q11 as AT&T gets an updated
iPhone, and settle at 28-29% in 2012 and beyond.
Verizon should pick up some share. We believe that Verizon will initially take
some gross add share from AT&T and from Sprint and T-Mobile as well. While
Sprint and T-Mobile USA customers generally spend less than they would have to at
Verizon, some would like an iPhone but havent switched to AT&T due to networkquality concerns. This expected share gain at Verizon isnt as evident in the figure
above, however, because of Sprints forecast steady grind higher in gross addition
share which would otherwise cut more into Verizons share. We expect any Sprint
and T-Mobile gross add share loss to be short-lived.
What if Sprint were also to carry iPhone? Note that if Sprint were also to carry the
iPhone, say roughly in mid-summer, we anticipate it could pick up perhaps as much
35

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

~100 basis points of share against AT&T and perhaps a little more against T-Mobile,
and return to its 3Q10 share against VZ.
Churn shifts could be more significant than gross adds
At AT&T, iPhone contract levels are high, but so is dissatisfaction. AT&T in the
last year has made an aggressive push to get its iPhone customers on contract. In
May 2010 AT&T raised its contract termination fee from $200 to $325, and when the
iPhone 4G came out last summer the company offered a very aggressive policy to
allow 3G and 3GS customers to upgrade early in exchange for a new contract. The
company recently dropped the price of the iPhone 3GS to $49 from $99, which could
be a last effort to saturate the iPhone demand while it has exclusivity. While this
large contract percentage should insulate AT&T somewhat from competitive iPhone
offerings, historical customer dissatisfaction and record-low Consumer Reports
ratings recently have us concerned that some customers may be willing to drop their
contract and switch carriers.
On average we estimate that over 80% of AT&Ts postpaid customers are on
contract, and that this over-indexes to iPhone subs given the companys aggressive
push to upgrade in the last year. By contrast, a larger percentage of off-contract
customers are likely to still be using voice and text-based feature phones and likely
arent interested in new services. After considering the number of customers who are
with an enterprise and cant churn (~66%) or are on a family plan (~70%) with
someone on a contract (or enterprise), we believe the immediate churn potential is far
less than the 20% of customers off contract.
We estimate >90% of iPhone
customers are on contract

For iPhone subscribers, if we assume that 92% are on contract, that implies there are
roughly 1.5 million off-contract iPhone subs, likely split between two buckets, plus a
third bucket of on-contract iPhone subs.
Those most at risk signed with AT&T more than two years ago and are likely still
carrying an iPhone 3G despite an ability to get a subsidized upgrade to an
iPhone 3GS or iPhone 4 for as little as $49 in exchange for a new contract. Their
unwillingness to sign a new contract with AT&T probably indicates they are
dissatisfied with the service, and are likely to churn quickly once Verizon offers
the device. We estimate such subscribers could represent more than half of the
~1.5 million off-contract iPhone subscribers.
The second group, possibly less at risk, comprises customers who have activated
an older iPhone that was either passed on from a friend or bought through a
channel like eBay. These retread handsets accounted for 9% of the 5.2 million
iPhones activated on the AT&T network in 3Q10 and we estimate totaled one
million activations in the last year. While some of these customers are probably
on contract (from a subsidy on a different device) and just using the phone until
they qualify for a subsidy, we estimate that half are off contract and prefer not to
pay even the subsidized iPhone price.
Finally, iPhone customers on contract would need to pay a disconnect fee,
something between $200 and $325 depending on when they signed the contract,
to switch carriers. While some customers likely will be willing to walk away
from their contract and pay the disconnect fee (or perhaps think they can get
away without paying), the size of disconnect fee plus the prospect of another
handset fee at Verizon is likely too high for most to stomach and we dont expect
many of these subscribers to switch.

36

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 15: iPhone Retread Analysis 9% of iPhone Activations in 3Q10 Were Retreads
Total retail handset sales
Postpaid handset sales

1Q09
8,769
7,537

2Q09
9,232
8,282

3Q09
9,976
9,071

4Q09
9,735
8,656

1Q10
8,406
7,330

2Q10
9,137
7,648

3Q10
11,265
9,848

4Q10E
10,191
8,645

1Q11E
9,015
7,671

2Q11E
8,934
7,991

3Q11E
10,627
9,198

4Q11E
10,276
8,548

iPhone subscribers
iPhone handset activations
retread %
# retreads

7,859
1,600
3%
48

9,387
2,400
4%
96

11,545
3,200
6%
192

13,364
3,100
5%
155

14,179
2,700
6%
162

15,593
3,200
8%
256

18,360
5,200
9%
468

19,771
4,000
8%
320

19,876
3,100
10%
310

20,056
3,400
10%
340

20,917
4,200
10%
420

20,966
3,500
10%
350

Source: Company reports and J.P. Morgan estimates.

Verizon churn to AT&T could drop marginally. AT&T garners 44% of nonVerizon gross additions in the postpaid space. Assuming that every postpaid
disconnect at Verizon ends up as a postpaid gross add someplace else, customers
churning from Verizon contribute ~40% of non-Verizon industry gross additions, or
~17% of AT&Ts gross addition traffic. One has to assume that many of these
customers are leaving Verizon the highest-quality network in the US with prices
similar to AT&Ts to get the iPhone.
Sprint, T-Mobile USA could see some churn, but we dont expect much. Some
Sprint and T-Mobile customers may have bee on the fence about going to Verizon,
and were waiting for the company to get the iPhone before moving. However, we
dont think this is a significant number given Verizons higher prices, but such
subscribers could impact churn at Sprint and T-Mobile USA in the first half of 2011.
Note that Sprint continues to have a smaller percentage of on-contract customers than
AT&T or Verizon, so Sprint customers that did want to switch carriers likely would
have a better chance of being able to leave immediately.
Net-net, we expect AT&T to lose subs in 1H11 and add only 100k in 2011
By our model, between fewer gross adds and higher churn, AT&Ts postpaid
business should lose subscribers in 1Q and 2Q 2011, and then get back to even by the
end of the year resulting in 0.15% subscriber growth in 2011, after 6% growth in
2010. Key variables will include not just the timing of Verizons iPhone but its
pricing, any promotions, and AT&Ts reaction.
Table 16: AT&T Postpaid Overview
Postpaid
Beginning subscribers
+ Gross additional subscribers
- Disconnects
+ Acquisition subscribers
Ending subscribers
% change, y-t-y
% of Total subs

1Q10
64,627
2,594
(2,082)
0
65,108
8%
74.8%

2Q10
65,108
2,497
(2,001)
1,366
66,970
9%
74.3%

3Q10
66,970
3,048
(2,303)
0
67,688
8%
73.0%

4Q10E
67,688
2,869
(2,344)
0
68,213
6%
71.6%

2010E
64,627
11,008
(8,730)
1,366
68,213
6%
71.6%

1Q11E
68,213
2,558
(2,658)
0
68,113
5%
70.5%

2Q11E
68,113
2,554
(2,854)
0
67,813
1%
69.8%

3Q11E
67,813
2,747
(2,547)
0
68,013
0%
69.1%

4Q11E
68,013
2,754
(2,454)
0
68,313
0%
68.1%

2011E
68,213
10,613
(10,513)
0
68,313
0%
68.1%

Average subs

64,868

66,039

67,329

67,951

66,420

68,163

67,963

67,913

68,163

68,263

Churn rate, monthly

1.07%

1.01%

1.14%

1.15%

1.10%

1.30%

1.40%

1.25%

1.20%

1.28%

Net additional subscribers


% change, y-t-y
% of total net adds

512
-43%
27.6%

496
-56%
31.8%

745
-44%
28.3%

525
-38%
20.4%

2,278
-46%
26.4%

(100)
NM
NM

(300)
NM
NM

200
-73%
16.3%

300
-43%
15.8%

100
-96%
2.0%

Source: Company reports and J.P. Morgan estimates.

Upgrade rates likely to pick up


overall

We expect Verizons upgrade rate to ramp up for two quarters. The company
was already running at a 9.5% upgrade rate in 3Q10 very high by historical levels,
but we expect that to pop to 10% in 1Q11 and 11% in 2Q11 as customers shift over
to iPhone. This will depend somewhat on how willing Verizon is to allow customers
on contract to upgrade. The company doesnt appear to be subsidizing quick
upgrades, but may be willing to in certain situations depending on the competitive
situation. In general, we dont expect Verizon to allow upgrades nearly as early as
AT&T did recently with the iPhone 4 launch.
37

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

We expect AT&Ts upgrade rate to drop back to the 1H10 level, but not much
below. AT&T should continue its aggressive push to upgrade customers, and is
likely to offer good discounts on competitive devices to keep customers interested.
We expect to see a marginal shift to other handsets like those with the Windows
Phone 7 operating system, but dont see new AT&T customers abandoning the
iPhone just because AT&T isnt likely to push it as aggressively. We model AT&Ts
quarterly upgrade rate declining to 7.7% on average in the first half of 2011 versus
the very high 10.1% in 3Q10. With a typical iPhone refresh in mid-summer the
upgrade rate and Apple share will likely rebound but not should not be as extreme as
in 3Q10 as AT&T tried to lock down its base.
What will be the impact on carrier financials?
We expect AT&T margins to rebound ~600 basis points from 3Q10 levels.
AT&T should see some benefit from lower subsidies as sales shift to other devices
and the upgrade rate slows. The company should also see some benefit from slower
growth and lower gross adds. Overall, we expect the total subsidy amount that
AT&T pays peaked in 3Q10 at $2.55 billion and will decline to closer to $1.8 billion
per quarter in 1H11, for a pickup of about 500-600 basis points in EBITDA margin.
Given AT&Ts large customer base and heavily recurring revenue nature, our EPS
estimate for AT&T is not drastically impacted by more or fewer customer additions.

Table 17: AT&T Handset Analysis


iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Postpaid handset sales

1Q09
1,552
2,744
435
104
2,702
7,537

2Q09
2,304
2,665
208
3,105
8,282

3Q09
3,008
3,160
138
159
2,606
9,071

4Q09
2,945
3,114
136
2,460
8,656

1Q10
2,538
2,593
141
2,058
7,330

2Q10
2,944
2,563
118
2,023
7,648

3Q10
4,732
3,146
119
1,851
9,848

4Q10E
3,680
3,323
119
1,524
8,645

1Q11E
2,790
3,577
129
1,175
7,671

2Q11E
3,060
3,573
120
1,239
7,991

3Q11E
3,780
4,039
120
1,260
9,198

4Q11E
3,150
4,116
120
1,162
8,548

Handset sales split


iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Postpaid handset sales

20.6%
36.4%
5.8%
1.4%
35.8%
100.0%

27.8%
32.2%
0.0%
2.5%
37.5%
100.0%

33.2%
34.8%
1.5%
1.8%
28.7%
100.0%

34.0%
36.0%
0.0%
1.6%
28.4%
100.0%

34.6%
35.4%
0.0%
1.9%
28.1%
100.0%

38.5%
33.5%
0.0%
1.5%
26.5%
100.0%

48.1%
31.9%
0.0%
1.2%
18.8%
100.0%

42.6%
38.4%
0.0%
1.4%
17.6%
100.0%

36.4%
46.6%
0.0%
1.7%
15.3%
100.0%

38.3%
44.7%
0.0%
1.5%
15.5%
100.0%

41.1%
43.9%
0.0%
1.3%
13.7%
100.0%

36.9%
48.1%
0.0%
1.4%
13.6%
100.0%

iPhone as % of integrated sales

36.1%

46.4%

48.8%

48.6%

49.5%

53.5%

60.1%

52.6%

43.8%

46.1%

48.3%

43.4%

Subsidy analysis
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Total postpaid subsidy

$350
$250
$100
$100
$58
$191

$350
$180
$100
$100
$65
$182

$325
$175
$100
$100
$68
$192

$325
$125
$100
$100
$54
$173

$325
$135
$100
$100
$57
$178

$325
$160
$100
$100
$61
$196

$350
$250
$100
$100
$52
$259

$325
$225
$100
$100
$54
$236

$325
$200
$100
$100
$143
$235

$325
$200
$100
$100
$53
$224

$325
$200
$100
$100
$61
$231

$325
$200
$100
$100
$81
$228

Aggregate handset subsidy expense (million


% of service revenue

1,439
12.4%

1,507
12.6%

1,738
14.0%

1,494
11.9%

1,307
10.2%

1,502
11.4%

2,550
18.7%

2,038
14.8%

1,803
13.0%

1,787
12.7%

2,125
14.9%

1,952
13.8%

Source: Company reports and J.P. Morgan estimates.

38

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

We expect higher subsidies to cut ~500 basis points from Verizons margin.
Verizon should see some commensurate pressure on margins from faster growth and
upgrades, as well as ramping smartphone share and higher subsidies on iPhones. We
estimate that the companys gross subsidy expense will increase to $2.5 billion in
2Q11 from $1.7 billion per quarter in 2H10. While this is somewhat due to iPhone,
the companys shift to more expensive smartphones is equally to blame. In 2010
Verizon was able to maintain margins by cutting other costs to offset higher subsidy
expense, and our estimate of a 500 basis point hit could be conservative.
Table 18: Verizon Handset Analysis
Handset sales split
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet
Feature phone
Postpaid handset sales

1Q09
2,520
291
485
6,398
9,693

2Q09
1,936
1,568
553
5,164
9,221

3Q09
2,359
3,077
615
4,205
10,257

4Q09
3,054
3,461
611
3,054
10,178

1Q10
3,562
2,375
693
3,265
9,895

2Q10
4,096
1,741
614
3,789
10,239

3Q10
4,731
1,430
770
4,071
11,002

Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet
Feature phone
Postpaid handset sales

0.0%
26.0%
3.0%
5.0%
66.0%
100.0%

0.0%
21.0%
17.0%
6.0%
56.0%
100.0%

0.0%
23.0%
30.0%
6.0%
41.0%
100.0%

0.0%
30.0%
34.0%
6.0%
30.0%
100.0%

0.0%
36.0%
24.0%
7.0%
33.0%
100.0%

0.0%
40.0%
17.0%
6.0%
37.0%
100.0%

0.0%
43.0%
13.0%
7.0%
37.0%
100.0%

iPhone as % of smartphone sales

4Q10E
5,513
1,607
899
2,469
10,507

1Q11E
2,283
4,122
31
980
3,850
11,414

2Q11E
3,710
4,443
424
1,019
2,626
12,368

3Q11E
3,446
4,381
(197)
1,030
2,656
11,486

4Q11E
3,306
5,003
(486)
1,051
1,976
11,020

0.0%
52.5%
15.3%
8.6%
23.5%
99.8%

20.0%
36.1%
0.3%
8.6%
33.7%
100.0%

30.0%
35.9%
3.4%
8.2%
21.2%
100.0%

30.0%
38.1%
-1.7%
9.0%
23.1%
100.0%

30.0%
45.4%
-4.4%
9.5%
17.9%
100.0%

0.0%

0.0%

35.6%

45.5%

44.0%

39.8%

iPhone subsidy
Smartphone subsidy (non-iPhone)
3G Multimedia subsidy
Internet subsidy
Feature phone subsidy
Total postpaid subsidy

1Q09
$350
$250
$100
$100
$147
$170

2Q09
$350
$250
$100
$100
$194
$184

3Q09
$350
$250
$100
$100
$200
$176

4Q09
$350
$250
$100
$100
$242
$187

1Q10
$350
$250
$100
$100
$140
$167

2Q10
$350
$250
$100
$100
$131
$171

3Q10
$350
$250
$125
$125
$69
$158

4Q10E
$350
$230
$110
$125
$54
$161

1Q11E
$350
$225
$100
$120
$44
$177

2Q11E
$350
$200
$100
$120
$51
$201

3Q11E
$350
$200
$100
$120
$56
$203

4Q11E
$350
$180
$100
$100
$51
$201

Aggregate handset subsidy expense (million


% of service revenue

1,647
12.9%

1,696
13.1%

1,802
13.7%

1,908
14.5%

1,654
12.3%

1,756
12.7%

1,740
12.3%

1,690
11.9%

2,017
14.1%

2,485
17.0%

2,335
15.6%

2,215
14.8%

Source: Company reports and J.P. Morgan estimates.

Sprint and T-Mobile should not see any material change in financials. If Sprint were
to get the iPhone at some point mid-year, its upgrade rates and subsidies would not
be expected to increase substantially since they are already very high.
New HSPA+ iPhone in mid-2011 could mitigate AT&Ts pain
We expect Apple to introduce the next version of the GSM/HSPA iPhone on its
typical upgrade schedule of mid-summer. We would not expect to see an updated
CDMA iPhone at that time rather we would expect Apple to likely stagger versions
with Verizon possibly getting a CDMA upgrade (with a slim possibility of including
LTE) in early 2012. The LTE question likely will be a big topic of debate in 2H11.
The latest iPhone is already HSPA 7.2, and an upgrade could move to HSPA version
14.4 or even 21, for significantly higher speeds than Verizons CDMA network can
support. If AT&Ts iPhone offering had a great new feature set that isnt available on
the CDMA version, it could mitigate somewhat AT&Ts subscriber loss in 2H10, but
also may keep more of a lid on margins than we currently model. We would not
expect an LTE-enabled iPhone in 2011 since the global ecosystem is very new, but
think one could be introduced as early as 2012.

39

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Smartphones Are Driving Network Data Usage Higher


We expect Verizon to have the capacity to support the iPhone
We estimate that the average AT&T iPhone uses ~350 MB/month on the 3G network
vs. ~400 MB for Android devices and ~50-100 MB for Blackberrys. Given Verizons
rapidly increasing smartphone adoption rate in the last year which started with the
Droid, we believe that Verizon is prepared for the demands that will be created by a
shift to iPhone as well as continued ramp in smartphone penetration in general.
The key to Verizons performance will be capacity on its CDMA network since we
do not expect an LTE iPhone in 2011. While the company will offer some LTEenabled smartphones in mid-2011 (these are dual-chip devices the single-chip
CDMA/LTE devices wont be available until 2012), we dont expect them to be
significant enough sellers to reduce demand on the CDMA network. What could be
important, however, is the upcoming shift in datacards at Verizon from CDMA to
LTE-enabled. While Android devices typically use ~400 MB/month or higher on
average, datacard usage at most carriers (not including Clearwire) has averaged 12 GB/month, so getting that demand off the CDMA network and on to LTE would
create significant capacity. Verizon has 5.5 million datacards in its base each using
an average of more than 2 GB per month a substantial amount of use such that the
iPhone likely wont be a big incremental burden.

40

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 19: Verizon 3G Network Capacity Demand Grew >50% TTM


Subscribers
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet*
Feature
Total postpaid base
Data usage (MB)
3G iPhones
3G Smartphones (ex-iPhone)
3G Multimedia
Internet*
Feature
Average postpaid sub
Total subscriber data demand (GB)
% change y/y

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10E

8,664
4,641
64,204
77,354

9,407
1,568
4,704
62,716
78,395

10,652
5,013
4,595
59,111
79,371

12,345
8,632
4,818
54,700
80,495

14,137
10,828
5,113
50,833
80,912

16,234
12,126
5,311
47,902
81,573

18,681
12,956
5,524
45,096
82,257

1Q09
100
100
50
1,500
20
118
9,112

2Q09
150
100
50
1,600
20
125
9,799

3Q09
200
100
50
1,700
20
130
10,310

4Q09
250
150
50
1,800
20
150
12,049

1Q10
300
175
50
1,900
20
170
13,748
51%

2Q10
325
200
50
2,000
20
189
15,434
57%

1,900

2,000

LTE internet device MB/mo (on all networks)


3G internet device MB/mo
% of data use on LTE network for LTE devices
LTE Monthly network data usage (GB)
3G monthly network data usage (GB)
% change y/y
LTE mix within handset sales
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet*
Feature
Total LTE mix of sales
LTE handset sales
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet*
Feature
Total LTE handset sales
% of postpaid handset sales
LTE subs
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet*
Feature
Total LTE subs
LTE mix of subs
Smartphone (iPhone)
Smartphone (non-iPhone)
3G Multimedia
Internet*
Feature
Total LTE subs

9,112

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

9,799

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

10,310

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

12,049

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

21,129
13,671
5,800
42,257
82,857

1Q11E
2,283
21,934
12,944
6,013
40,334
83,507

2Q11E
5,724
22,924
12,639
6,235
36,736
84,257

3Q11E
8,494
23,714
11,866
6,442
34,242
84,757

4Q11E
10,798
24,968
11,070
6,642
31,678
85,157

3Q10
350
200
50
2,000
20
199
16,334
58%

4Q10E
350
225
50
2,100
20
223
18,463
53%

1Q11E
350
225
50
2,200
20
244
19,617
43%

2Q11E
400
250
50
2,300
20
282
21,438
39%

3Q11E
400
275
50
2,400
20
315
23,259
42%

4Q11E
425
300
50
2,500
20
351
25,283
37%

2,000

3,500
2,078

3,700
2,128

4,000
2,137

4,200
2,074

4,200
1,915

50.0%
363
19,254
40%

55.0%
733
20,705
34%

60.0%
1,515
21,744
33%

65.0%
2,540
22,744
24%

13,748
51%

15,434
57%

16,334
58%

40.0%
126
18,337
52%

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
10.0%
0.0%
0.9%

0.0%
0.0%
0.0%
20.0%
0.0%
1.7%

0.0%
10.0%
0.0%
30.0%
0.0%
6.1%

0.0%
30.0%
0.0%
50.0%
0.0%
15.9%

0.0%
25.0%
0.0%
80.0%
0.0%
19.0%

90
90
0.9%

196
196
1.7%

444
306
750
6.1%

1,314
515
1,830
15.9%

1,251
841
2,092
19.0%

90
90

274
274

444
544
989

1,690
989
2,679

2,681
1,701
4,382

0.0%
0.0%
1.5%
0.0%
0.1%

0.0%
0.0%
0.0%
4.6%
0.0%
0.3%

0.0%
1.9%
0.0%
8.7%
0.0%
1.2%

0.0%
7.1%
0.0%
15.3%
0.0%
3.2%

0.0%
10.7%
0.0%
25.6%
0.0%
5.1%

0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%

0.0%
0.0%
0.0%
0.0%
0.0%

Source: Company reports and J.P. Morgan estimates.

In addition, we believe that Verizon has an advantage with a CDMA iPhone in that
it can only conduct a voice or data session at any one time, not both. AT&T has
indicated that the demand on its network from a simultaneous voice and data session
(talking on the phone and surfing the web at the same time) is far more than just
double the capacity demand, and when a third app kicks in (like a scheduled
automated weather or email update request) the complexity expands exponentially.

41

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

While AT&T has a marketing advantage over Verizon with simultaneous voice and
data, we dont believe that many consumers find it practical.
Verizons 3G network load is
~20% higher than AT&Ts

AT&T network data capacity demand is less than Verizons, even with iPhone
While AT&T has far more smartphone customers than Verizon, it has a much
smaller number of laptop customers, and those customers use only ~1 GB of data per
month vs. ~2 GB at Verizon. The result is that we estimate AT&T supports about 1415 terabytes of data per month vs. more than 18 terabytes per month for Verizon in
4Q10.

Table 20: AT&T Data Usage Trends


Postpaid subscribers
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Total postpaid base
Postpaid data usage (MB)
iPhone
3G Integrated devices (non-iPhone)
2G Integrated devices (non-iPhone)
Laptop cards
Feature
Average postpaid sub (MB)
Total monthly subscriber data demand (GB)
% change y/y
iPad analysis (in prepaid business)
iPad device net adds
iPad embedded base
iPad data usage (MB)
Total monthly iPad data usage on network
Total network data usage/month (GB)
% change y/y
iPhone as % of data usage

1Q09
7,859
5,112
6,218
1,300
40,045
60,535

2Q09
9,387
7,021
5,785
1,400
38,054
61,647

3Q09
11,545
9,371
5,527
1,445
35,072
62,961

4Q09
13,364
11,740
5,142
1,465
32,917
64,627

1Q10
14,179
14,039
4,857
1,488
30,545
65,108

2Q10
15,593
15,355
4,681
1,487
29,855
66,970

3Q10
18,360
16,058
4,367
1,487
27,416
67,688

4Q10E
19,771
17,336
4,161
1,487
25,457
68,213

1Q11E
19,876
18,516
3,838
1,497
24,386
68,113

2Q11E
20,056
20,056
3,288
1,497
22,916
67,813

3Q11E
20,917
20,830
3,142
1,497
21,627
68,013

4Q11E
20,966
22,700
2,787
1,497
20,363
68,313

300
150
50
1,000
20
91
5,536

300
175
50
1,000
20
105
6,495

300
200
50
1,000
20
123
7,761

325
200
50
1,000
20
140
9,072

325
200
50
1,000
20
150
9,758
76%

325
225
50
1,000
20
162
10,841
67%

350
225
50
1,000
20
182
12,293
58%

350
275
50
1,000
20
204
13,892
53%

350
300
50
1,000
20
216
14,688
51%

375
325
50
1,000
20
238
16,159
49%

400
350
50
1,000
20
261
17,744
44%

425
375
50
1,000
20
285
19,467
40%

9,758
76%
46%

400
400
600
240
11,081
71%
44%

400
800
600
480
12,773
65%
47%

600
1400
600
840
14,732
62%
45%

350
1750
600
1050
15,738
61%
44%

300
2050
600
1230
17,389
57%
43%

250
2300
600
1380
19,124
50%
43%

300
2600
600
1560
21,027
43%
42%

5,536

6,495

7,761

9,072

40%

40%

45%

Source: Company reports and J.P. Morgan estimates.

Total network data usage/month (GB)

Figure 18: Network Usage Trends at T and VZ


30,000
25,000
20,000
15,000
10,000
5,000
1Q09 2Q09 3Q09 4Q09 1Q10 2Q10 3Q10 4Q10E 1Q11E 2Q11E 3Q11E 4Q11E
VZ

Source: Company reports and J.P. Morgan estimates.

Expect Continued Move to Capped Data Plans


We expect wireless data packages to increasingly skew toward usage caps and/or
speed throttles for heavy users. In our view, the potential for abuse is just too high,
and customers who abuse the network can use a massive amount of data if left
unchecked.
42

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Note that we dont see carriers trying to reduce usage for the vast majority of
customers, and think the art in caps will be in finding and explaining levels that most
customers realize they will never reach. The worst abusers, however, often are doing
more than just watching video on their cell phones AT&T believes that the worst
abusers have jail-broken their phones and are likely running some kind of
networked application or group server off the device.
AT&Ts top 1% of users each generate ~4 GB/month of traffic
AT&T has said that the top 1% of users account for 20% of data usage, and the top
3% use 40% of capacity. Using our estimate of ~15 terabytes of data traffic per
month in 4Q10 (see above table) implies that the top 1% of customers is each using
~4 GB of data, the next 2% about 2.5 GB, and the other 97% an average of only 115130 MB/month.
Clearwires 7 GB/month average
is >3x greater than that of
Verizons laptop customers

Clearwire has had to add capacity far earlier than expected


While Clearwires original build called for such a dense network that it did not
expect to increase capacity for years, the company has already had to go back and
add capacity in some areas far earlier than we originally expected. This could be
partly due to better customer uptake overall, but we think it is mostly due to the very
high usage the company is seeing upwards of 7 GB/month per customer. While this
represents mostly broadband laptop and desktop customers rather than smartphones,
it is a tremendous amount of data to support on a wireless network.

Emerging Devices to Augment Growth


We consider emerging wireless devices to be everything from wireless-enabled
dog collars to smart utility meters and 3G iPads. From the carriers perspective, the
big categories we see are 1) consumer devices that will be directly billed by carriers
like the iPad or other tablets, 2) consumer devices that have a middleman like
Amazons Kindle, and 3) enterprise devices that are mostly run by third-party firms
like utilities and security firms. While the biggest number of connections are likely to
come from the third category (e.g., thousands of utility meters), these generate very
little data or revenue per connection, and even less for the carrier after a middleman
takes a large slice. Instead, for the carrier the vast majority of revenue is likely to be
generated from consumer devices like tablets and GPS machines which the consumer
pays directly to the carrier.
Can generate better margins depending on the sales model
Wholesale relationships require no upfront subsidies or follow-on customer service
or billing. We estimate margins are likely in the 50-80% range depending on the
business vs. 20-40% for retail wireless.

Consumer Devices Are the Most Important for Revenue


Netbooks, laptop cards, tablets command $15-80/month
We expect the most common consumer data devices to be products like tablets that
are wirelessly enabled, such as the iPad or Kindle, laptop cards/MiFi devices that add
connectivity to one or many externally purchased devices, and GPS dashboard
devices that add wireless connectivity for information, such as traffic reports and
weather.

43

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Among tablets, the most successful so far have been Amazons Kindle (reported as
wholesale) and Apples iPad (reported as prepaid), which are good examples of the
different ways these devices can be booked. At this years CES, though, tablets were
the key device rolled out by almost every handset and PC vendor, so we expect a lot
of fragmentation from here.
Verizon and AT&T both sell iPad, but book subscriber and revenue differently
The 3G iPad, which is sold mostly through consumer electronics retailers but also
through AT&T on an unsubsidized basis, has an option for a $25/month prepaid data
plan that gives the user up to 2 GB/month. Because it is prepaid, the payment is
reported through AT&Ts prepaid subscriber line; we estimate the iPad accounted for
~400,000 net adds in 2Q and 3Q 2010, and we expect higher sales in 4Q10. While
the iPad ARPU is similar to AT&Ts typical prepaid ARPU of ~$28, this is
obviously a very different customer as the device uses no voice but a potentially
large amount of data, but also requires no upfront subsidy or commission from the
carrier.
Verizon charges premium to
bundle iPad and MiFi

Verizon also sells the iPad, but today at least this is a WiFi-only device and comes
bundled with a Verizon 3G MiFi device to give it wireless connectivity if the
customer wishes. Verizon has said it will sell a 3G iPad in the future. The
combination costs $130 more than a comparable iPad alone (vs. free for the MiFi
device alone on the Verizon website) and requires a postpaid contract of $20-80 per
month for 1-10 GB of data. We would not be surprised to see Apple eventually build
a CDMA or LTE-based iPad to sell through Verizon without the MiFi, but to date the
companies have not indicated the potential for such a device.
Table 21: Verizon iPad and MiFi Pricing
Data Cap

Monthly Plan Price

Overage

1GB

$20

$20

3GB

$35

$10

5GB

$50

$10

10GB

$80

$10

Source: Company reports and J.P. Morgan.

iPad has cannibalized Kindle


less than we expected

44

The Kindle is a usage-based wholesale device


In contrast with iPad, the Kindle is a device sold by Amazon with an optional 3G
data connection, in which the 3G relationship is invisible to the customer. Instead,
Amazon pays the carrier a one-time activation fee plus a monthly amount based on
how much data the device consumes. The carrier does not have to manage or bill the
customer, and the customer only pays for the content downloaded not the
transmission. Kindle was first sold with a CDMA connection on Sprints network,
but devices sold in the last year have been WCDMA-enabled on AT&Ts network
instead. Both carriers report revenue from the devices as a wholesale relationship,
AT&T under a special category called Connected Devices. We estimate AT&T gets
$3-4 per month for each Kindle subscriber.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 19: iPad, Galaxy Tab, and Kindle

iPad

Galaxy Tab

Kindle

Source: Company reports (used with permission).

Competition and New Entrants


The wireless industry remains highly competitive, though less so for the best
customers who mostly choose between AT&T and Verizon.
We see the possibility of consolidation among traditional players outside of AT&T
and Verizon, but emphasize that we dont think it is likely any time soon. In our
view, consolidation would likely be a function of higher stock prices as companies
could consider paying substantial premiums with a strong equity currency, rather
than out of weakness. Note that balance sheets in the wireless industry have never
been stronger traditional carriers have little risk of a near-term cash crunch and
most have record cash levels. Maturities have been pushed out and covenants are
almost non-existent.

Emerging Carriers Prospects for Clearwire and


Lightsquared
In our view, other companies considering 4G wireless wholesale deals like T-Mobile
USA, DTV, and DISH could lean toward Clearwire rather than Lightsquared, given
the greater risk of growing pains.
Clearwire has some excellent anchor tenants
We believe that Clearwire is a very important strategic asset for Sprint and only
slightly less so for the cable partners. With Sprint as the anchor tenant, the company
has been adding wholesale customers very quickly and we believe that eventually the
majority of Sprint customers will have 4G capabilities.
We remain wary of Clearwires
retail strategy

On the retail side we remain wary. We think the company will struggle to grow in the
retail business in a saturated broadband and wireless voice market against very strong
incumbents. The planned launch of its own smartphones will likely be a challenge to
Clearwires financials as upfront subsidies and commissions to sales channels
increase in hopes for higher long-term ARPU.
We believe that Clearwire is likely to take additional equity from Sprint and possibly
from additional wholesale partners. Clearwire has also stated that it is the process of
evaluating the sale of some of its spectrum. We would not expect the company to sell
spectrum unless it was the only funding opportunity as this would likely create
additional competition (or reduce wholesale partner potential) in the long term and
possibly increase the cost of adding capacity.

45

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

We expect Lightsquared to struggle


The company as yet does not have an anchor tenant, and the timing to high capacity
coverage could keep any potential big anchor tenant from committing significant
capital.
Financing still a significant hurdle. In 2010 Lightsquared announced two funding
rounds the first in July of up to $1.75 billion in additional debt and equity
financing, and in October a debt round of $850 million. While the debt is backed by
the companys spectrum, the company also has a very aggressive buildout
requirement to keep that spectrum. However, Lightsquared expects to eventually
need a total of ~$5 billion for the full buildout, including $1.75 billion raised in July
2010.
We believe that Lightsquared will eventually need at least ~$5 billion for the full
buildout, including $1.75b raised in July 2010. This is less than we expect Clearwire
to need, but by not pursuing a retail strategy Lightsquared can avoid the expensive
marketing and store build that Clearwire plans. There may be some potential for
vendor financing (Nokia Siemens Networks is building the network), but likely a
small minority and probably not cash.
Lightsquared has access to 59 MHz of spectrum. Lightsquared claims access to
59 MHz of spectrum in the US. This is broken down between 28 MHz of L-Band
(1.6 GHz) bought with SkyTerra, 18 MHz of L-Band leased from Inmarsat, 8 MHz
of 1.4 GHz terrestrial spectrum leased from Terrestar, and 5 MHz of terrestrial
spectrum owned in the 1.6 GHz band. The network will focus on the 46 MHz of
L-Band with 6 MHz allocated to the satellite link and 40 MHz for terrestrial uses.
Buildout requirements are a whip behind the network build. Lightsquareds
build requirements include four cities (Baltimore, Phoenix, Denver, and Las Vegas)
to be launched by 2H11, 100 million pops covered with terrestrial service by 2H12,
and 92% pop coverage by 2015. Nokia Siemens Networks got to work planning the
network after being named the vendor in July, and it appears the build is progressing
with tower leases being signed currently.
Satellite issues have cleared up. Lightsquared launched its first satellite in
November. While it originally had problems opening its arms, this seems to have
been resolved. The company has another satellite that was reserved for capacity that
it could launch in 2011. Each satellite acts like ~300 cell sites, directing beams of
data to different parts of the country, though at lower bit rates than terrestrial LTE.
With 23 meter mirrors, these antennas are the biggest ever deployed on commercial
satellites.

46

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Spectrum Auctions Will Come Eventually


Spectrum is the lifeblood of the wireless industry a finite resource that is hard to
come by and guarded carefully once it is won. In the US there currently is roughly
214 MHz of spectrum allocated to the 2G and 3G technologies, while the 4G
allocation is roughly 300 MHz depending on how one counts it.
Table 22: Wireless Spectrum Summary
Common band name
Cellular
PCS
AWS
700
2.5/2.6
1.4-1.5 GHz
2G/3G total
4G total

Frequency
Bandwidth (MHz) 2G/3G/4G
800/850 MHz
64
2G/3G
1900 MHz
130
2G/3G
1.7/2.1 GHz
90
3G/4G
700 MHz
52
4G
2.5/2.6 GHz
150
4G
1.4/1.5 GHz
59
4G
~214
~300

Comments
50 MHz given to industry in the 80's. 14 MHz owned by Sprint through Nextel spectrum swap
120 MHz auctioned in mid-90's. Includes 10 MHz of G-Block spectrum owned by Sprint not in use.
2006 Auction. T-Mobile owns ~30 MHz for 3G, VZ and T own ~30 MHz for 4G, cable owns ~20 MHz unbuilt
2008 Auction. Mostly owned by AT&T, Verizon, Echostar
Owned by Clearwire, allocated to 4G
Owned by Lightsquared, allocated to 4G

Source: Company reports and J.P. Morgan estimates.

In general, the lower the frequency the farther a signal will travel and be useful
(propagate) at a given power. Thus, lower frequencies like 700 MHz and 800 MHz
bands are better for covering large distances and penetrating buildings than the PCS
or AWS bands. More debatable is the value of high-frequency spectrum when trying
to cover an urban area that needs high cell-site density anyway. How well
Clearwires 4G network loads and its economies over time will be interesting to
watch against Verizons lower-frequency 700 MHz build.
Figure 20: Illustration of Signal Propagation at 800 MHz and 1900 MHz

Source: J.P. Morgan.

47

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Spectrum Spending History Done for Now


Spectrum expense is mostly behind us
Since 2005 the US wireless carriers spent roughly US$38 billion in major sales of
spectrum in the PCS, AWS, and 700 MHz bands. Spectrum spending is almost
always a contra to valuation in our models. While there has been speculation about
more spectrum coming to market, we estimate it could be 2012 or later before
substantial blocks of spectrum come to market.
Table 23: Major Wireless Spectrum License Spending by Carrier, 2005-Present
$ millions

Company
Auction 58
AT&T
184
Verizon Wireless
697
252
T-Mobile USA
SpectrumCo LLC (Cable co's)
Sprint
194
Echostar
Qualcomm*
MetroPCS
Leap Wireless
235
Cox Wireless
US Cellular
152
Cellular South
CenturyLink
1714

AWS
1400
2809
4182
2378

1391
1076
170

13406

700 MHz Aloha (700) QCOM (700)


6637
2500
1925
9363

Total
12646
12869
4434
2378
194
712
558
1704
1311
305
622
192
149
37516

712
558
313
305
300
192
149
18529

2500

1925

* Later sold to AT&T


Source: Company reports and FCC.

Table 24: History of Spectrum Prices in Major Transactions, 2001-2010

AT&T/Qualcomm sale
Auction 73 (700 MHz)
AT&T/Aloha
Auction 66 (AWS)
Auction 58 (PCS)
Auction 35 (PCS, cancelled)

Date Frequency Total $ amount ($m)


2010
700 MHz
1,925
2008
700 MHz
14,827
2007
700 MHz
2,500
2006 1.7/2.1 GHz
13,879
2005 1900 MHz
2,250
2001 1900 MHz
16,300

$/MHz/pop
$0.86
$1.29
$1.06
$0.54
$1.06
$2.11

Source: FCC and J.P. Morgan estimates.

We do not expect a major


spectrum auction for 2-3 years

What Spectrum Bands Are Likely to Come? And When


Could They Be Available?
The national broadband plan was delivered to Congress in March 2010. It
recommends finding 500 MHz of new spectrum for broadband services 300 MHz
in the next five years and another 200 MHz in the five years after that. We think that
finding 300 MHz in the next five years will be a real challenge for the FCC.
700 MHz Auction 92. The FCC has scheduled Auction 92 for July 2011, but the
spectrum for sale represents the remnants of licenses that either didnt sell in the
last 700 MHz auction or were turned back in to the Commission already. Most
are rural and have little impact on the overall spectrum market, by our analysis.
700 MHz D-Block. This 10 MHz block did not hit the minimum bid requirement
in the 2008 auction, and has been in limbo since. There is movement in Congress

48

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

to give this spectrum to public safety instead of auctioning it, so we do not count
on it coming to market.
Lower 700 MHz band. One source of very good spectrum could be to repurpose
more spectrum from the TV broadcasters in the 700 MHz band. Channels 52-69
were the sources for the 2008 700 MHz auction (6 MHz per channel), and
channels 31-51 could yield another 120 MHz. However, most of the big
broadcasters have pushed back against this, and some argue that they could better
use the spectrum for mobile video than could the wireless carriers. Whichever
way this is decided its likely to be a battle, and we dont expect a resolution for
3-5 years.
AWS bands. Another source could be additional AWS spectrum. AWS-2 is two
5 MHz pairs (H-block is 1915-1920/1995-2000 and J-block is 2020-2025/21752180) that are close to Sprints PCS spectrum and werent auctioned in the 2006
AWS-1 auction. AWS-3 is an unpaired 20 MHz that could be paired with lowerband military spectrum to make it more attractive.
Microwave? Additional commercial microwave could also be sourced, pushing
the microwave broadcasters further up the bands. This, however, is not seen as
very attractive spectrum by the carriers but could become more so if it was paired
with something in a lower band.
Satellite spectrum? Thus far the FCC has not talked about re-allocating any
satellite spectrum for terrestrial uses, but given Lightsquareds recent request to
make its system a totally terrestrial application rather than have the satellite
option, the FCC could decide to count Lightsquareds 40 MHz against the
300 MHz. However, given that Lightsquareds previous entities were given that
spectrum rather than having it go through a public auction, we would expect the
wireless carriers to protest against such a giveaway.
Government bands. Finally, there is a lot of spectrum that currently is being
used (or at least allocated to) the Federal government. However, entities like the
FBI, the Department of Defense, and others are likely to lobby heavily against
being moved from their current bands and it would take congressional action to
move most of this into the FCCs domain.

2011 Is the Year of 4G Transitions


Although no one can seem to agree on a definition for 4G, we consider it to be the
next evolution of wireless data speeds from the 100 Kpbs to 1.5 Mbps speeds of 3G,
to a data connection that is steadily above 1 Mbps and can burst to 20 Mbps or
higher. Eventually, 4G will enable both data and voice at faster speeds, lower
latency, and lower cost to support than current 3G technologies.
31 flavors of 4G
There are two main flavors of 4G supported by two different standards bodies. The
Third Generation Partnership Project (3GPP) likely will be the most common and is
constructed of traditional wireless cellular vendors and operators. It has shepherded
the GSM standard through 3G with WCDMA and into what is now known as LTE or
Long Term Evolution. Alternatively, the Institute of Electrical and Electronic
Engineers (IEEE) supports a technology called WiMAX (Worldwide Interoperability
for Microwave Access) which we expect to remain the smaller standard.
49

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

The standards, however, are very similar in that both work best on wide spectrum
channels and utilize the same technologies like Orthogonal Frequency Division
Multiplexing (OFDM), Multiple-In/Multiple-Out (MIMO) antennas, and 64QAM
which work together to deliver higher speeds and an incrementally more-efficient
wireless experience.
Note, however, that some carriers have been claiming 4G speeds even without
having the above technologies. T-Mobile USA, which has a WCDMA/HSPA 14.4
network, claims today to have Americas largest 4G network because its real-world
speeds rival those of Clearwires and Verizons network. While we dont consider
T-Mobiles network to be true 4G, the speeds are close enough that T-Mobile can
likely minimize any marketing disadvantage it may have had, and muddy the waters
for Clearwire and Verizon as they try to differentiate on network speeds.
Our 4G network is capable of theoretical speeds up to 21 Mbps and we have seen
average download speeds approaching five Mbps on our myTouch 4G phone in some
cities with peak speeds of nearly 12 Mbps. Further, independent reviewers have seen
average download speeds on our webConnect Rocket between 5 and 8 Mbps with
peak speeds up to 8-10Mbps. Neville Ray, Chief Technology Officer, T-Mobile
USA
Figure 21: Wireless Technologies Converging Towards LTE
US/
Lat Am

iDen

US/Asia

cdmaOne

Japan

PDC

Global

GSM

US/
Lat Am

TDMA

2001

1XRTT

GPRS

1X EVDO

WiMAX

W-CDMA

HSPA+

Likely evolution

EDGE

2003-2005

2006-2007

2008

LTE

Unlikely evolution

2009

2010+

Source: J.P. Morgan.

What Speeds Can We Expect from 4G?


Since the launch of Verizons LTE network press reports have been flooded with
anecdotal stories of speeds up to 20 Mbps. While these speeds are possible on a
brand-new network that is virtually empty of other users, we expect real-world
mature speeds on Verizons network to be more in the 2-6 Mbps range once the
network is loaded. Similarly, we have experienced speeds on Clearwires 4G
WiMAX network in the range of 1-8 Mbps depending on the time of day and
location, with an average in the 1-3 Mbps range.

50

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Few applications run differently


at 4G vs. 3G speeds

So who uses all this speed?


The debate about wireless speeds for 4G today is somewhat counterproductive, in
our view. For handsets, there are few common applications for which users will
notice a difference between a speed of 1 Mbps and 5 Mbps. And for laptop-toting
customers, Verizons faster 4G speed is a great help in getting customers to their
5GB or 10 GB monthly limit faster, after which extra GBs are available for $10 each.
We think that in the near term the great 4G debate will be won by the carrier with the
most attractive devices an area in which Sprint and Clearwire have some advantage
over Verizon at least to mid-2011, but in which T-Mobile USA may actually be best
off because it can claim 4G speeds but on lower-priced 3G device technology and
scale.

Carrier Upgrade Paths to 4G


The ease of the upgrade path to 4G depends on a carriers current network
technology, the age of the equipment, and its available spectrum. Today more than
80% of carriers worldwide run 3GPP-standard GSM/WCDMA networks, with which
LTE is designed to be backwards-compatible. While some very new WCDMA
networks have the ability to easily upgrade to LTE with minimal changes, most
WCDMA carriers will need additional network hardware at the base station, on the
tower, and at the switch to enable LTE.
Of the remaining 20% of carriers globally, most are on the CDMA standard. This
standard has no easy upgrade path to 4G and will require a full network overlay of
LTE or WiMAX to get to LTE. This is already in the works at most CDMA carriers
MetroPCS and Verizon have launched their LTE networks in some cities and
Sprint continues to work with Clearwire to roll out WiMAX. We would not be
surprised to see Clearwire eventually throw WiMAX over for LTE as well.
Spectrum determines ease of and performance of 4G deployment
4G can be rolled out on as little as 3 MHz of free spectrum, but true 4G speeds
require 10+ MHz. Some carriers like AT&T, Clearwire, and Verizon already have
substantial allocations of empty spectrum that they can use, while T-Mobile and
others will need to re-farm spectrum that is currently in use or buy more in
upcoming auctions.
Table 25: US Carrier 4G Deployment Status
Company
AT&T
Verizon Wireless
T-Mobile USA
SpectrumCo LLC (Cable co's)
Sprint
Clearwire
Echostar
MetroPCS
Leap Wireless
Cox Wireless
US Cellular
Cellular South

Legacy network
WCDMA/HSPA
CDMA
WCDMA/HSPA
NA
CDMA
WiMax
NA
CDMA
CDMA
CDMA
CDMA
CDMA

4G Status
75M pops LTE EOY 2011, Complete 2013
38 live markets
HSPA+ for now
NA
WiMax with CLWR
120 M pops today
NA
LTE deployed today
2011 LTE trials
NA
2011 LTE trials
4Q11 LTE target

4G spectrum plan
30-40 MHz in AWS and 700 bands
40-50 MHz in AWS and 700 bands
Refarm 2G/3G spectrum
20 MHz AWS
Clearwire, refarm 2G/3G spectrum
~150 MHz
~10 MHz of PCS or AWS
Refarm PCS/AWS

Source: Company reports and J.P. Morgan.

51

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Cash Flow Solid Across Wireless Space


Wireless margins are a story of haves, have-nots
Margins in the wireless space have diverged in the last three years, with AT&T and
Verizon margins strengthening while those at Sprint, T-Mobile, and lesser carriers
have in general shrunk. In some cases this is a function of disappearing roaming
revenue (T-Mobile and US Cellular), while Sprint has seen substantial negative
leverage as its customer base shrank.

Cash EBITDA margin

Figure 22: Wireless EBITDA Margin by Carrier 2007 vs. 2010


60%
50%
40%
30%
20%
10%
0%

45% 47%

38% 41%

55% 56%
32%

32% 29%
17%

Verizon

AT&T

Sprint

Wireless

Mobility

Nex tel

T-Mobile

28%

33% 30%
21%

US

nTelos

MetroPCS

Cellular
2007

28%

23%

Leap
Wireless

2010E

Source: Company reports and J.P. Morgan estimates.

Capital intensity has moderated


Capital spending as a percentage of revenue has generally declined in the industry, as
businesses have scaled and technology transitions been less painful.
Figure 23: Wireless Capital Spending by Carrier 2007 vs. 2010E
Capex % of revenue

50%

40%

40%
30%
20%

17% 15%

10%

16%

17%

16% 15%

15% 15%

T-Mobile

US

17%

22%

18%

6%

10%
0%
Verizon

AT&T

Sprint

Wireless

Mobility

Nex tel

Cellular
2007

Source: Company reports and J.P. Morgan estimates.

52

36%

30%

2010E

nTelos

MetroPCS

Leap
Wireless

North America Equity Research


13 January 2011

Wireline Industry

Wireline Industry

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

53

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

54

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireline Key Investment Points


Consumer revenue losses have been well below access line losses, as data
revenue growth provides cushion
Within the consumer revenue stream, telecom companies have been proactive to
offset the secular headwinds facing consumer access lines. Carriers have enhanced
their value offerings by bundling voice with high-speed internet and video. Not only
are bundles accretive to ARPU, as consumers sign up for additional services
providing carriers with a higher share of their telecom/cable spending, but they also
lower churn significantly and help minimize subscriber losses to cable competitors.
Video remains an opportunity
Video rollout through fiber deployment has been a major growth driver for AT&T
and Verizon over the past two years, and both companies have generated significant
traction. Although most of the network buildout phase for both companies is
expected to end in 2011, significant growth opportunity exists as both T and VZ
expect to reach 30%+ video penetration in the regions in which the product is
marketed.
Increasing exposure to business revenues provides leverage to an eventual
macro turnaround
Through acquisitions of CLECs and datacenters serving business customers, as well
as internal initiatives through a marketing focus, some wireline providers have
shifted their business model to be increasingly leveraged to business vs. consumer.
While there has yet to be a meaningful macroeconomic recovery, the business
segments at wireline carriers have started to stabilize as y/y comparisons have
become easier. With unemployment still hovering near 10% and businesses cautious
with their spending, we expect this segment to remain stable over the next few
quarters but grow meaningfully with any macroeconomic rebound.
Ongoing industry consolidation has stabilized EBITDA with synergies driving
savings
The secular shift away from consumer voice, exacerbated by tough macro conditions,
has accelerated the consolidation wave in the wireline industry. Despite ongoing
pressure on the top line, M&A activity has helped to keep margins stable and
allowed companies to maintain their cash flow and dividend levels.
Rurals better on operating metrics, opportunity to drive further penetration
Given a lower level of competition in their operating markets and their local
marketing approach, rural carriers have maintained much better operating metrics,
with access line loss rates as much as 500-600 bps below those of their more urban
counterparts. As the industry consolidation trend within RLECs has expanded to
include more urban assets (CenturyLink buying Qwest following Embarq, Frontier
buying Verizon assets), we see significant opportunity for increased data penetration
levels and carriers to keep access line losses under control. We have seen a good
turnaround in the acquired urban markets in Embarq territories, and would expect
RLECS local engagement approach and increasing broadband reach to help improve
operating metrics in acquired regions in a meaningful way.

55

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireline Stock Recommendations


AT&T Initiating coverage with Overweight rating and $33 YE11 price target
AT&T is well positioned on the wireless side to capture the coming wave of data
devices and currently supports both Apples iPad as well as all of the eReaders that
are sold wirelessly enabled in the US. We expect postpaid wireless subscribers to
grow more slowly than the industry in 2011 due to AT&T losing exclusivity on the
iPhone, but still expect wireless top-line growth of 5.5% and EBITDA growth of
10% as a result of additional margin expansion. In wireline, AT&T has seen slowing
consumer revenue losses as data and video become more important, and stabilization
in enterprise trends which should improve with the economy.
Among the largest telecom players we prefer AT&T over Verizon given its better
cash flow characteristics and far lower EPS multiple, which we believe offset the
risks created by the changes in iPhone exclusivity. Not only should AT&Ts EPS
grow faster in 2011 but the multiple should also expand once the iPhone risk has
been quantified. We model $2.50 in EPS for AT&T in 2011 which represents 9%
growth from estimated 2010 EPS of $2.29.
Verizon Initiating with Neutral rating and $38 YE11 price target
Verizon is also in a strong position in the wireless space, and postpaid subscriber
growth should exceed that of the industry in 2011 due to both its overall network
quality and the addition of the iPhone to its handset lineup. Margins though are likely
to compress somewhat with higher smartphone subsidy costs, so we expect no
wireless EBITDA growth vs. top-line growth of 7%. In wireline, we also see
stabilizing revenue trends as video and data growth should outweigh voice losses,
and we expect some rebound in margins.
While we like Verizons business, the recent move in the stock price to 16.1x 2011E
EPS forces us to recommend avoiding the stock. We would consider coming back to
the stock once the excitement about the iPhone has receded further and if the stock
retreats to more attractive multiples of EPS and free cash flow. We model $2.23 in
2011 EPS, or 4% growth from estimated 2010 EPS of $2.14 (ex SpinCo).
Frontier Initiating with Overweight rating and $11 YE11 price target
We prefer Frontier within our RLEC coverage as we find the 7.8% dividend yield,
the highest within our telecom coverage universe and the highest in the S&P 500,
attractive and believe it is sustainable. The company is undergoing a major transition
to integrate the acquired Verizon assets, which increases its risk profile, though early
signs from initial integration milestones have been comforting. Nevertheless, we
expect Frontier to put in the necessary capital investment as well as implement its
local engagement strategies to begin turning around these markets as early as 1H11.
The roadmap to an operational turnaround, likely upward revisions on synergy
guidance, as well as an attractive dividend make Frontier our top pick within RLECs.
Windstream Initiating with Neutral rating and $14 YE11 price target
Windstream has a rural footprint with the lowest access line density within our
RLEC coverage, which has enabled the company to deliver better operating metrics
than its peers. Moreover, management has been shifting its revenue stream away
from the declining consumer voice business and towards business and data segments,
which we think positions the company well for top-line stabilization when macro
56

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

trends improve. While the current dividend yield of 7.4% remains one of the highest
within telecom, the companys higher leverage and EV/EBITDA multiple compared
to those of wireline peers, as well as ongoing acquisition risk, keep us sidelined.

Wireline Investment Risks


Further deterioration in economic fundamentals could impact revenue
Further deterioration in economic fundamentals poses risk to access line and revenue
trends as consumer markets could continue to deteriorate at current high levels and
the expected recovery in business markets could be pushed further out. This would
decrease the traction telecoms could capture with their video and broadband growth
initiatives.
Further acquisitions could be risky for near-term free cash flow
The potential for additional acquisitions poses risk for the wireline companies. While
we typically like consolidators within the deteriorating wireline industry from a
synergy realization and margin maintenance perspective, a potential acquisition in
the near to medium term could impact near-term free cash flow fundamentals and
could increase risks around dividends.
Regulatory revenue declines could be significant
Wireline companies have significant exposure to regulatory revenue, which increases
the more rural a companys footprint gets. While we expect the regulatory revenue
stream to continue to diminish gradually rather than see a major short-term
disruption, a faster-than-expected decline in USF subsidies and access rates could
cause a significant hit to top lines and detract from companies ability to maintain
dividends.

Wireline Industry Overview


Negative Voice Trends Persisted Through Past Decade
Access line losses to wireless substitution and cable VoIP competition have been the
overarching challenge facing the wireline industry over the past decade. Wireline
providers have seen a significant secular decline over this period, with telecom
penetration levels dropping to ~40% of homes in urban regions. Further, the secular
shift was exacerbated by tough macro conditions in the past few years. Consolidation
and merger synergies, as well as significant organic cost reductions, were the main
initiatives taken by wireline carriers which battled to preserve their dividends and
cash flow.

57

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 24: Y/Y Access Line Losses in 3Q10 Were Worst at More Urban Carriers
0.0%
-2.0%
-4.0%
-6.0%

-3.7%

-4.3%

-8.0%

-6.9%

-10.0%

-7.8%
-9.3%

-12.0%
WIN

CNSL

CBB

FTR

CTL

-8.5%
VZ

-11.3%
T

-10.4%
Q

Y/Y CHANGE IN ACCESS LINES


Source: Company reports.

Voice now 60-70% of wireline industry consumer revenues, with battles for
share between telecom and cable
Over the last decade the wireline space has mostly been a battle of telecom vs. cable,
and the distinction between the bundles offered by the two has diminished further
recently, as telecoms have been rolling out video service, traditionally dominated by
the cable MSOs and satellite companies. Cable has already captured a significant
market share in consumer voice and is more recently showing more presence in the
small/medium business and enterprise segments which have bee served traditionally
by telecoms.
Telecom still dominates a smaller voice pie. Currently telecom and cable split
access lines with roughly 70% and 30% share, respectively, and we believe the
market is near equilibrium for cable stealing access lines from telecom. However, we
expect the pie to continue to contract due to ongoing wireless substitution. We
estimate that currently over 30% of urban territories are wireless-only households,
and we expect ongoing wireless substitution to drive a 40%+ wireless-only ratio in
urban regions by 2012, with ~25% level in rural territories.

58

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

300

Wireless -Subs

250
200

Wireline -Access
Lines

150
100
50

20
12
E

20
10
E

20
08

20
06

20
04

20
02

20
00

19
98

19
96

19
94

19
92

19
90

19
88

19
86

0
19
84

Access Lines & SUbscribers (m)

Figure 25: Wireless Subscriptions Have Been Replacing Access Lines

Source: Company reports and FCC.

Cable voice gains have stabilized


Cable MSOs initially attacked the telecom space by rolling out voice over internet
protocol (VoIP) service in 2004, and have reached ~20% penetration across their
markets since then. Their bundled offerings typically include voice as a $10 add-on
for a triple play with a data and video plan. The highly regulated telecom incumbents
had very little ability to fight back given the low price offered by cable. With much
of the regulation around voice pricing relaxed in recent years, and the availability of
a VoIP solution within the U-verse and FiOS product suites, line losses to cable have
contracted.
Figure 26: Cable Providers Have Gained Meaningful Share of the Voice Market
100.0%
80.0%
60.0%
40.0%
20.0%
0.0%
1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08
Telco

4Q08

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

Cable

Source: Company reports and J.P. Morgan.

Wireless substitution continues to be key to voice access line losses


Wireless substitution has been a material component of access line loss trends, and
became the major source of line losses over the past couple of years as telecom/cable
share has essentially stabilized but unlimited wireless continued to gain traction.
Leaps and Metros aggressive builds in 2009, increasing unlimited voice plan offers
by MVNOs, Life Line substitution offers from Assurance and Safe Link, as well as
tough macro conditions have kept wireless substitution trends robust. We estimate
59

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless is gaining ground in


even the lowest-credit wireline
segment

that presently ~30% of households across the US have disconnected wireline service
altogether, and we expect wireless substitution to become more prevalent as the
population ages, housing moves rebound, and unlimited wireless offers become more
competitive.

Figure 27: Telecoms Continue to Lose Access Lines, Though Cable Gains Are Well Below Historical Levels
2,000
1,000
0
(1,000)
(2,000)
(3,000)
2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

Telco

Cable

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

Source: Company reports and J.P. Morgan.

Rural voice has performed better going local is key


Access line losses at rural carriers have been much lower than peer levels, while
broadband penetration levels are also tracking higher. A major component of this is
the lower level of cable competition as well as weaker wireless coverage in regions
with lower population density. Nevertheless, the most rural player within our
coverage, Windstream, has ~65% cable VoIP alternative coverage as well as ~7580% broadband alternative coverage. We believe a major component of the better
operating metrics for the rural carriers is the local management and marketing
approach they implement in their markets.
We expect access line trends as well as broadband penetration levels to improve
significantly in the urban territories of the acquired Embarq markets after
CenturyLink implemented its local strategy. As rural players continue to acquire
more urban assets Frontiers acquisition of Verizon assets and CenturyLinks
pending acquisition of Qwest we will watch the metrics in the more urban regions
for evidence the acquirers are capitalizing on the opportunity to turn around the
urban markets. We expect to see meaningful improvements in Frontiers acquired
Verizon territories as early as 1H11 as we believe the companys handholding
strategy coupled with aspirational gift campaigns, such as computers for new
customers, will resonate well with customers.
Local approach has helped rurals on the business front as well. Over the past
couple of years cable MSOs have started marketing to the SMB customer, though
telecoms have not lost meaningful share on the SMB front. We believe the local
approach by telecoms has been a significant driver in the maintenance of market
share.

60

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 28: Access Line Losses as of 3Q10


0.0%
-2.0%
-4.0%
-6.0%

-3.7%

-4.3%

-8.0%

-6.9%

-7.8%

-10.0%

-9.3%

-12.0%
WIN

CNSL

CBB

FTR

CTL

-8.5%
VZ

-11.3%
T

-10.4%
Q

Y/Y CHANGE IN ACCESS LINES


Source: Company reports and J.P. Morgan.

High-Speed-Internet Split More Even Between Telco, Cable

We believe HSI penetration can


exceed 100% of voice lines

On the high-speed-internet (HSI) front, we see the current split of ~50%/50%


telco/cable as a reasonable equilibrium point. While cable has been the dominant
winner in this area over the past decade, telecom initiatives to deploy high-capacity
fiber across their footprints has stemmed the losses to cable and we expect
improvements in copper line capacity starting in 2011 to improve non-fiber market
metrics as well. While we continue to expect fluctuations on the share front as certain
players promote heavily in any given period, we believe the currently comparable
cable and telecom offers will keep overall market share near current levels.
Figure 29: HSI Penetration of Access Lines

50%
45%

44%
39%

40%

37%

36%

35%

31%

VZ

29%

30%

34%

32%

25%
20%
15%
10%
5%
0%
CNSL

WIN

CTL

FTR

CB B

Source: Company reports and J.P. Morgan.

61

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Telco Video Is Just Getting Warmed Up


Over the past two years AT&T and Verizon have conducted major builds for video
service. Since the beginning of 2009, T and VZ cumulatively added 3 million subs,
while cable (CMCSA, TWC, CHTR, and CVC) cumulatively lost 2.2 million subs,
~5% of their base, and satellite held its own with 1.9 million subscriber additions.
We expect telecom to double its current 7.3% share of the pay TV market over the
next few years. In their respective regions, Verizon and AT&T are expected to reach
~35% and ~30% terminal video penetration.

Consumer Buoyed by Data, Video


Voice Has Fallen to 60-70% of Consumer Wireline Revenue
As wireline voice service has lost value from a consumer standpoint, consumer voice
revenue has also become less relevant for the financial performance of wireline
providers. As we see little differentiation between a typical triple-play offer from a
telecom or cable MSO, retaining the existing customer base and upselling higher
bundles has become the key focus to minimize top-line losses in the industry.
Cables move into the voice space was the initial dominant force affecting the
cable/telecom landscape; however, over the past two years larger telecoms have
expanded their video offerings through aggressive fiber rollouts that compete directly
with those of cable companies. AT&T and Verizon are pushing video content to
homes, and Qwest is enabling speeds for over-the-top video.
AT&T and Verizon currently hold ~5% of the total pay TV subscriber base, which
we think could double over the next 18 months. AT&T is expecting to pass
30 million homes with fiber by the end of 2011 and we expect its current low-teens
penetration level with U-verse to reach ~30% over time. Verizon currently has
12.5 million homes open for sale for FiOS broadband and 12.1 million for video, and
plans to pass ~70% of its homes. Penetration levels in territories that have been open
for sale for 24 months have reached 30%, and management expects markets could
see 40-50% penetration levels over time. Qwests fiber footprint has reached
4.5 million homes, with a rate of 1 million incremental homes per year. CenturyLink
is testing video deployments more cautiously and expects to pass 1 million homes by
the end of 2011.
Despite the crowding in the video space created by the entrance of telecoms, the
pricing environment has remained healthy and we have not seen cable battling back
with price cuts to maintain subscriber base. We expect pricing to remain healthy
going forward, and expect to see further price increases as programming costs
continue to ramp in the cable landscape.

62

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 30: HSI Penetration of Total Access Lines


50%

Figure 31: Video Penetration of Total Access Lines

44%
39%

40%

14%

37%

29%

30%

13%
12%

36%

32%

31%

34%

12%

12%

11%
10%

10%

9%

8%

20%

6%
4%

10%

3%

2%

0%

0%

CNSL

WIN

Source: Company reports and J.P. Morgan.

CTL

FTR

VZ

CBB

WIN

CNSL

VZ

FTR

CB B

Source: Company reports and J.P. Morgan.

Video/Broadband Should Offset Consumer Voice Declines


While we expect the consumer voice revenue erosion to continue, we believe video
and broadband revenue gains will meaningfully offset the loss. Currently wireline IP
revenue (U-verse plus non-U-verse broadband) drives more than 42% of AT&Ts
consumer wireline revenues, and FiOS-related revenues constitute about half of
Verizons consumer revenue base. Verizons more aggressive push with its FiOS
service offering could increase penetration levels to 40%+ in its mature markets,
while at AT&T we look for penetration levels of ~30%.

Broadband Key Driver of Revenue, Customer Retention


Broadband drives a significant portion of consumer revenues, and has become the
mainstream solution offered by telecoms to fight the secular shift away from wireline
voice. We estimate broadband revenues currently make up ~20% of total consumer
wireline revenues. At AT&T U-verse and DSL, which represent total wireline
consumer IP revenue, now represent more than 42% of total consumer revenue. The
higher ARPU on higher-speed subscribers along with relatively low penetration of
higher speeds present a significant revenue opportunity for wireline players, in our
view, which will be key to offsetting the expected losses from the shrinking
consumer voice revenue stream.

63

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 32: Telecom and Cable Both See Growth on HSI


3,000
2,500
2,000
1,500
1,000
500
0
(500)
2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

Telco

Cable

4Q08

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

Source: Company data and J.P. Morgan.

Naked/Semi-Naked DSL as Well as Bundles for Retention


The telecom carriers primary focus is to increase the competitiveness of their
offerings through triple-play bundles that include a voice line, high-speed internet,
and video (core offers or as resellers) as well as wireless in some cases. For those
households that have cut the cord completely, however, most telcos have introduced
a naked DSL offering, which includes a broadband line usually paired with a
metered emergency voice line.

Wireless-only voice subs still


need a hard-line HSI connection

For example, Windstreams Greenstreak offer bundles naked DSL with an


emergency line that charges ~$0.10/minute for outgoing calls and enables free
incoming calls. Such offers have been in the market since 2007-2008. These products
are marketed to a targeted customer base, usually by direct mail, in order to not
cannibalize other wired voice subs, and they enable value-oriented customers to
decrease their monthly bill to less than US$35, which is the minimum they likely
would have to pay for cable. We estimate that naked DSL customers constitute a
low-single-digit percentage of customers in most rural regions in which the service is
offered and a high-single-digit percentage in more urban areas, but ongoing wireless
substitution could drive penetration to ~10%+ over time.

Telco Fiber in Urban Regions Going Head-to-Head with


Cable
AT&T and Verizon are on track to pass 60-65% of their footprints with fiber (for
Verizon) or enhanced copper service (for A&T) by the end of 2011. Qwest has also
been deploying a fiber-to-the-node (FTTN) strategy in its footprint and reached
~4.5 million homes, with plans to reach ~ 6 million homes (over half of its footprint
of 11 million homes). AT&Ts FTTN offer and Verizons fiber-to-the-home (FTTH)
strategy which goes all the way to the premise enable the companies to compete
head-to-head with the cable companies video offers. While Qwest is not pushing
video content, its broadband speeds are sufficient for over-the-top video streaming.

64

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Growth Coming from Video Opportunity


Video offers have been a typical addition to the telco triple-play bundles, and have
been essential to minimize subscriber churn. Video solutions offered by telco
providers vary from facility-based video products such as Verizons FiOS which
takes content all the way to the home, AT&Ts U-verse, and CenturyLinks IPTV
trials, to satellite resale offers with DirecTV or DISH which are insignificant to
revenue but critical from a customer retention perspective.
Over the past couple of years AT&T and Verizon have implemented aggressive
video builds, and gained significant traction with their offerings. AT&T expects to
pass 30 million homes by the end of 2011, and continue to extend to adjoining areas
thereafter, with the current low-teens penetration expected to reach ~30% over time.
Verizon currently has 12.5 million homes open for sale for data and 12.1 million for
video, and plans to pass ~70% of its footprint with fiber. CenturyLink is testing video
deployments more cautiously and expects to pass 1 million homes by the end of
2011.
Penetration levels in territories that have been open for sale for 24 months have
reached 30%, and management expects markets could see 40-50% penetration levels
over time. Despite the increasing crowdedness in the video space created by the
entrance of telcos, the pricing environment has remained healthy and we have not
seen cable battling back with price cuts to maintain subscriber base. Overall, we
expect telcos to hold a mid-teens share of the video market by 2016.
Bundling mobile with video could create telco advantage. Telcos have an
opportunity to innovate and provide themselves an edge over cable MSOs by
creating better value for consumers through more integrated services which bring
video content on the go from home DVRs to mobile devices like smartphones and
tablets. This likely would enable carriers to further increase prices, rather than
compete on price, which we have yet to see in the video space.
Satellite remains partner outside of fiber footprint. Where a facility-based video
offer is not a feasible solution due to heavy capital requirements, telcos are
partnering with satellite players to include video as an add-on to their offers, which
makes their bundles more comparable to the triple-play offers from cable companies.
Moreover, fixed-price agreements for a customers lifetime for voice/HSI are
resonating well with customers, and telcos can add video to the fixed-price voice/HSI
bundles, subject to price changes by satellite providers.

65

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 26: Estimated Terminal Penetration of Voice, Video, and Data by Technology
VZ-In Build (15.4m homes today)
No subscription

Voice

Data

TV

AT&T-In Build (26m homes passed today)

0%

10%

5%

No subscription

Voice

Data

TV

0%

10%

5%

Wireless

40%

5%

NA

Wireless

40%

5%

NA

Copper (circuit switched and DSL)

25%

0%

NA

Copper (circuit switched and DSL)

20%

5%

NA

Satellite
Cable

NA

NA

25%

Satellite

20%

45%

35%

Cable

FiOS

15%

40%

35%

19.0

100%

100%

100%

VZ-Out of Build (11.6 mm lines today)

Voice

Data

TV

0%

10%

15%

No subscription

U-Verse

NA

NA

25%

25%

50%

40%

15%

30%

30%

31.3

100%

100%

100%

AT&T-Out of Build (20mm today)

Voice

Data

TV

0%

10%

15%

No subscription

Wireless

30%

5%

NA

Wireless

30%

5%

NA

Copper (circuit switched and DSL)

40%

35%

NA

Copper (circuit switched and DSL)

40%

40%

NA

Satellite
Cable
FiOS

NA

NA

35%

Satellite

30%

50%

50%

Cable

NA

NA

NA

7.7

100%

100%

100%

VZ-Average lines

U-Verse

NA

NA

40%

30%

45%

45%

NA

NA

NA

21.7

100%

100%

100%

Voice

Data

TV

0.0

5.3

4.8

Voice

Data

TV

AT&T-Average lines

No subscription

0.0

2.7

2.1

No subscription

Wireless

9.9

1.3

Wireless

19.0

2.7

Copper (circuit switched and DSL)

7.8

2.7

Copper (circuit switched and DSL)

15.0

10.3

Cable

6.1

12.4

10.5

14.3

25.4

FiOS

2.9

7.6

6.7

4.7

9.4

9.4

26.7

26.7

26.7

26.7

53.0

53.0

53.0

Satellite

7.5

Satellite

16.5

Cable
U-Verse
53.0

22.3

Source: Company reports and J.P. Morgan estimates.

Figure 33: Telcos Video Offering Gaining Share from Cable


100.0%
80.0%
60.0%
40.0%
20.0%
0.0%

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

Satellite

35.1%

35.3%

35. 5%

35.8%

35.8%

35. 8%

35.9%

36.0%

36.1%

36. 2%

36.5%

39.0%

39.1%

39.2%

40.3%

Cable

64.5%

64.0%

63. 5%

62.8%

62.4%

61. 9%

61.4%

60.6%

60.0%

59. 2%

58.5%

55.2%

54.7%

54.1%

52.4%

Telco

0.4%

0.7%

1.0%

1.4%

1.8%

2.2%

2.8%

3.4%

3.9%

4.5%

5.0%

5.7%

6.2%

6.6%

7.3%

Source: Company reports and J.P. Morgan.

66

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 34: Telcos Winning Video Share from Cable and Satellite
800
600
400
200
0
(200)
(400)
(600)
1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08
Telco

4Q08
Cable

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

Satellite

Source: Company reports and J.P. Morgan.

Business, Regulatory 60%+ of Wired Revs


Business, regulatory, and wholesale revenue now make up over 60% of wireline
industry revenues within our coverage, with AT&T and Verizon having the highest
levels and the rest of the wireline providers closer to 55%. In the business space,
enterprise continues to be dominated by AT&T and Verizon, while the small and
medium-sized business space has become somewhat more competitive as cable has
entered the mix. On the regulatory side, we expect USF revenues to remain under
pressure. Wholesale revenues though are growing on the strategic side as the cell
backhaul business attracts any carrier with a nearby fiber footprint to light up as
many sites as it can.

Small/Medium Business Seeing Increasing Push from Cable


Although the capabilities required by small and medium-sized businesses (515 lines) are similar those for consumers, the public cable players have been much
slower to create a presence in SMB market than on the consumer side. While their
push into this segment has been more meaningful recently, they have still yet to
achieve significant market share. The lack of competition that RBOCs have seen in
this space, other than the presence of CLECs, has enabled higher margins on
products like T-1 data and voice lines than seen in the consumer space.
We see cable MSOs marketing more towards the low-end SMB market, with usually
5-10 lines, which could impact the rural players business revenue stream more
meaningfully than those of the larger telcos whose urban markets have been under
attack by CLECs for years.

67

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 35: SMBs Account for 13% of AT&Ts

Figure 36: And 7% of Verizons LTM Revenue

GEM

Consumer

solutions

35%

Other
2%

Global

26%

Consumer

Wholesale

32%

Serv ices
Small

20%

business &
alternate
channels
13%

Mass

Global
enterprise
solutions

Source: Company data, October 2008.

23%

Markets
Other
1%

2%

Global
Enterprise

SME

39%

7%

Source: Company data, October 2008.

Enterprise Dominated by the Big Two


Unlike the consumer and SMB spaces, the crowd in the enterprise market remains
small essentially a duopoly between AT&T and Verizon. Outside the big two,
Sprint has some presence (~5-6% share) and Qwest is positioned as the #2 or #3
telecom provider to its local Fortune 500 companies, holding ~4% of the enterprise
market. CenturyLink currently holds ~2% share of the enterprise market as legacy
Embarq properties have some exposure to enterprises in Las Vegas, Orlando, Florida,
and Carolina regions.
The global footprint and quality of service (QOS) requirements of the enterprise
customer act as barriers to entry for cable or smaller telco players. While pricing has
been stable and data volumes have been growing meaningfully, this segment has
seen some negative revenue trends in the past couple of years, largely from
enterprises cutting back on headcount as well as telecom spending budgets, and
demand shifting towards lower-priced IP products. On the competition front though,
we have heard very few complaints lately about aggressive pricing to steal customer
wins.
AT&Ts business revenue <40% voice. Currently voice represents less than 40% of
AT&Ts business revenue. From older offerings (circuit-switched voice) and legacy
packet technologies (ATM and frame relay), there has been a strong uptake of IPbased services for voice and data to the point that over 80% of AT&Ts enterprise
customers are on IP technologies today. Data revenues continue to gain importance,
as IP-based services and managed services grow in volume. As companies move
more of their applications and their content into the cloud, hosting and cloud
computing are becoming more important as well.

68

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Business Revenues Should Grow Following a Macro


Recovery
Business revenues have seen significant pressure due to the challenging
macroeconomic environment. Pressure on volumes given low business activity from
customers, who have been deferring potentially costly initiatives and being cautious
on headcount, has been exacerbated by pressure on the pricing front as the migration
to IP from ATM and Packet has progressed. There has yet to be a meaningful
comeback on the macro front, though we expect volumes to come back as
unemployment trends improve. With the migration to lower-cost IP-based services
close to completion, pricing should stabilize as well, which positions business
revenues to potentially see top-line growth, possibly as early as 2H11.
Nevertheless, largely from easier comparisons, y/y changes have shown some
recovery in the past few quarters, and we look for further stabilization, and
eventually growth, following a macroeconomic stabilization.
Figure 37: Unemployment Rate Is Hovering at ~10%
15%

500

13%

400
300

9%
7%

200

5%

100

y/y change (bps)

Unemployment

11%

3%
0

1%
-1%

-100

Jan 07

Jun 07 Nov 07 Apr 08

Sep 08 Feb 09

Jul 09

Unemploy ment

Dec 09 May 10 Oct 10

y /y change

Source: BLS and J.P. Morgan.

Figure 38: Unemployment

Figure 39: Y/Y Change in Unemployment (bps)

30%

Source: BLS and J.P. Morgan.

10
Se
p

ay
10
M

Ja
n1
0

09
Se
p

ay
09

10
Se
p

Ja
n1
0
M
ay
10

09
Se
p

08
Ja
n0
9
M
ay
09

Se
p

Ja
n0
8
M
ay
08

0%

10%

Ja
n0
9

700
500
300
100
-100

20%

16-19 y ears

20-24 y ears

16-19 y ears

20-24 y ears

25-34 y ears

Population

25-34 y ears

Population

Source: BLS and J.P. Morgan.

69

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 40: Unemployment

Figure 41: Y/Y Change in Unemployment (bps)

African American

Hispanic

Source: BLS and J.P. Morgan.

Population

ay
M

African American

10
Se
p

10

0
n1
Ja

09
M

ay

n0
Ja

Se
p

10
M

ay

n1
Ja

Se
p

09
ay
M

n0
Ja

Se
p

08
ay
M

n0
Ja

Population

Se
p

-100
9

0.0%
10

100

5.0%

09

300

10.0%

08

500

15.0%

09

20.0%

Hispanic

Source: BLS and J.P. Morgan.

Regulatory Revenue Is a Diminishing Stream


Regulatory revenues have been a steadily diminishing revenue stream as lower
access minutes of use and access rates have pressured switched access revenues.
Over the past couple of years there has been a lot of noise on the regulatory front, as
inter-carrier compensation and USF subsidies were under review for potential
reform, and as the national broadband plan was proposed.
A key risk for most of the rural telcos has been inter-carrier compensation reform,
which could lower access rates to a fixed level well below current rates, and hit
access revenues in a major way.
USF support likely to shift from voice to data
On the USF subsidies front, the increased focus on broadband will likely result in
reallocation of subsidy money away from rural voice and towards broadband
providers.
On the flip side though, the broadband stimulus plan has provided opportunity for
telcos to apply for grant money to increase broadband reach in rural territories.
Windstream, for example, has applied for broadband stimulus money and won
$180 million in grants which it must match with only $60 million of its own
spending.
Nevertheless, as the FCC plans to reduce the funding allocated to voice service and
shift it to broadband, rural carriers should eventually take some level of hit.
Generally, the more rural a carriers footprint, the higher the reliance on regulatory
revenues. Network access revenue accounts for ~16% of CenturyLinks revenue, and
is composed of intercarrier compensation and USF funds. The pending merger with
Qwest will decrease net exposure to this revenue stream significantly, making
Windstream the most highly exposed player within our coverage. Frontier had ~17%
exposure to this stream before the Verizon transaction, but has reduced its exposure
significantly, to ~10% after the deal.
Table 27: Regulatory Revenue Exposure of Wireline Providers
Switched access and
USF exposure

FTR

CTL

WIN

<6% switched acces


and <4% USF

~16%

net payer of switched


access, <3% USF

~16%

Source: Company reports and J.P. Morgan estimates.

70

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

While we havent gotten a firm answer on the USF and ICC fronts, we expect
ongoing discussions on the topics will continue to pose risk factors for the free cash
flow prospects of the RLECs, especially given the high-margin structure of
regulatory revenues. Nevertheless, we do not expect to see a resolution or an
implementation that should impact the carriers over the next year. While ongoing
modest declines in this revenue stream would likely not be a trigger for dividend
cuts, any change in regulation could be significantly detrimental to free cash flow in
the medium term. We model this revenue stream to continue to grind down over the
next decade, at a high-single-digit rate, as the FCC implements its broadband plan.

Special Access Revs to Grow with Fiber Rollout to Cell


Sites
Carriers receive special access revenues from wholesale customers (usually wireless
carriers in region) for providing backhaul services over their copper or fiber
footprints. Special access revenues are following a healthy growth trajectory, and
despite some slowdown in the growth in a tough macro environment, we expect
growth to reaccelerate as data needs of wireless carriers are consistently growing.
Rural players such as Windstream mostly offer special access service on a copper
footprint but, given the necessity of extra bandwidth for data usage on cell sites, the
companies have initiatives to extend fiber plant to cell sites. Qwest has outlined a
major fiber-to-the-cell-site initiative, and it has signed contracts for 4,000 of the
17,000 sites in territory. These contracts provide 5-10 years of guaranteed minimum
payments a high margin and stable cash flow.

Consolidation to Insulate Cash Flow


RLEC Consolidation Could See a Short Pause
The wireline industry has been going through a wave of consolidation as access line
losses caused significant top-line pressure and squeezed profitability prospects. In
order to scale up and generate synergies, larger players in the segment have emerged
as repeated acquirers. CenturyLink has quadrupled in size over the past four years
with its Madison River (2006) and Embarq (2009) acquisitions, and is in the process
of closing its acquisition of Qwest which was 1.5x its size. Frontier recently acquired
Verizon access lines which have tripled the size of the company, and Windstream
continues to acquire and integrate tuck-in deals while looking opportunistically at
further deals.
While the main rationale for RLEC acquisitions has typically been cost savings
through headcount reduction as well as the transfer of traffic to more efficient
networks, there have also been efforts to gain more exposure to the more sticky
business revenue stream, which was a catalyst for Windstreams recent CLEC/data
center acquisitions which have replaced RLEC tuck-ins over the past year.
While there has been a significant amount of M&A in the RLEC space over the past
18 months, we expect a slowdown on this front for most of 2011. With Frontiers and
CenturyLinks hands full with integration of their large acquisitions, and Windstream
turning more towards CLEC acquisitions which would increase its presence on the

71

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

business services front, we do not expect any large RLEC deals over the next 12
months.

CLEC Acquisitions Pose New Growth Opportunities


At the end of 2009 Windstream announced its acquisition of NuVox, a competitive
local exchange carrier (CLEC), followed by the acquisition of another CLEC, as well
as a data center, in an effort to increase its exposure to business customers.
CLEC business is inherently a lower-margin business than RLEC operations, and
this business line faces more competition from larger carriers. However, there is
significant opportunity for growth in business revenues, which makes these assets
more valuable. Moreover, as we have seen with Windstream, acquiring CLEC assets
with a complementary footprint allows the acquirer to strengthen its local presence in
existing territories and positions the company better for growth opportunities in the
small and medium-sized business segments.
Given the increasing pressure in the consumer business, we view Windstreams
potential reorientation toward business markets as an attractive medium-term
strategy. We believe Windstreams recent initiative could eventually be replicated by
other RLECs as the consumer space continues to get tougher. CenturyLink already
has CLEC operations in 12 states and has explicitly committed to a CLEC edge-out
strategy, while Frontier screens as a less likely prospect as it sold its Electric
Lightwave CLEC business and has its hands full with Verizon integration.

Margins Stabilizing, Cash Flow to Recover Somewhat


Wireline cash flow has been under substantial pressure over the past couple of years.
As revenue losses pressured margins across the industry, carriers have rationalized
their cost bases to minimize margin pressures. For AT&T and Verizon, top-line
pressures and pension costs were exacerbated by the opex component of U-verse and
FiOS deployments, which resulted in ~300 bps of margin declines from 2008 levels.
While we do not expect margins to recover to historical levels, we do model a
modest ~30-40 bps of margin recovery for the two carriers for 2011 and another
~50 bps for 2012 from 2010 levels, which likely would look conservative if the
economy rebounds. For Ts and VZs wireline operations combined, we look for
0.6% and 1.3% EBITDA growth for 2011 and 2012, respectively.
For the wireline-only players, we look for margin expansion as well, especially for
Frontier, as we expect significant synergies from the acquisition of Verizon access
lines coupled with some level of macroeconomic relief to more than offset the
ongoing top-line pressures over the next few years.
Table 28: EBITDA Trends Across Wireline Carriers
AT&T
T Margin

1Q09
5,234
32.4%

2Q09
5,069
31.7%

3Q09
4,991
31.7%

4Q09
4,984
31.9%

2009
20,278
31.9%

1Q10
4,809
31.2%

2Q10E
5,007
32.5%

2Q10E
4,957
32.5%

3Q10E
4,753
31.5%

2010E
19,526
31.9%

1Q11E
4,767
31.9%

2Q11E
4,799
32.1%

3Q11E
4,831
32.3%

4Q11E
4,809
32.4%

2011E
19,205
32.2%

1Q12E
4,843
32.6%

2Q12E
4,850
32.6%

3Q12E
4,884
32.8%

4Q12E
4,859
32.9%

2012E
19,436
32.7%

Verizon
VZ Margin

2,401
22.6%

2,329
22.0%

2,266
21.2%

2,243
21.2%

9,239
21.8%

2,008
19.4%

2,133
20.8%

2,164
21.0%

2,234
21.8%

8,539
20.7%

2,220
21.8%

2,232
21.9%

2,254
22.0%

2,315
22.8%

9,021
22.1%

2,286
22.7%

2,305
22.8%

2,270
22.3%

2,294
22.7%

9,154
22.6%

Frontier
FTR Margin

753
48.2%

737
47.8%

733
48.4%

655
45.0%

2,878
47.4%

679
46.6%

639
44.6%

574
40.9%

640
46.0%

2,532
44.5%

644
47.0%

643
47.0%

632
47.5%

629
47.5%

2,548
47.2%

629
47.7%

630
48.0%

624
48.5%

621
48.5%

2,503
48.2%

Windstream
Win Margin
Total
Change y/y

470
46.9%
8,858

453
45.7%
8,588

441
45.0%
8,431

467
47.4%
8,349

1,830
46.3%
34,225

469
48.2%
7,964
-10.1%

459
47.9%
8,239
-4.1%

462
47.8%
8,157
-3.3%

474
49.0%
8,100
-3.0%

1,863
48.2%
32,460
-5.2%

469
48.7%
8,100
1.7%

459
48.4%
8,133
-1.3%

462
48.3%
8,179
0.3%

474
49.5%
8,227
1.6%

1,865
48.7%
32,639
0.6%

464
48.5%
8,221
1.5%

454
48.2%
8,239
1.3%

459
48.3%
8,236
0.7%

471
49.5%
8,245
0.2%

1,847
48.6%
32,941
0.9%

Source: Company reports and J.P. Morgan estimates.

72

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Capital spending ready to ease in wireline. The major part of U-verse and FiOS
build phases is expected to finish in 2011. We expect capex to remain under control
for the rest of wireline space as well, given that Qwest is more than two-thirds
complete with its fiber deployment, CenturyLink is maintaining its cautious stance
on IPTV, and Windstream will be deploying success-based capex to enhance its
broadband availability but has no plans to push video content. Frontier plans to ramp
capex over the next few quarters to increase broadband availability as quickly as
possible, but eventually expects capex levels around 10% of revenues. Net-net, we
look for capex to decline 8% in 2011 and 4% in 2012.

We expect Verizon wireline


capex to decline 12% in 2011

Table 29: Capex Trends Across Wireline Carriers


AT&T Inc.
% of rev
Verizon
% of rev
Frontier PF
% of rev
Windstream
% of rev
Total
Change y/y

1Q09
2,479
15%
2,003
19%
190
12%
101
10%
4,773

2Q09
2,706
17%
2,338
22%
200
13%
103
10%
5,347

3Q09
2,536
16%
2,273
21%
191
13%
94
10%
5,094

4Q09
3,336
21%
2,278
22%
231
16%
114
12%
5,959

2009
11,057
17%
8,892
21%
812
13%
412
10%
21,173

1Q10
2,122
14%
1,566
15%
166
11%
67
7%
3,921
-18%

2Q10E
2,535
16%
1,781
17%
195
14%
104
11%
4,615
-14%

2Q10E
2,796
18%
1,751
17%
175
12%
113
12%
4,835
-5%

3Q10E
3,275
22%
1,844
18%
290
21%
120
12%
5,529
-7%

2010E
10,728
18%
6,942
17%
825
15%
404
10%
18,900
-11%

1Q11
2,388
16%
1,527
15%
202
15%
103
11%
4,221
8%

2Q11E
2,389
16%
1,529
15%
201
15%
102
11%
4,221
-9%

3Q11E
2,388
16%
1,537
15%
196
15%
102
11%
4,223
-13%

4Q11E
2,959
20%
1,523
15%
197
15%
102
11%
4,782
-14%

2011E
10,124
17%
6,116
15%
796
15%
410
11%
17,447
-8%

1Q12
2,352
16%
1,510
15%
176
13%
102
11%
4,140
-2%

2Q12E
2,357
16%
1,515
15%
175
13%
101
11%
4,148
-2%

3Q12E
2,359
16%
1,524
15%
171
13%
101
11%
4,155
-2%

4Q12E
2,342
16%
1,510
15%
171
13%
102
11%
4,125
-14%

2012E
9,410
16%
6,059
15%
692
13%
407
11%
16,567
-5%

Source: Company reports and J.P. Morgan estimates.

73

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

74

North America Equity Research


13 January 2011

North America Equity Research


13 January 2011

Tower Industry

Tower Industry

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

75

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

76

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Tower Key Investment Points


Sprints Network Vision plan could drive upside to 2011-12 activity
While the iDEN shutdown will eventually reduce tower demand, in the near term
Sprint will need to augment all of its 48,000 remaining cell sites to enable the
800 MHz frequency band for its CDMA network. We expect Sprint to start
augmenting its cell sites in 2011 and finish in 2013, and model a tower lease expense
increase of 6.5% in revenue in 2011 and 2.2% in 2012. We expect Sprint to begin
taking down iDEN cell sites in 2014 with the full impact in 2016 when net dilution is
16% to our base case with CDMA augmentations adding 10 percentage points and
iDEN dilution of 26 percentage points. We believe the industry could lose about 3%
or more in site revenue from Sprint Nextel once its network is fully optimized.
Strong growth continues from increased density of 3G network and 4G builds
We forecast strong wireless capacity demand growth driven by subscriber adoption
of smartphones and data devices like tablets and wireless broadband. 4G network
builds have begun in earnest and amendment activity should drive tower revenue
growth through 2013. Verizon has driven most amendment activity so far but we
expect Sprint and AT&T amendments to accelerate in 2011. Clearwire has stalled
lately but could reaccelerate its WiMAX build with additional funding. We dont
expect much from Lightsquared, but if it finds partners and funding this could be a
significant driver in 2012.
International expansion driven by higher potential returns
Some tower companies have looked to international markets for growth as the
domestic market has matured. With fewer opportunities to build towers in the US
and recent domestic deals commanding multiples as high as ~20x tower cash flow,
tower companies are looking to international markets for higher rates of return.
Although we consider these investments higher-risk than domestic towers, a diverse
portfolio of international markets can spread risk and allow investors to focus more
on IRRs as much as 7-9% higher than those for domestic towers.
REIT conversion should lead to continued low effective tax rates
Tower companies have historically used the fast depreciation of tower assets to lower
effective tax rates. As operating losses run out for American Tower in the next two
years, we expect AMT to convert to REIT status to continue its low effective tax
rates. Interestingly, REIT companies tend to trade at higher multiples than tower
companies, despite tower companies better growth and customer base. The
J.P. Morgan REIT coverage universe is trading at 20.8x on a price-to-2011E AFFO
basis versus about 20.0x for the tower companies on a price-to-2011E RFCF basis
(net of straight-line impacts). American Tower has the biggest potential downside
risk if it does not convert to a REIT due to its potential tax liabilities. We value AMT
at $60 per share assuming REIT conversion versus $45 without conversion.
Improved debt structures after 2010 refinancing actions
During the financial crisis, the tower companies were faced with not being able to
raise debt at all and only later at unattractive rates with near-term maturities
weighing on the companies. The highly leveraged, high-multiple tower stocks fell
51-72% vs. the S&P 500 down 38%. While the tower companies remain leveraged at
4-7x, all the companies have taken advantage of a friendly debt financing
environment and have raised a total of $6.3 billion of fixed-rate debt with an average
77

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

rate of 4.9% since the start of 2010. In addition, maturities were extended and the
first tranches are not due until 2015 and some do not mature until 2020. Each
company has laddered out debt schedules so an overbearing amount of principal does
not come due in one year.

Tower Stock Recommendations


American Tower Initiating with Overweight rating and $60 YE11 price target
American Tower, with its scale, conservative financial position, and focus on
profitability, has been the most stable tower company and offers an attractive
risk/reward at this level, in our view. International opportunities could drive higher
returns in comparison to those in domestic markets and the company has diversified
into eight countries already, with no country but the US accounting for more than
10% of revenue. The firms relatively low leverage at 4x EBITDA provides us with
additional comfort and gives the company the ability to consider a larger tower deal
or buy back stock. Finally, we see REIT conversion as the most likely case for
American Tower and include it in our target valuation of $60 per share; however, if
this does not occur, our estimated YE11 valuation for the company would be only
$45, implying 11% downside from the current level.
Crown Castle International Initiating with Overweight rating $55 YE11 target
Crown Castle has substantial scale and an attractive tower portfolio in the top 100
markets. We are confident Crown Castle will be able to turn the positive business
trends into solid revenue, EBITDA, and cash flow growth and view shares of the
company as attractive at current levels. In addition, leverage at Crown Castle is
relatively modest at about 5x, which provides us with additional comfort and gives
the company the ability to execute a larger tower deal or buy back stock. Finally, we
see REIT conversion as the most likely case for Crown Castle given NOLs are
forecast to run out in 2016 and include it in our target valuation of $55 per share. We
see little downside risk to our valuation if a conversion is delayed due to minimal net
impact of tax changes.
SBA Communications Initiating with Overweight rating and $53 YE11 target
SBA has been a superior operating and financial performer; as a result, we believe
SBA warrants a premium valuation. SBA has decided to not sign a master lease
agreement (MLA) with AT&T, and so should report faster growth in 2H11 and 2012
than its peers as AT&T signs LTE amendments. SBA has started to expand
internationally with a focus on markets that are fairly mature but still offer higher
growth and better return prospects than the US. While we see REIT conversion as a
possibility when the companys NOLs run out, we dont think this will happen in our
forecasting time frame (before 2020).

78

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Tower Investment Risks


iDEN shutdown could be a 3%-plus headwind for the tower industry
We expect Sprint Nextel to shut down its 20,000 iDEN cell sites in the 2012-2016
time frame with the bulk coming in 2013-2015. We model iDEN site shutdowns
dragging our base-case revenue scenario of 68,000 sites by 4% in 2012, and we
expect an incremental ~600 bps in dilution annually until 2015. Peak dilution is
expected in 2016 at 26%, a substantial negative impact if CDMA augmentations do
not partially offset the revenue loss. We believe the tower industry could lose about
3% or more of gross industry revenue once Sprint Nextels network is fully
optimized.
Carrier consolidation likely would reduce growth
While we do not expect near-term consolidation for major national wireless carriers,
we would view another major merger as a long-term negative for tower companies.
Consolidation of a regional carrier could be a small headwind for tower companies,
but a large national consolidation likely would have a significant negative impact.
The last substantial merger that led to cell site and tower consolidations was AT&T
Wireless and Cingular, which shaved off about a percentage point in growth for the
tower companies for about five years.
Debt refinancing could be at risk if credit markets close or tighten
The tower companies have substantial leverage that ranges from 4-7 times with debt
maturities in the 2014-2020 time frame. The companies need to be able to refinance
their debt as it matures. If the credit markets were closed or refinanced debt bore
substantially higher interest rates, the tower companies would be negatively affected.
International expansion could create uncertainty
The major independent tower operators either own towers already or could consider
acquiring towers in international markets. We believe that site management and
ownership in foreign markets offer an opportunity for growth but could also bring
incremental risks not experienced in the United States. The tower ownership model
in the US is clear in terms of the structure of leases, pricing, operating and capital
costs, property ownership rights, and zoning issues. This model has not been as well
defined in other developing countries. Building up a bigger presence in any emerging
market, in which the wireless industry is still developing and country-specific risks
are very different from those in the United States, could negatively impact a
companys risk profile and multiple.
Wireless technology innovation could lead to less reliance on towers
Technological advances in wireless equipment could reduce the number of sites
required or the amount of space needed on a tower. The unlimited carriers have used
Distributed Antennae Systems (DAS) to build out parts of some markets, instead of
traditional tower and rooftop builds. Being able to share one antenna for multiple
wireless technologies is another potential risk for the tower companies. Other
innovations in technology and equipment could lead to negative impacts for the
tower companies.

79

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

High multiple valuations could come under pressure in a volatile environment


The tower companies trade at high forward EV/EBITDA and EV/revenue multiples
of about 14-17x and 9-11x, respectively, substantially higher than those of the
S&P 500. Disruptions in the financial markets or negative earnings results could
dramatically reduce the multiples investors are willing to pay for tower company
stocks. Any operational missteps could also lead to a higher risk profile for the tower
companies, especially given their reputation for stability.

Tower Industry Overview


The wireless transmission tower companies continue to benefit from solid
momentum due to positive wireless industry trends. Data growth is driving tower
revenue with more robust 3G network builds and new 4G builds. Carriers continue to
expand wireless coverage and improve network quality to reduce churn. We expect
the continued adoption of smartphones and expansion of network capacity to support
data usage to drive the need for more towers and equipment on existing towers.
AT&T and Verizon are leading the way with robust 3G and 4G activity and other
carriers are a distant third, but could provide incremental upside. Until AT&T or
Verizon slow down their leasing activity, the tower companies will continue to face a
positive operating environment, in our view.

Carriers Continue to Spend on Wireless Networks


Growing data use and new
technologies keep tower traffic
high

The wireless industry remains extremely competitive and key differentiators include
network quality, capacity, and speed. The rapid adoption of smartphones has
burdened the data capacity of wireless networks, which were originally optimized for
voice services. Carriers are racing to upgrade their wireless networks to be able to
handle the data demand generated by smartphone usage. Furthermore, new 4G
networks are being built with the main purpose of being able to handle faster data
speeds and higher capacity.
Table 30: US Carrier 4G Deployment Status
Company
AT&T
Verizon Wireless
T-Mobile USA
SpectrumCo LLC (Cable co's)
Sprint
Clearwire
Echostar
MetroPCS
Leap Wireless
Cox Wireless
US Cellular
Cellular South

Legacy network
WCDMA/HSPA
CDMA
WCDMA/HSPA
NA
CDMA
WiMax
NA
CDMA
CDMA
CDMA
CDMA
CDMA

Source: Company reports and J.P. Morgan.

80

4G Status
75M pops LTE EOY 2011, Complete 2013
38 live markets
HSPA+ for now
NA
WiMax with CLWR
120 M pops today
NA
LTE deployed today
2011 LTE trials
NA
2011 LTE trials
4Q11 LTE target

4G spectrum plan
30-40 MHz in AWS and 700 bands
40-50 MHz in AWS and 700 bands
Refarm 2G/3G spectrum
20 MHz AWS
Clearwire, refarm 2G/3G spectrum
~150 MHz
~10 MHz of PCS or AWS
Refarm PCS/AWS

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 31: Capex by Carrier


In millions
Verizon Wireless
AT&T Mobility
Sprint Nextel
T-Mobile
US Cellular
nTelos
MetroPCS
Leap Wireless
Total
National carrier total

2008
6,510
5,869
2,259
3,603
586
88
1,201
796
20,911
18,241

2009
7,152
5,907
1,161
3,662
547
45
832
807
20,112
17,882

2010E
8,405
8,521
1,471
2,741
589
45
810
418
23,001
21,138

2011E
8,842
8,740
2,350
2,850
620
46
698
450
24,596
22,782

2012E
9,228
8,745
2,850
2,850
620
50
676
555
25,573
23,673

Source: Company reports and J.P. Morgan estimates.

Amendment Activity Should Dominate Again in 2011


Carriers that want to change out the equipment on their tower lease may have to pay
an amendment fee to the tower company that is tacked on to the monthly lease
payment and is usually subject to the same annual escalators as the lease. As Verizon
and AT&T overlay their networks with LTE, they often need larger or heavier
antennas as well as more ground space for the extra base station. These typically cost
$300-500 per tower per month to add to the lease. We expect Sprint to start
amendment activity in 2011 as well and need larger antennas as well as add a
Remote Radio Head (RRH) to the antenna array which adds weight and cost to the
lease.
2010 closed strong; 2011 outlook driven by VZ, with T pulled forward
Site rental revenue grew an estimated 10% for the tower industry in 2010 and we
forecast 8% growth in 2011. For 2010, this broke down to 2% escalator growth (net
of 1% churn), 4% tower portfolio growth, 3% amendment-related growth, and 1%
from Clearwire, for 10% growth. For 2011, we forecast a slower growth rate for
tower builds and lower amendment benefit due to the pull-forward of AT&T into
2010. Verizon should continue a high level of amendment activity in 2011 due to the
continued buildout its LTE network. We model 2% escalator growth (net of 1%
churn), 3% tower portfolio growth, 2% amendment growth, and 1% from Clearwire,
for 8% net growth. Upside could come from significant activity from T-Mobile USA,
Sprint amendments, LightSquared expansion, or additional coverage builds by Leap
Wireless or MetroPCS.
Master lease agreement amendment for AT&T pulled revenue into 2010
A major carrier (we believe AT&T) signed new Master Lease Agreements (MLAs)
with American Tower and Crown Castle in 3Q10 to accelerate its LTE overlay.
While the individual agreements vary somewhat, the major provisions seem likely to
have a similar impact on each companys revenue.
Use fee tacked on to escalator for roughly six years. The MLA is structured to
add an additional use fee of ~3% to the regular ~3.5% lease escalator for six years
which allows the carrier to add whatever is necessary to its existing towers and RAD
centers during that period, after which the escalator drops back to the standard
~3.5%. Thus, rather than renegotiating the lease for a specific equipment package on
a site-by-site basis, the carrier can change out equipment much faster with a similar
overall cost during the period in which the use fee is accruing.

81

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

On at least one of the MLAs the carrier can only do this work for free on its towers
during the six years in which the use fee is accruing. If it doesnt update the towers
during that six years and wants to at a later date, that would require an additional
amendment process and most likely payment.
We do not think that any
significant upfront cash came
with the new MLAs

Straight-line accounting will drag 2011+ revenue for AMT, CCI. Because of
straight-line accounting rules, AMT and CCI calculate the total incremental revenue
created by the MLA over the lease period and recognize it on a straight-line basis
through the contract. Thus, American Tower guided to $21 million in additional
revenue in 4Q10 due to accounting on the new MLA, but we believe this will not
have any cash impact on its business near term. Assuming the 3% additional use fee
applies for six years of a ten-year average lease life results in incremental revenue of
12.6%. Now that the revenue has been booked upfront for all AT&T amendments,
however, the new MLAs will negatively impact GAAP revenue growth for CCI and
AMT from 2H11.

We expect SBA to come out inline or ahead after not signing


the MLA

SBA prefers pay-as-you-go approach. SBA Communications has chosen to not


sign a similar MLA as have CCI and AMT, and noted that it believes it could see a
better outcome by completing one-by-one augmentations in a short amount of time.
Our analysis estimates that a site-by-site augmentation that hit 100% of sites over
five years could mean 16.8% more revenue for SBA from that carrier, better than the
12.6% return of the new MLAs for CCI and AMT. However, a ten-year lease-up
scenario would lower the return to 10.6%, about 200bps below those estimated for
the new MLAs. We estimate the breakeven scenario of the new MLAs versus a siteby-site lease up is about 7-8 years.
Some may ask if AMT and CCI could potentially attract more business from the
carrier because of the new MLAs. While AT&T may be able to hit the CCI or AMT
sites earlier in its 4G overlay because of pricing and the MLAs, we expect that over
time AT&T will need to upgrade nearly 100% of its towers for 4G, whether they
belong to SBA, AMT, or CCI. Since that 4G upgrade probably hits 90%+ in the next
3-5 years, we cant fault SBA for holding back.

Table 32: Comparison of New MLA for AMT and CCI vs. Base Case
Base tower rent ($2000 month)
Escalator
Use fee
Base case (3.5% escalator)
Base case (3.5% escalator and 3% use fee for 6 years)

$24,000
3.50%
3.00%
0
1
$24,000 $24,840
$24,000 $25,560

2
$25,709
$27,221

3
$26,609
$28,991

4
$27,541
$30,875

5
$28,504
$32,882

6
$29,502
$35,019

7
$30,535
$36,245

8
$31,603
$37,514

9
$32,710
$38,827

Total
$281,553
$317,134

Annual
$28,155
$31,713

Monthly
$2,346
$2,643

Source: Company reports and J.P. Morgan estimates.

CDMA Augmentations Reduce iDEN Pain


The market so far appears to
have ignored the potential
positive from Sprints network
upgrade for towers

Sprint Nextel currently has the largest tower lease count of any American carrier
68,000 sites even though it has only about 50% of the customers of Verizon
(43,000 sites). This excess tower count is due to the still-unconsolidated combination
of Nextels iDEN network, which was built very densely in the early 2000s to add
capacity, and Sprints CDMA network in the 1900 MHz band which required more
sites for coverage than a cellular (800 MHZ) network.
Sprint recently announced its Network Vision blueprint a full modernization of its
network that will entail upgrading site and core network hardware. The plan includes
putting CDMA service on Nextels 800 MHz spectrum for higher quality and
coverage, deploying next-generation PTT service on CDMA, and starting to shut

82

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

down cell sites on its iDEN network in 2013. The company has a goal to eliminate
roughly 20,000 of its 68,000 sites late in that transition, and we believe the majority
of sites would roll off in the 2013-2015 period. The iDEN network could go
completely dark in 2015-2016, 3-4 years after Sprint starts the migration process.
Table 33: Sprint Nextel Exposure by Tower Company as % of Total Revenue
American Tower
Crown Castle
SBA Communications

Sprint Nextel Exposure


18%
22%
22%

iDEN Exposure
Less than 3%
Less than 3%
Less than 3%

Source: Company reports and J.P. Morgan estimates.

Sprints Network Vision plan could drive 6.5% upside to 2011 Sprint lease
spend
The near-term impact of Sprints Network Vision plan could actually drive upside to
tower company results. Sprint will need to augment all of its 48,000 remaining cell
sites to handle the 800 MHz frequency band for its CDMA network. We expect
Sprint to start augmenting these cell sites in 2011 and finish the majority in 2013,
and begin taking down iDEN cell sites in 2012 with final shutdowns in 2016. While
the net negative impact will be greater from shutting down 20,000 iDEN sites,
CDMA augmentations should come first and soften the effects of iDEN cell site
decommissions.
iDEN shutdown is four-plus
years away

iDEN shutdown could cut Sprint tower spending by 26%...


In our iDEN shutdown scenario, we model Sprint Nextel shutting down 20,000
iDEN cell sites from 2012 to 2016 with the bulk coming in 2013-2015. In 2012, we
model iDEN site shutdowns dragging our base-case revenue scenario of 68,000 sites
by 4% and we expect an additional ~600 bps in dilution a year until 2015. Peak
dilution is expected in 2016 at 26%, a substantial negative impact if CDMA
augmentations do not lessen the blow.
but net impact closer to 16%, translating to ~3% of tower industry revenue
Sprint Nextels CDMA augmentations could add about 10% in revenue to our basecase scenario starting in 2013, which partially offsets the losses from the iDEN site
shutdowns. On a net basis, we see low-single-digit revenue gains in 2011 and 2012
with only 1% dilution in 2013. The full impact of the iDEN shutdown is likely to be
felt in 2016 with net dilution of 16%, with iDEN dilution of 26 percentage points and
CDMA augmentations adding 10 percentage points. Overall these estimates translate
into the industry possibly losing a net 3% or more in site revenue once Sprints
network is fully optimized.
In the table below we outline the impact of the CDMA augmentation spending as
well as the iDEN rolloff. We start with a base case of 68,000 sites with a 3%
escalator as if nothing would change from here. We then add the CDMA
augmentation assuming 16,000 sites amended each year from 2011 to 2013 at
$300 per amendment, which also grows at the standard escalator. Finally, we start
removing iDEN sites in 2012 with 2,000 sites, but ramp that to 5,000 sites each year
from 2013 to 2015.

83

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 34: Net Effect of Sprints Network Vision Is ~16% Reduction in Site Leasing Expenses
A - Base Case
Sprint Nextel sites
Annual lease cost
Revenue from sites
Escalator

2010
68,000
$24,000
1,632,000,000
3%

2011

2012

2013

1,680,960,000

1,731,388,800

1,783,330,464

2011
16,000
$3,600
57,600,000

2012

2013

59,328,000
16,000
$3,600
57,600,000

2010
B - CDMA Augs.
Annual amendment fee
48,000

57,600,000
2010

116,928,000

2011

2012
(3,000)
$24,000
(72,000,000)

C - iDEN shutdown
$24,000
(20,000)

B&C net impact


A
Resulting annual Sprint revs
% change y/ y

1,632,000,000
1,632,000,000

2014

2015

2016

2017

2018

2019

2020

2021

1,891,935,289

1,948,693,348

2,007,154,148

2,067,368,773

2,129,389,836

2,193,271,531

2,259,069,677

2014

2015

2016

2017

2018

2019

2020

2021

61,107,840

62,941,075

64,829,307

66,774,187

68,777,412

70,840,735

72,965,957

75,154,935

77,409,583

59,328,000
16,000
$3,600
57,600,000
178,035,840

61,107,840

62,941,075

64,829,307

66,774,187

68,777,412

70,840,735

72,965,957

75,154,935

59,328,000
183,376,915

61,107,840
188,878,223

62,941,075
194,544,569

64,829,307
200,380,906

66,774,187
206,392,334

68,777,412
212,584,104

70,840,735
218,961,627

72,965,957
225,530,476

2013

2014

2015

2016

2017

2018

2019

2020

2021

(74,160,000)
(5,000)
$24,000
(120,000,000)

(76,384,800)

(78,676,344)

(81,036,634)

(83,467,733)

(85,971,765)

(88,550,918)

(91,207,446)

(93,943,669)

(123,600,000)
(5,000)
$24,000
(120,000,000)

(127,308,000)

(131,127,240)

(135,061,057)

(139,112,889)

(143,286,276)

(147,584,864)

(152,012,410)

(123,600,000)
(5,000)
$24,000
(120,000,000)

(127,308,000)

(131,127,240)

(135,061,057)

(139,112,889)

(143,286,276)

(147,584,864)

(123,600,000)
(2,000)
$24,000
(48,000,000)

(127,308,000)

(131,127,240)

(135,061,057)

(139,112,889)

(143,286,276)

1,836,830,378

(49,440,000)

(50,923,200)

(52,450,896)

(54,024,423)

(55,645,156)

(72,000,000)

(194,160,000)

(319,984,800)

(449,584,344)

(511,071,874)

(526,404,031)

(542,196,151)

(558,462,036)

(575,215,897)

(592,472,374)

57,600,000
1,680,960,000
1,738,560,000
6.5%

44,928,000
1,731,388,800
1,776,316,800
2.2%

(16,124,160)
1,783,330,464
1,767,206,304
-0.5%

(136,607,885)
1,836,830,378
1,700,222,493
-3.8%

(260,706,121)
1,891,935,289
1,631,229,168
-4.1%

(316,527,305)
1,948,693,348
1,632,166,043
0.1%

(326,023,124)
2,007,154,148
1,681,131,024
3.0%

(335,803,818)
2,067,368,773
1,731,564,955
3.0%

(345,877,932)
2,129,389,836
1,783,511,904
3.0%

(356,254,270)
2,193,271,531
1,837,017,261
3.0%

(366,941,898)
2,259,069,677
1,892,127,779
3.0%

3.4%
0.0%
3.4%

6.8%
-4.2%
2.6%

10.0%
-10.9%
-0.9%

10.0%
-17.4%
-7.4%

10.0%
-23.8%
-13.8%

10.0%
-26.2%
-16.2%

10.0%
-26.2%
-16.2%

10.0%
-26.2%
-16.2%

10.0%
-26.2%
-16.2%

10.0%
-26.2%
-16.2%

10.0%
-26.2%
-16.2%

Accretion
Dilution
Net

Source: Company reports and J.P. Morgan estimates.

Most of Sprints iDEN customers


will naturally migrate off in the
next 2-4 years

Who are Sprints 11 million remaining iDEN customers?


As of 3Q10, Sprint Nextel reported almost 11 million customers that rely on the
iDEN network with 6.1 million postpaid iDEN customers, 440,000 Power Source
customers (CDMA/iDEN), and 4.5 million prepaid subscribers. The most important
segment of the iDEN customer base is the postpaid customers, comprising roughly
50% enterprise and SMB subscribers and 50% government and public safety-related
customers. The company lost 383,000 postpaid iDEN customers in 3Q10 and we
expect its base to continue to dwindle, but Sprint needs to minimize this loss both
before and during the network transition.
We estimate that by the end of 2014 the company will be down to ~2 million
postpaid iDEN (and no Powersource) customers, which it can then start migrating
over to the next-generation Q-CHAT service on CDMA. Prepaid iDEN subscribers
as well should be quickly migrated over from iDEN due to normally high churn and
handset replacement rates.

Table 35: iDEN Subscriber Model


Beginning iDEN subscribers
+ iDEN gross adds
- Disconnects
+ Acquisition subscribers
Ending iDEN subscribers
Implied net additional iDEN subcribers
iDEN churn
PowerSource subs
PowerSource net adds
iDEN legacy
iDEN legacy net adds
% change y/y

2002
8,666
4,346
(2,400)
0
10,612
1,946
2.1%

2003
10,612
4,507
(2,237)
0
12,882
2,270
1.6%

Source: Company reports and J.P. Morgan estimates.

84

2004
12,882
4,790
(2,625)
0
15,047
2,165
1.6%

2005
15,047
5,015
(3,304)
0
16,758
1,711
1.7%
0
0
16,758
16,758

2006
16,758
3,428
(2,647)
2,092
17,600
781
1.2%
100
100
17,500
742

2007
17,600
(125)
(4,404)
0
13,246
(4,529)
2.4%
1,400
1,300
11,846
(5,654)
-32.3%

2008
13,246
(8)
(3,629)
0
9,609
(3,637)
2.7%
1,400
0
8,209
(3,637)
-30.7%

2,009
9,609
141
(2,495)
0
7,255
(2,354)
2.5%
725
(675)
6,530
(1,679)
-20.5%

2010E
7,255
196
(1,765)
0
5,642
(1,569)
2.3%
339
(386)
5,303
(1,227)
-18.8%

2011E
5,642
543
(1,368)
0
4,817
(825)
2.2%
0
(339)
4,817
(486)
-9.2%

2012E
4,817
426
(1,126)
0
4,117
(700)
2.1%
0
0
4,117
(700)
-14.5%

2013E
4,117
362
(962)
0
3,517
(600)
2.1%
0
0
3,517
(600)
-14.6%

2014E
3,517
(803)
(697)
0
2,017
(1,500)
2.1%
0
0
2,017
(1,500)
-42.6%

2015E
2,017
(1,068)
(332)
0
617
(1,400)
2.1%
0
0
617
(1,400)
-69.4%

2016E
617
(581)
(36)
0
0
(617)
2.1%
0
0
0
(617)
-100.0%

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

What is required before Sprint can shut down the iDEN network?
Sprint is currently nearing completion of the spectrum swap with public safety in the
800 MHz band, after which it will have 14 MHz of contiguous spectrum in that band.
While the pre-swap spectrum was good only for the iDEN technology, this new
spectrum is excellent, paired spectrum that can be used for standard wireless
technologies like CDMA or LTE. The company has already reserved some of this
spectrum for CDMA service, and is testing the CDMA radios in the field today.
Table 36: Spectrum Swap Changes for Legacy Nextel Position
Pre-swap Spectrum Holdings
Final Settlement

700 MHz
4 Contiguous
0

800 MHz
18.5: 10 Contiguous
14 Contiguous

900 MHz
4 Non-contiguous
4 Non-contiguous

1.9 GHz
0
10 Contiguous

Source: Company reports and J.P. Morgan estimates.

The laws of physics dictate that, all else being equal, a signal on a lower frequency
will travel further and penetrate buildings better than a higher-frequency signal.
Sprints service in marginal areas is disadvantaged by its use of 1900 MHz spectrum
vs. the 850 MHz that AT&T and Verizon use, and the company spends ~$1 billion
annually on roaming with Verizon to augment its coverage. While unlimited capital
spending could fix this problem, that is unfeasible, and we believe that only by
deploying CDMA in the 800 MHz band can Sprint improve its coverage offering, as
well as penetrate deeper into buildings to improve in-home quality. That this
possibility is finally close to coming true, five years after the merger of Sprint and
Nextel, could be the help that Sprint needs to move into the ranks of high-quality
carriers, in our view.
Figure 42: Illustration of Signal Propagation at 800 MHz and 1900 MHz

Source: J.P. Morgan.

This CDMA at 800 MHz rollout will be part of Sprints Network Vision plan that
will start in 2011 and end in the 2014 time frame, but likely covering the majority of
the country by 2013. At the same time, Sprint will begin to address its handset lineup
and start offering CDMA handsets with wider 800 MHz capability. Sprints devices
can look at the 800 MHz cellular frequencies today where Verizon runs its CDMA
network, but new devices will need to be retuned to access the Nextel spectrum as
well. We expect Sprint to start selling these devices into its base in the second half of
2011.
85

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

CDMA at 800 MHz should boost


Sprints network quality
dramatically

Note that CDMA handsets with wider 800 MHz capability offered in 2011 and 2012
are likely to have better CDMA coverage, but arent likely to offer the nextgeneration Q-CHAT technology until the CDMA at 800 MHz overlay is mostly
complete. Once that is mostly done in 2014 we estimate Sprint Nextel could still
have 2-3 million postpaid iDEN customers and begin an aggressive handset
migration effort to get them off the network by the end of 2015. As a result, we do
not see a full iDEN network shutdown until at least 2016, but the company should be
able to remove capacity sites starting in 2012.
Finally, Sprint Nextel will have to decide whether to let leases expire or pay the
termination fee. Our checks in the tower industry have not indicated any holding off
of re-signing iDEN tower leases, and we estimate the average iDEN tower still has 57 years remaining on its lease. Alternatively, Sprint could choose to pay to terminate
the leases, which is typically the NPV of the remaining lease payments.

International Expansion Offers Growth, with Some More


Risk
Some tower companies are looking to international markets for growth as the US
market continues to mature. With fewer opportunities to build towers in the US and
recent small deals commanding multiples of roughly 20x tower cash flow, American
Tower and SBA Communications are turning to international markets for higher
rates of return. Crown Castle has a sizable Australian portfolio, but has not
completed any significant international tower transactions recently.
Build or buy? Here or there? A tower company has four options to expand its
tower portfolio, build in the US or internationally or buy in the US or internationally.
Below we have provided a framework as an example of how a tower company could
analyze a build-versus-buy decision in the US and internationally. Given our
assumptions, a tower company would see the highest returns from either building or
buying internationally with domestic returns being 700-900bps lower.
Table 37: Potential Return Scenarios for Domestic and International Towers
Acquisition multiple
Starting tenant leases
Annual Lease-Up
Tenant target
Ending Annualized Revenue
Operating Cost
Tower Cash Flow
Tower Cash Flow Margin
Avg. New Build or Acquisition Cost per Tower
Monthly Rent per Tenant
Escalator
Terminal Multiple
Implied Leveraged IRR

US Tower
NA
1.0
0.2
2.5
$45,000
11,000
34,000
76%
$287,500
$1,500
3.0%
15.0x
21%

Acquired US Tower
20x
2.5
0.1
3.5
$62,607
11,000
51,607
82%
$680,000
$1,500
3.0%
15.0x
14%

International Tower
NA
1.0
0.2
2.5
$33,750
5,500
28,250
84%
$150,000
$1,125
3.0%
12.5x
28%

Acquired International Tower


15x
1.5
0.2
3.0
$40,658
5,500
35,158
86%
$221,250
$1,125
3.0%
12.5x
23%

Note: We assume 50% leverage at a 6% interest rate.


Source: Company reports and J.P. Morgan estimates.

US build-versus-buy assumptions
New tower build opportunities remain limited, but we estimate an attractive
IRR. We assume a new tower costs $250,000-325,000 to develop and the tower
starts with a single anchor tenant. We model an average lease-up rate of 0.2 per year
at a final 2.5 tenancy at $1,500 in monthly rent per tenant. Our tower operating cost
estimate of $11,000 a year ($917/month) remains relatively flat regardless of
tenancy. The ending new tower cash flow margin would then be 76%. Given our
86

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

assumptions, we model an IRR of 21% for a new tower build with 50% financing at
a 6% interest rate.
Recent domestic deals have
been fairly small, but also
expensive given high demand,
limited supply

Domestic deals have occurred at ~20x TCF. The market for smaller private tower
acquisitions has returned to robust levels and asking prices are high. We estimate a
20x purchase price of tower cash flow or $680,000 for a mature tower with a tenancy
of 2.5, and assume that increases to 3.5 over ten years. With the same general
assumptions of a new tower build, we estimate an IRR of 14%.
The higher price and lower lease-up potential of an acquired tower make the build
decision a better one for a tower company, given our assumptions. Unfortunately, the
opportunity for attractive US tower builds continues to shrink, leaving acquisitions as
the alternative domestically.
Incremental tenants increase the return on a tower. We assume one (1.0) incremental
tenant per tower in our base case. For a 2.0 incremental tenant lease-up scenario, our
IRRs go to 27% and 15% from 21% and 14%, respectively, in our build vs. acquire
scenarios.

Table 38: Potential Return Scenarios for Building or Acquiring Domestic Tower
Acquisition multiple
Starting tenant leases
Annual Lease-Up
Tenant target
Ending Annualized Revenue
Operating Cost
Tower Cash Flow
Tower Cash Flow Margin
Avg. New Build or Acquisition Cost per Tower
Monthly Rent per Tenant
Escalator
Terminal Multiple
Implied Leveraged IRR (1)

US Tower
NA
1.0
0.2
2.5
$45,000
11,000
34,000
76%
$287,500
$1,500
3.0%
15.0x
21%

Base Case
Acquired US Tower
20x
2.5
0.1
3.5
$62,607
11,000
51,607
82%
$680,000
$1,500
3.0%
15.0x
14%

Higher Lease-Up
US Tower
Acquired US Tower
NA
20x
1.0
2.5
0.33
0.2
3.5
4.5
$62,550
$80,214
11,000
11,000
51,550
69,214
82%
86%
$287,500
$680,000
$1,500
$1,500
3.0%
3.0%
15.0x
15.0x
27%
15%

Note: We assume 50% leverage at a 6% interest rate.


Source: Company reports and J.P. Morgan estimates.

International offers higher returns to offset higher country risks


The public tower companies expect higher potential returns from international tower
builds or purchases, despite higher risk. New builds are cheaper due to lower soft and
construction costs. In addition, many markets are still developing and need new
tower builds. Collocation in some markets is starting to be adopted by more carriers
and some carriers are selling tower portfolios to focus on core strategies, like most
US carriers. Our analysis sees a substantially higher return potential (difference of
700-900 bps) for international sites than domestic ones.
Opportunities for new tower builds are substantially greater than in the US. We
assume a significantly lower tower development cost, mostly due to lower soft costs
(site acquisition, zoning, etc.) and construction costs. We model an international
tower could only cost $130,000-190,000, or about 50% less than a US tower. In
addition, we assume a lower rent rate of about $1,125 per tenant, 75% of the US
rental rate, and a lower operating cost structure of $5,500 annually ($458/month),
about 50% less than in the US. At an ending tenancy of 2.5, we calculate a tower

87

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

cash flow margin of 84% vs. 76% for a US tower build. Given our assumptions, we
model an IRR of 28% for a new tower build with 50% financing at a 6% interest rate.
The public tower companies see ample opportunities to partner or purchase carrier
tower portfolios internationally, especially in Central and South America, Africa, and
parts of Asia. In some international wireless markets, collocation is still a new
market dynamic and is starting to gain acceptance, providing an opportunity for an
independent third party, such as the US public tower companies, to run a tower
portfolio. For an international acquired tower, we estimate a 15x purchase price of
tower cash flow or $221,250 for a tower. In our international scenario, we assume a
tower is purchased with 1.5 tenants and we model an average lease-up rate of 0.2 to
3.0 ending tenants in ten years. With the same general assumptions of a new tower
build, we estimate an IRR of 23%.
Incremental tenants increase the return on a tower. We assume one (1.0)
incremental tenant per tower in our base case. For a 2.0 incremental tenant lease-up
scenario, our IRRs go to 35% and 26% from 28% and 23%, respectively, in our build
vs. acquire scenarios.
Table 39: Potential Return Scenarios for Building or Acquiring International Tower

Acquisition multiple
Starting tenant leases
Annual Lease-Up
Tenant target
Ending Annualized Revenue
Operating Cost
Tower Cash Flow
Tower Cash Flow Margin
Avg. New Build or Acquisition Cost per Tower
Monthly Rent per Tenant
Escalator
Terminal Multiple
Implied Leveraged IRR

Base Case
International Tower
Acquired International
Tower
NA
15x
1.0
1.5
0.2
0.2
2.5
3.0
$33,750
$40,658
5,500
5,500
28,250
35,158
84%
86%
$150,000
$221,250
$1,125
$1,125
3.0%
3.0%
12.5x
12.5x
28%
23%

Higher Lease-Up
International Tower
Acquired International
Tower
NA
15x
1.0
1.5
0.33
0.3
3.5
4.0
$46,913
$53,863
5,500
5,500
41,413
48,363
88%
90%
$150,000
$221,250
$1,125
$1,125
3.0%
3.0%
12.5x
12.5x
35%
26%

Note: We assume 50% leverage at a 6% interest rate.


Source: Company reports and J.P. Morgan estimates.

We use some broad assumptions as a base example, knowing actual scenarios in each
country around the world could be dramatically different from our assumptions. In
addition, we are not providing a framework that includes demographics, political
environment, regulations, wireless market structure, and other country-specific issues
that could impact wireless tower returns.
American Tower has most aggressive international expansion plan
American Tower now has 28% of its wireless towers in Mexico, Brazil, Chile, Peru,
Colombia, and India. In addition, the company has agreed to purchase towers from
Cell C in South Africa and announced a joint venture with MTN Group in Ghana
called TowerCo Ghana (51% ownership) to purchase MGN Ghanas existing towers.
Both deals are expected to close in early 2011. All in all, total international tower
count for AMT could near 50% of total towers by the end of 2011. In 2009,
international site leasing revenue accounted for about 15% of revenue, but we expect
it to be closer to 20% in 2011.
88

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 40: American Tower Global Tower Portfolio and Credit Metrics by Country
US
International
Mexico
Brazil
Chile
Peru
Colombia
India
South Africa
Ghana

Towers
20,333
12,347
2,619
1,659
113
131
225
7,600
1,400*
1,876*

S&P Debt Rating


AAA

5-Yr CDS Spread


41

BBB
BBBA+
BBBBBBBBBBBB+
B

113
110
85
113
112
NA
125
NA

* Denotes pending acquisition.


Note: American Tower has 199 broadcast towers in Mexico and 2 DAS networks.
Source: Company reports, Bloomberg, and J.P. Morgan estimates.

Crown Castles international assets


Crown Castle has 1,595 towers, or about 7% of its tower portfolio, in Australia. The
sites generate about 5% of Crown Castles site revenue. The company has not started
further international expansion at this time. The majority of Crown Castles
Australian towers were purchased in 2000 and 2001 from Optus and Vodafone,
respectively. On average, the revenue per tower is $57,162 vs. $74,337 in the US and
the gross margin is 68.2% vs. 73.7%.
Table 41: Crown Castle Global Tower Portfolio and Credit Metrics by Country
US
Australia

Towers
22,265
1,595

S&P Debt Rating


AAA
AAA

5-Yr CDS Spread


41
49

Source: Company reports, Bloomberg, and J.P. Morgan estimates.

SBA so far has been cautious


about international expansion

SBAs international assets


SBA Communications has international assets mainly in Panama and Canada. The
company is also starting very small operations in El Salvador and Costa Rica. SBAs
international strategy is to focus on more mature and developed markets in which
development and zoning restrictions enable a stable, predictable return. The company
acquired Jade Tower in May 2009, giving SBA a foothold in Canada with 52 owned
towers and 360 managed communications sites. In Panama, SBA has a backlog of
290 potential tower acquisitions and it plans on building 100 towers in Costa Rica
with others in Canada and Central America.
Table 42: SBA Global Tower Portfolio and Credit Metrics by Country
US
International
Canada
Costa Rica
El Salvador
Panama

Towers
8,538
167
52
4
10
101

S&P Debt Rating


AAA

5-Yr CDS Spread


41

AAA
BB
BB
BBB-

NA
NA
NA
99

Source: Company reports, Bloomberg, and J.P. Morgan estimates.

89

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

REIT Conversion Impact


The tower operators, as has long been discussed, are prime candidates for converting
their organizational structures to real estate investments trusts (REITs). We anticipate
that some tower companies will opt for REIT structures over the next few years at
varying times, once available net operating losses are fully utilized and they are in a
position to pay corporate cash taxes. Though they are enablers of wireless and
broadcast services, the tower companies are essentially owners of real estate assets,
including the physical towers and increasingly the land beneath the towers. A REIT
structure likely would shield the companies from paying most if not all corporate
taxes.

AMT should be the first tower


company to run through its NOLs

SBA likely will not work through


its NOLs for ten-plus years

AMT is expected to convert in 2012; valuation starting to reflect change


We expect American Tower to be the first tower company to opt to convert to a Real
Estate Investment Trust (REIT) structure to lower taxes as its NOL balance runs out,
which we expect to happen in early 2012, and the company has said it is exploring
the option. The company could hold a vote at its annual shareholder meeting in May
2011 for actual conversion in 2012 or beyond. The J.P. Morgan REIT coverage
universe is trading at 20.8x on a price-to-2011E AFFO basis versus about 20.0x for
the tower companies on a price-to-2011E RFCF basis. American Tower has the
biggest potential downside risk if it does not convert to a REIT due to its potential
tax liabilities. We value AMT at $60 per share with REIT conversion versus $45
without conversion.
CCI could convert in 2017; SBAC probably avoids REIT status
Crown Castle is unlikely to run out of its NOLs until after 2016 and we think Crown
Castle could then convert into a REIT. Our valuation for the company is $55 with
REIT conversion versus $54 without conversion due to the minimal net impact of
out-year tax changes. Finally, we do not forecast SBA running out of its NOL
position before 2020 and expect no change in status for the foreseeable future.
Below we address concerns over REIT requirements, potential challenges to
pursuing investments, potential changes to investor bases, valuation, and possible
delays to conversion.
REIT requirements
In order for a corporation to qualify for REIT classification, the company must meet
a variety of criteria. Below we highlight some key requirements:
Taxes. Perhaps most attractive about the REIT structure, no less than 90% of
taxable income must be distributed as dividends to equity holders. Any remaining
taxable income would face standard corporate taxes. With the remaining 10%, a
company would have to consider the tax consequences in any decision to invest
(at a taxed level) rather than return cash to shareholders untaxed.
Ownership. The shareholder base of a REIT must consist of at least 100 different
equity holders. In addition, 50% of the equity cannot be held among five owners
or less. American Towers top five owners control about 23% of shares
outstanding.
Real estate tests. At least 75% of gross income (defined as revenue and passive
income) as well as company assets must be attributed to property or interest on
loans for property. Income from other businesses, unrelated to real estate, cannot

90

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

represent more than 5% of total revenue, though passive income such as


dividends and interest on bank deposits can contribute up to 20%. However,
REITs can hold ownership in taxable subsidiaries, which are excluded from
standard REIT rules, though the value of such subsidiaries cannot represent more
than 25% of the REITs total asset value. In other words, the majority of the
companys assets must fall under a REIT classification, while a minority could be
held within a taxable subsidiary.
Treatment of ancillary businesses. We believe the tower operators would likely
elect to place a portion of their assets in a taxable subsidiary, within the 25%
threshold. For example, the tower operators are involved in ancillary businesses
that do not derive income directly from their real estate assets. The network
development services, network services, and site development businesses for
American Tower, Crown Castle, and SBA, respectively, primarily consist of
consulting and construction services, though this varies among each operator.
Each of these businesses represents 3-14% of consolidated revenue for the
operators.
Table 43: Services Business Contribution
$ in millions

AMT

CCI

SBAC

2010 Revenue
Site Leasing/Rental Revenue
Site Development/Services Revenue
Total 2010 Revenue

$1,924.7
51.7
1,976.4

$1,697.9
173.2
1,871.1

$535.4
88.8
624.3

Revenue Breakdown
Site Leasing/Rental Revenue
Site Development/Services Revenue
Total

97%
3%
100%

91%
9%
100%

86%
14%
100%

Source: Company reports and J.P. Morgan estimates.

In addition, a portion of the companies international operations may be candidates


for placement in a taxable subsidiary. We estimate American Towers international
operations represented approximately 15% of site leasing revenue in 2010, though
this is expected to grow in the future, while 5% of Crown Castles site leasing
revenue was generated outside the United States.
Table 44: International Business Contribution
$ in millions

2010 Site Leasing Revenue


US
International
Total 2010 Site Leasing Revenue
Breakdown
US
International
Total

AMT

CCI

SBAC

$1,636.0
288.7
1,924.7

$1,607.4
90.5
1,697.9

$530.1
5.4
535.4

85%
15%
100%

95%
5%
100%

99%
1%
100%

Source: Company reports and J.P. Morgan estimates.

91

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Payout ratio limits capital for growth, but high cash flow more than enough
The primary reason to opt for a REIT structure is the tax efficiency it provides.
REITs are required to distribute 90% of taxable income as dividends to shareholders,
limiting taxes payable. However, this could also potentially impact cash available to
pursue growth opportunities or reinvest in the business, which we consider to be the
primary drawback of converting to a REIT. For the towers specifically, we do not
believe this is a significant concern. Current taxable income is limited by heavy
levels of depreciation, driven by the size of the companies tower real estate holdings
which are depreciated over 15-20 years However, on a cash flow basis, we believe
there is ample cushion to support current and, if necessary, elevated capital spending
levels.
Figure 43: Capital Spending as % of Depreciation Tower Capex Continues to Run Below Depreciation
100%
80%
60%
40%
20%
0%
2007

2008
American Tow er

2009
Crow n Castle

2010E

SBA Communications

Source: Company reports and J.P. Morgan estimates.

We estimate that American Tower, Crown Castle, and SBA Communications might
only allocate 20-50% of cash flow after dividends towards capital expenditures in
2010. After including acquisitions, both Crown Castle and SBA have substantial cash
flow, though American Towers investments exceed estimated cash flow. However,
both Crown and SBA participated in fewer M&A deals through 3Q10, relative to
typical levels, as they addressed leverage challenges. American Tower has kept
leverage at the low end of target levels, limiting excess capital. It is unclear to us how
leverage would be handled as a REIT. It is possible that rating agencies could be
more stringent compared to today, given that the companies would be forced to pay
dividends, driving less of a cash cushion.

92

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 45: Investment Activity Compared to Available Cash Flow


$ in millions

AMT
$587.2
528.5
90%

CCI
$90.3
81.2
90%

SBAC
($127.7)
0.0
0%

58.7
22.3
38%

9.0
3.4
38%

0.0
0.0
0%

Investable Net Income


Depreciation and amortization
Stock-based compensation
Cash Flow after Dividends
Estimated 2010 Maintenance Capital Expenditures
Investable Cash Flow
Estimated 2010 Discretionary Capital Expenditures
Excess Cash Flow
Estimated 2010 Acquisitions
Cash Flow after Acquisitions

36.4
463.2
56.1
555.8
(42.9)
512.9
(271.9)
241.0
(584.3)
(343.2)

5.6
550.1
36.7
592.4
(21.7)
570.7
(197.9)
372.8
(127.0)
518.2

0.0
276.2
10.3
286.5
(10.0)
276.4
(62.1)
214.3
(281.5)
200.8

Capital Spend % of Cash Flow after Dividends


Estimated 2010 Maintenance Capital Expenditures
Estimated 2010 Capital Expenditures
Estimated 2010 Acquisitions
Total % of Cash Flow after Dividends

7.7%
48.9%
105.1%
161.8%

3.7%
33.4%
21.4%
58.5%

3.5%
21.7%
98.3%
123.4%

Adjusted 2010 Taxable Income


Minimum Dividend Distribution
% of Taxable Income
Remaining Taxable Income
Corporate Taxes
Tax Rate

Note: We do not expect SBAC to generate positive normalized taxable income in 2010.
Source: Company reports and J.P. Morgan estimates.

Investor base implications


Should the tower equities convert to REIT structures, there could be a shift in the
investor base. The potential investor base could actually expand, as the companies
join REIT indices. In fact, American Tower and Crown Castle could be two of the
larger REITs in the market, with market capitalizations of approximately
$19.5 billion and $12.2 billion, respectively, representing approximately 10% of the
MSCI US REIT Index on a pro forma basis.
However, the stocks in general could also experience some volatility as non-REIT
funds may be forced to sell holdings. Unfortunately, it is tough to gauge the potential
risk or upside, as disclosures of top holders do not specify whether ownership is held
by a REIT fund or telecom fund within a larger family of funds.
There are already 15 REITS in
the S&P 500; AMT is bigger than
14 of them

For American Tower specifically, its membership in the S&P 500 is unlikely to be
affected, in our view. Given it already has membership and it would not be
undergoing any significant restructuring or change in its business model, we
anticipate it likely would be kept in the index. However, future admission into the
S&P 500 as REITs could be more difficult for Crown Castle or SBA, as additional
REIT representation could be offered to larger REIT operators instead.

93

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 46: REITs in the S&P 500 Index


REIT
Apartment Investment & Management Co
AvalonBay Communities Inc
Boston Properties Inc
Equity Residential
HCP Inc
Health Care REIT Inc
Host Hotels & Resorts Inc
Kimco Realty Corp
Plum Creek Timber Co Inc
ProLogis
Public Storage
Simon Property Group Inc
Ventas Inc
Vornado Realty Trust
Weyerhaeuser Co

Weighting in S&P 500


2.7%
8.5%
10.6%
13.1%
11.7%
6.0%
10.4%
6.4%
5.3%
7.2%
12.4%
25.6%
7.3%
11.9%
8.9%

Source: Company reports and J.P. Morgan estimates.

Other scenarios
A variety of factors could slow the process of the tower companies converting to
REITS. The tower operators could pursue additional acquisitions, which could
increase net operating losses, delaying the need for conversion. In addition, if
leverage were to increase, we could see a larger drag from interest expense,
increasing losses. Finally, should leasing revenue growth slow more than expected, it
would take longer for the companies to consume their NOLs.
Valuation as a REIT
We believe many investors already compare the tower equities to REITS, using
recurring cash flow as a benchmark for adjusted funds from operations (AFFO). We
calculate recurring cash flow as Adjusted EBITDA less the total of net interest
expense, cash taxes, preferred dividends, and maintenance capital expenditures, as a
standardized measure of cash flow across the tower operators.
REIT-specific cash flow measures, such as funds from operations (FFO) and adjusted
funds from operations (AFFO), are similar. FFO is calculated by taking net income
and adding depreciation and amortization, but deducting preferred dividends. AFFO
adjusts FFO by subtracting straight-line rent and maintenance capital expenditures.
We cannot include the impact of straight-line rent in our recurring cash flow
calculation as all of the tower operators do not provide the data.
When comparing tower valuations to REIT valuations, the tower equities continue to
appear very attractive. REITS are trading at 20.8x on a price-to- 2011E AFFO basis
using J.P. Morgans REIT team estimates, while the tower stocks are trading at 20.0x
recurring cash flow (net of straight-line impacts). However, we expect tower
recurring cash flow to grow 8.9% while AFFO growth for the REITS is expected to
be 8.4%.

94

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 47: Tower Valuation Summary


$ in millions

AMT
$50.50
387
$19,540
4,184
23,724

CCI
$42.60
286
$12,192
5,531
17,723

SBAC
$39.66
114
$4,540
2,853
7,393

$36,272
12,568
48,840

13.8%
16.2%
6.8%

15.0%
8.3%
7.7%

13.7%
11.3%
9.9%

14.3%
12.4%
7.5%

EV / Adj. EBITDA - 2010E


EV / Adj. EBITDA - 2011E
EV / Adj. EBITDA - 2012E

17.6x
15.2x
14.2x

15.2x
14.0x
13.0x

19.2x
17.3x
15.7x

16.9x
15.0x
14.0x

Recurring Cash Flow - 2010E


Recurring Cash Flow - 2011E
Recurring Cash Flow - 2012E

946
1,026
1,164

602
638
761

228
269
309

1,775
1,933
2,235

Recurring Cash Flow Y/Y Growth - 2010E


Recurring Cash Flow Y/Y Growth - 2011E
Recurring Cash Flow Y/Y Growth - 2012E

12.3%
8.4%
13.5%

17.9%
6.1%
19.3%

12.2%
18.0%
15.0%

14.1%
8.9%
15.6%

20.7x
19.0x
16.8x

20.3x
19.1x
16.0x

19.9x
16.9x
14.7x

23.1x
20.0x
16.7x

Recent Price
Shares outstanding
Market Cap- Equity
Year-end Net Debt (Cash)
Enterprise Value
Adj. EBITDA Y/Y Growth - 2010E
Adj. EBITDA Y/Y Growth - 2011E
Adj. EBITDA Y/Y Growth - 2012E

Price / Recurring Cash Flow - 2010E


Price / Recurring Cash Flow - 2011E
Price / Recurring Cash Flow - 2012E

Group

Source: Company reports and J.P. Morgan estimates.


Note: Pricing as of 1/7/2011. Share count and net debt 2011E.

Table 48: REIT Industry Summary Valuation


Property Type
Health Care
Industrial
Lodging
Manufactured Housing
Multifamily
Office
Regional Mall
Self Storage
Strip Center
Triple Net Lease
REIT Industry Weighted
Average

EV/
EBITDA
16.5x
17.1x
23.2x
15.4x
20.3x
15.4x
16.1x
17.6x
15.8x
12.9x
16.9x

P/FFO
2011E
2012E
14.6x
13.4x
20.9x
18.9x
19.0x
14.4x
13.5x
12.6x
20.3x
18.5x
14.6x
13.7x
15.0x
14.3x
18.0x
16.8x
15.9x
15.1x
12.7x
12.6x

P/AFFO
2011E
2012E
15.9x
15.1x
30.9x
26.8x
24.1x
19.6x
15.3x
14.2x
24.6x
21.9x
21.8x
20.5x
18.4x
17.5x
19.7x
18.3x
20.1x
19.0x
13.2x
12.9x

2010E Growth
FFO
AFFO
3.9%
4.2%
11.8% (35.6%)
6.7%
13.6%
0.0%
0.3%
(0.9%)
(3.2%)
6.7%
0.6%
(5.3%)
(9.6%)
(0.0%)
(2.0%)
(1.8%) (18.2%)
0.4%
(5.9%)

2011E Growth
FFO
AFFO
11.9%
12.4%
10.9%
22.7%
24.1%
29.8%
9.6%
10.1%
8.9%
9.6%
3.3%
2.4%
9.6%
7.2%
7.9%
9.7%
4.3%
0.0%
5.3%
6.1%

2012E Growth
FFO
AFFO
8.7%
5.7%
10.0%
15.4%
34.2%
41.7%
6.6%
7.7%
10.2%
12.4%
7.0%
6.6%
4.8%
4.7%
7.3%
7.7%
4.6%
5.7%
0.6%
2.5%

16.3x

20.8x

1.8%

8.3%

8.5%

15.0x

19.2x

(4.3%)

8.4%

9.2%

Source: FactSet, SNL Financial, and J.P. Morgan REIT team estimates. As of 01/10/2010.

High Leverage, but Manageable Debt Structures


During the financial crisis, the tower companies were faced with not being able to
raise debt at almost any rate as near-term maturities weighed on the companies. The
highly leveraged, high-multiple tower stocks fell 51-72% vs. the S&P 500 down
38%.

95

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 44: Stock Prices for the Public Tower Companies from 2007 to 2010
60%
SBAC, 48%

40%

AMT, 35%
CCI, 36%

20%

0%
S&P 500, -11%

-20%

-40%

-60%

AMT

Source: Company reports and J.P. Morgan.

96

CCI

SBAC

S&P 500

11/3/2010

9/3/2010

7/3/2010

5/3/2010

3/3/2010

1/3/2010

11/3/2009

9/3/2009

7/3/2009

5/3/2009

3/3/2009

1/3/2009

11/3/2008

9/3/2008

7/3/2008

5/3/2008

3/3/2008

1/3/2008

11/3/2007

9/3/2007

7/3/2007

5/3/2007

3/3/2007

1/3/2007

-80%

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 45: Tower Company Trailing-Twelve-Month Leverage from 2007 to 2010


12.0x

10.0x

8.0x

6.0x

4.0x

2.0x

0.0x
1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

American Tow er

4Q08

1Q09

Crow n Castle

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10E

SBA Communications

Source: Company reports and J.P. Morgan estimates.

While the tower companies remain leveraged at 4-7x, all the companies have taken
advantage of a friendly debt financing environment and have raised a total of
$6.3 billion of fixed-rate debt with an average rate of 4.9% since the start of 2010. In
addition, maturities were extended and the first tranches are not due until 2014 and
some do not mature until 2020. Each company has laddered out debt schedules so an
overbearing amount of principal does not come due in one year.
Figure 46: Tower Company Debt Maturity Schedule
$ millions

4000
3000
2000
1000
0
2010

2011

2012

2013

2014
AMT

2015
CCI

2016

2017

2018

2019

2020

SBAC

Source: Company reports and J.P. Morgan estimates.

97

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Of the three public tower companies, American Tower or Crown Castle could
increase leverage some, in our view, to either purchase towers more aggressively or
buy back more stock. SBA is at about 7x levered already and we do not think
management wants to increased leverage to 8-10x unless a transformative tower deal
was available. In our view, American Tower has the ability to lever up by 1 or 2
multiple points to 5-6x net debt/EBITDA from 4x today, which translates into $23 billion or more of additional debt capacity with no acquired EBITDA. At current
prices, this could allow AMT to repurchase approximately 7-14% of its outstanding
stock. We think Crown Castle could add about one additional turn of leverage to 6x
from about 5x today or about $1.2 billion more debt capacity. At that level, the
company could repurchase about 10% of its outstanding stock at current prices.

Wireless Tower Overview


We estimate that there are between 100,000 and 120,000 leaseable towers in the US
today. For purposes of our tower industry model, we use about 112,000 wireless
towers with about 46% being owned by the three public tower companies. AT&T,
T-Mobile, Verizon, and US Cellular have carrier tower portfolios adding up to about
23% of the total. Two large private tower companies are Global Tower and TowerCo
with 4,400 and 3,200 towers, respectively. The remaining towers are generally
owned by mom-and-pop owners. We forecast that the independent tower operators
could build 2,500-3,000 leaseable towers for 3G/4G purposes over the next few
years.
The wireless industry has added 73,593 sites over the past five years. This includes
sites on rooftops, collocations on existing towers, anchor tenants on newly built
towers, and collocations on newly built towers. We have used the Cellular
Telecommunications Industry Associations (CTIAs) published figures (the results
of CTIAs semiannual survey) to track cell sites, which at the end of June 2010
totaled 251,618 cell sites in the United States. These are total cell sites and we
assume 25% are on rooftops and other alternative structures to towers, leaving about
75% on towers. We estimate the average tenant per tower in the US at ~1.7.

98

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 49: Tower Industry Model


Towers
1Q09
Crown Castle
22,481
American Tower
14,339
AT&T Towers
9,676
SBA Communications
7,805
T-Mobile Towers
5,820
Verizon Wireless Towers
5,570
US Cellular Towers
4,210
Global Tower Partners
3,650
TowerCo
3,120
Mobilite LLC
2,761
Other
22,500
Total
101,932
y/y % change
3.8%
q/q % change
0.5%
Market share
Crown Castle
American Tower
AT&T Towers
SBA Communications
T-Mobile Towers
Verizon Wireless Towers
US Cellular Towers
Global Tower Partners
TowerCo
Mobilite LLC
Total

1Q09
22.1%
14.1%
9.5%
7.7%
5.7%
5.5%
4.1%
3.6%
3.1%
2.7%
100.0%

2Q09
22,425
14,339
9,726
7,924
5,855
5,575
4,225
3,750
3,140
2,766
22,800
102,525
3.2%
0.6%

3Q09
22,385
16,039
9,776
8,004
5,890
5,580
4,240
3,850
3,160
2,771
23,100
104,795
4.3%
2.2%

4Q09
22,365
18,039
9,826
8,241
5,925
5,585
4,255
3,950
3,180
2,776
23,400
107,542
6.1%
2.6%

2009
22,365
18,039
9,826
8,241
5,925
5,585
4,255
3,950
3,180
2,776
23,400
107,542
6.1%
6.1%

1Q10
22,338
20,039
9,876
8,296
5,955
5,590
4,270
4,050
3,190
2,781
23,700
110,085
8.0%
2.4%

2Q10
22,321
20,162
9,926
8,502
5,985
5,595
4,285
4,150
3,202
2,786
24,000
110,914
8.2%
0.8%

3Q10
22,265
20,333
9,976
8,618
6,000
5,600
4,300
4,400
3,217
2,791
24,150
111,650
6.5%
0.7%

4Q10E
22,280
20,383
10,026
8,742
6,015
5,605
4,315
4,450
3,232
2,796
24,300
112,144
4.3%
0.4%

2010E
22,280
20,383
10,026
8,742
6,015
5,605
4,315
4,450
3,232
2,796
24,300
112,144
4.3%
4.3%

2Q09
21.9%
14.0%
9.5%
7.7%
5.7%
5.4%
4.1%
3.7%
3.1%
2.7%
100.0%

3Q09
21.4%
15.3%
9.3%
7.6%
5.6%
5.3%
4.0%
3.7%
3.0%
2.6%
100.0%

4Q09
20.8%
16.8%
9.1%
7.7%
5.5%
5.2%
4.0%
3.7%
3.0%
2.6%
100.0%

2009
20.8%
16.8%
9.1%
7.7%
5.5%
5.2%
4.0%
3.7%
3.0%
2.6%
100.0%

1Q10
20.3%
18.2%
9.0%
7.5%
5.4%
5.1%
3.9%
3.7%
2.9%
2.5%
100.0%

2Q10
20.1%
18.2%
8.9%
7.7%
5.4%
5.0%
3.9%
3.7%
2.9%
2.5%
100.0%

3Q10
19.9%
18.2%
8.9%
7.7%
5.4%
5.0%
3.9%
3.9%
2.9%
2.5%
100.0%

4Q10E
19.9%
18.2%
8.9%
7.8%
5.4%
5.0%
3.8%
4.0%
2.9%
2.5%
100.0%

2010E
19.9%
18.2%
8.9%
7.8%
5.4%
5.0%
3.8%
4.0%
2.9%
2.5%
100.0%

Source: Company reports and J.P. Morgan estimates.

99

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 50: Cell Sites by Tower Company


CTIA
244,021 245,912
Sites on towers
183,016 184,434
Tenants
1.80
1.80

246,497
184,872
1.76

247,081
185,311
1.72

247,081
185,311
1.72

249,350
187,012
1.70

251,618
188,714
1.70

245,818
184,364
1.65

248,018
186,014
1.66

248,018
186,014
1.66

Cell sites by tower compan


1Q09
Crown Castle
59,799
American Tower
35,977
AT&T Towers
9,676
SBA Communications
19,289
T-Mobile Towers
7,566
Verizon Wireless Towers
5,570
US Cellular Towers
4,210
Global Tower Partners
4,745
TowerCo
4,368
Mobilite LLC
2,761
Other
29,055
Total
183,016

2Q09
59,987
35,145
9,726
19,750
7,612
5,575
4,225
4,875
4,553
2,766
30,221
184,434

3Q09
60,216
38,486
9,776
20,010
7,657
5,580
4,240
5,005
4,740
2,771
26,392
184,872

4Q09
60,497
42,789
9,826
20,602
7,703
5,585
4,255
5,135
4,929
2,776
21,215
185,311

2009
60,497
42,789
9,826
20,602
7,703
5,585
4,255
5,135
4,929
2,776
21,215
185,311

1Q10
60,759
48,094
9,876
20,741
7,742
5,590
4,270
5,265
5,104
2,781
16,791
187,012

2Q10
61,048
48,389
9,926
21,255
7,781
5,595
4,285
5,395
5,283
2,786
16,971
188,714

3Q10
61,229
46,766
9,976
21,545
7,800
5,600
4,300
5,720
5,469
2,791
13,168
184,364

4Q10E
61,604
46,881
10,026
21,854
7,820
5,605
4,315
5,785
5,818
2,796
13,510
186,014

2010E
61,604
46,881
10,026
21,854
7,820
5,605
4,315
5,785
5,818
2,796
13,510
186,014

2Q09
32.5%
19.1%
5.3%
10.7%
4.1%
3.0%
2.3%
2.6%
2.5%
1.5%
16.4%
100.0%

3Q09
32.6%
20.8%
5.3%
10.8%
4.1%
3.0%
2.3%
2.7%
2.6%
1.5%
14.3%
100.0%

4Q09
32.6%
23.1%
5.3%
11.1%
4.2%
3.0%
2.3%
2.8%
2.7%
1.5%
11.4%
100.0%

2009
32.6%
23.1%
5.3%
11.1%
4.2%
3.0%
2.3%
2.8%
2.7%
1.5%
11.4%
100.0%

1Q10
32.5%
25.7%
5.3%
11.1%
4.1%
3.0%
2.3%
2.8%
2.7%
1.5%
9.0%
100.0%

2Q10
32.3%
25.6%
5.3%
11.3%
4.1%
3.0%
2.3%
2.9%
2.8%
1.5%
9.0%
100.0%

3Q10
33.2%
25.4%
5.4%
11.7%
4.2%
3.0%
2.3%
3.1%
3.0%
1.5%
7.1%
100.0%

4Q10E
33.1%
25.2%
5.4%
11.7%
4.2%
3.0%
2.3%
3.1%
3.1%
1.5%
7.3%
100.0%

2010E
33.1%
25.2%
5.4%
11.7%
4.2%
3.0%
2.3%
3.1%
3.1%
1.5%
7.3%
100.0%

Market share
Crown Castle
American Tower
AT&T Towers
SBA Communications
T-Mobile Towers
Verizon Wireless Towers
US Cellular Towers
Global Tower Partners
TowerCo
Mobilite LLC
Other
Total

1Q09
32.7%
19.7%
5.3%
10.5%
4.1%
3.0%
2.3%
2.6%
2.4%
1.5%
15.9%
100.0%

Source: Company reports and J.P. Morgan estimates.

Positive characteristics of investing in tower companies


Tower ownership offers investors a scalable model with long-term contracted
revenues with built-in escalators from a diverse, creditworthy customer base and
effectively fixed direct operating costs and overhead costs. Tower companies provide
a mission-critical service to the operation of wireless networks, so churn rates are
very low and generally not even reported. We estimate industry churn rates at about
1% per annum (not to be confused with monthly churn rates) which are typically
driven today by broadcast and paging, not wireless, companies exiting. As a result,
we expect these companies to experience solid top-line growth and expanding cash
flow margins over the next several years. Almost all capital expenditures for tower
assets are discretionary and immediately provide incremental operating cash flow.
Towers ride the wireless growth
wave, without the competition

100

The tower sector is often compared to other industries that generally are better
understood by investors, including broadcasting, outdoor advertising, and real estate.
We maintain that the fundamental characteristics of the tower industry are superior to
those of all of the above-mentioned sectors. Below we compare these characteristics
with those of the wireless telephony business.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 51: Wireless Towers vs. Telephony


Characteristic

Towers

Wireless Telephony

Pricing

Increasing pricing for new tenants. Declining per-minute pricing model.


Existing tenant leases have built-in
price escalators.

Technology

Technology-agnostic. The greater


number of technologies used, the
better.

Technology choice creates risk of


obsolescence and requires
upgrades.

Churn

Less than 1% per annum.

20%-25% per annum.

Surety of Revenues

Based on long-term contracts


(average five years with multiple
automatic renewals).

Revenues generally based on


usage.

Potential EBITDA margins

70%-80%.

30%-40%.

Customer Base

National and regional wireless


providers. Service is critical to
wireless carriers network
operation.

Businesses with pricing power.

Access to capital/critical mass.

Multiple competitors licensed in


each market.

Barriers to Entry

Zoning/regulation.

Individuals.

Source: Company reports and J.P. Morgan estimates.

With few exceptions, the stock prices of public independent tower operators have
tracked each other fairly closely. In our opinion, part of the reason for this is that the
market is not distinguishing the different approaches and growth strategies that each
company has employed to take advantage of the tower ownership model. Key items
that differentiate these businesses include critical mass to achieve economies of
scale, liquidity position and access to capital, percentage of revenues and cash flow
from site leasing, tower build component of tower addition strategy, and sources of
acquired towers. We believe that all three public tower companies will benefit from
estimated solid demand for tower space for at least the next few years.

History of Towers
The original tower owners
The original owners of towers were mostly wireless network operators and
broadcasters. These companies viewed towers as a capital cost and an integral,
strategic part of their networks. Accordingly, carriers and broadcasters generally did
not allow competitors to collocate transmission equipment on their towers. The
mindset was that the carriers and broadcasters would provide their rivals with a cost
advantage by allowing them to avoid the cost of zoning, developing, and constructing
towers. Although collocation was a potential incremental source of revenue, it fell
below most carriers and broadcasters radar screens. With the introduction of several
new wireless competitors to each market throughout the US (mostly PCS licensees),
network operators, broadcasters, and others recognized the monetization opportunity
in their towers. In addition, zoning of new sites became increasingly difficult as
communities fought the proliferation of towers. This served to raise the value of
existing towers with excess capacity by forcing new wireless entrants and existing
carriers expanding their footprints to devise creative ways to add sites for both
network capacity and geographical coverage.

101

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Enter the independent tower operators


In the mid- and late 1990s, several entrepreneurs recognized the financial opportunity
in owning wireless transmission towers. Although it is a capital-intensive business,
they saw the potential demand for tower space, the potential for high margins, the
recurring revenue model, the pricing power, and the formidable barriers to entry.
These entrepreneurs had varied backgrounds, including wireless, broadcasting, cable,
real estate, and technology. In 1999 and 2000 alone, the five public independent
operators raised, in the aggregate, more than $13 billion in new capital and capital
commitments, by our estimates. These companies have raised capital from just about
every source available, including bank debt, high-yield debt, convertible debt,
preferred securities, private equity, and public equity. Even in difficult markets, they
have demonstrated the ability to secure significant amounts of capital. Each of the
three public companies has approached the tower opportunity in a different way, but
all have adhered to the same general guiding concepts that tower ownership is crucial
and that the best existing sites cannot be replicated.
Carrier deals transform the tower sector
Although conversations between wireless carriers and independent tower operators
about the transfer of tower ownership had been going on for years before, in 1999
and 2000 the carrier tower transactions became a reality. In that period, Bell Atlantic,
BellSouth, SBC, Powertel, Nextel, GTE, AirTouch, Triton PCS, Dobson Cellular,
ALLTEL, and other carriers all announced deals to sell their towers. These deals
transformed the tower industry. Wireless carriers recognized that tower ownership
and management was not a core competency and was, therefore, a good outsourcing
opportunity. The tower monetization relieved the carriers of tower operating costs
and provided capital to invest in their core business adding subscribers and driving
minutes of use through their networks.
Previously, the tower industry had been considered too small-cap for institutional
investors to consider. Suddenly, these companies were raising hundreds of millions
of dollars and garnering multi-billion-dollar valuations in the public market.
Management teams were (and continue to be) generally stretched to focus on
capitalizing on the land grab opportunity for the best towers in the country.
We maintain that many of the towers originally built and operated by wireless
carriers are among the best towers an independent tower operator can own. We
particularly like carrier towers for three main reasons.
1. These towers tend to be well-clustered around major markets and are often in
locations where the construction of a new tower would be virtually impossible.
2. These towers tend to be overbuilt. The capacity of towers built for cellular
networks was usually in excess of what was needed. As a result, these towers
generally have excess capacity and the ability to add multiple incremental
broadband tenants with no or minimal incremental capital expenditure.
3. Carrier towers tend to be underutilized because, under the carriers watch,
collocation attempts by competing wireless carriers often were ignored, if not
rebuffed. There is usually pent-up demand for space on carrier towers when they
are transferred to an independent tower operator, so, when a major carriers
portfolio of towers is sold to an independent tower operator, the tower company
often experiences strong initial lease-up rates.

102

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Tower companies turn to integration and execution


In 1999 and 2000, most of the public independent tower companies made companyaltering acquisitions and built towers in order to gain critical mass, including several
carrier tower deals and intra-industry combinations. The goal of the major tower
companies from that point was to integrate those acquisitions and demonstrate the
ability to execute on the lease-up potential of existing towers (2000 was supposed to
be the year of execution and integration, but the large carrier tower deals kept
coming). This integration process included hiring senior managers to handle the
added complexity of these companies and field workers to perform due diligence,
site management, and site-related services.
From boom to bust (almost)
The independent tower companies stock prices peaked in 2000 and started their
downward march through 2002, losing over 95% of their equity value from peak to
trough. Furthermore, Pinnacle Holdings and SpectraSite filed for bankruptcy. The
performance largely mirrors that of the dot-com, technology, and telecom companies
of the time. The tower companies suffered from slowing business conditions, heavy
debt burdens, and reduced capital spending from wireless carriers trying to remain
free cash flow positive. While operating performance for the tower companies
continued to improve, they were unable to grow fast enough to support untenable
capital structures. In 2003, the long road to recovery started for the tower companies.
The road to recovery
From 2003 to 2007, the tower companies addressed their once-toxic capital structures
and started to deleverage from double-digit net leverage. The companies underwent a
virtuous cycle of growing EBITDA, reducing interest expense and debt levels, and
reached positive free cash flow, which enabled the companies to further improve
their capital structures. As result of both financial recovery and continued operational
improvements, wireless carriers experienced solid growth with follow-on benefits for
the tower companies. By the end of 2007 the stocks of tower companies had come
back to 80-100% of their peak 2000 values with substantially improved balance
sheets, financials, and operational fundamentals.
Back from the brink of the financial crisis, and stronger too
The financial crisis of 2008 closed the credit markets to the tower companies, which
had substantial leverage and looming debt maturities. The fear of debt default
gripped investors and the tower stocks fell 40-75% from peaks to their 2008 troughs
vs. the S&P 500 which declined as much as 50% in the same period. American
Tower preformed the best in part due to its lower leverage. As the government
stepped in and credit markets improved, the tower companies started to recover and
Crown Castle was one of the first companies to raise new debt and re-open the credit
markets in 2009. Promisingly, through the decline the tower companies continued to
see relatively stable business trends, despite some pullback in builds from smaller
carriers.
The stock prices of the tower companies recovered to peak 2008 prices by early 2010
due to excitement about carrier health, LTE amendments at Verizon, and the
potential for new carriers like Clearwire and Lightsquared. The tower companies
have been able to take advantage of favorable credit markets and they refinanced
$6.3 billion of debt in 2010 at an average fixed interest rate of 4.9%. In addition,
maturities were extended and the first tranches are not due until 2015 and some do
not mature until 2020. Each company has laddered out debt schedules so an
103

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

overbearing amount of principal does not come due in one year. Finally, the tower
companies have started to look to international markets for incremental growth
opportunities as the domestic wireless market continues to mature. The tower
companies have been able to weather the financial crisis storm and now have come
out stronger on the other end.

Basic Tower Economics


Towers are a key component of any wireless communications network. Every
terrestrial wireless system consists of a network of cell sites on which antennae and
other electronic equipment are placed. Because these antennae must be elevated in
order for their signal to propagate to provide coverage, the majority of wireless
transmission antennae are located on existing structures such as rooftops. However,
if there is no available structure high enough that an antenna can be attached to it
within a particular cell, a tower must be erected to provide the requisite elevation.
The need for incremental sites to provide both coverage and capacity and the difficult
zoning environments are the two main drivers of demand for collocation and,
therefore, the value of wireless transmission towers. Competition among wireless
service providers appears to have migrated from pricing to the quality of coverage, a
situation that has greatly benefited the independent tower operators. As wireless
telephony subscriber penetration has accelerated, demand for cell sites and, thus,
towers has exploded. Wireless service providers are not only playing catch-up to
provide competitive coverage for todays voice and circuit-switched data offerings,
but are also trying to achieve the necessary cell site density to accommodate
continued subscriber growth and higher data speeds as 3G and 4G networks are
planned and developed. In addition, all cell sites require zoning approval from local
municipalities, and communities throughout the country have been fighting the
proliferation of towers, often delaying or blocking zoning approval.
A wireless carriers top network development priority is to deploy its sites in as
quick and cost-efficient a manner as possible, and the only two ways to provide
coverage are either to get new cell sites zoned or to rent space, or collocate, on
existing sites. Collocation provides faster network development time and significant
capital cost savings and is therefore almost always preferable.
Sources of revenue
A towers primary source of revenue is rental income from the leasing of tower space
to wireless service providers and broadcasters onto which those companies attach
antennae. A tower operators goal, therefore, is to lease up its towers by adding as
many collocation tenants onto each tower as possible to maximize leasing revenue.
The market for lessees is vast because potential tenants include any user of wireless
spectrum, including cellular, PCS, paging, wireless data, radio dispatch, radio and
TV broadcasting, and government and private networks. Lessees pay a fixed monthly
rent regardless of technology, coverage, or minutes of use. Commercial tenant leases
average around five years in duration and usually have multiple automatic renewals
and periodic rate escalations. Rental rates differ from location to location and from
market to market based on demand for the site and the local zoning environment.
The rule of thumb for rental rates for a wireless carrier tenant (cellular, PCS, etc.) is
generally $2,000 per month, although tougher zoning environments and carriers
104

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

urgent need to add sites has boosted the average rent for some towers and markets.
Typical rental rates for the increasingly scarce narrowband tenants (paging) average
around $500 per month, and radio and TV broadcasters rents generally range from
$5,000 to $10,000 per month. Lease rates for new wireless technology providers vary
significantly, but are often priced favorably for the tower owner relative to a
broadband equivalent on a lease rate-to-load ratio. Lease rates usually escalate on an
annual basis at a fixed rate of about 3-5% or at a variable rate based on an index such
as the CPI.
All three of the public independent tower operators also provide wireless network
development services, most of which are associated with the development of new
towers or the modification of existing antenna configurations. These services include
RF engineering, site acquisition, site development, site construction, and line and
antenna installation and modification. These services provide lower gross margins
than the site leasing business, but offer incremental cash flow.
We believe there are significant potential operating synergies from both owning
wireless transmission towers and providing these network development services. For
instance, providing site acquisition services to a wireless carrier can enhance the
companys relationship with the carrier and the companys knowledge of the
carriers network build plan, potentially driving lease-up of the companys owned
towers. Conversely, ownership of a multi-tenant tower on which an incremental
carrier wants to collocate gives the company leverage to provide and charge for line
and antenna installation service.
Network development revenues should continue to grow
Just a couple of years ago, the common view of network development services was
that it was a fee-based, project-type business that would shrink and possibly
disappear over time as the PCS buildout matured and the BTS model became more
prevalent. At any point in time over the past ten years it was assumed that network
development services revenue would diminish substantially in the following three to
five years. Our view is that, although network development revenues are not as
consistent as site leasing revenues, the independent tower operators will continue to
have significant and growing network development revenues. We expect these
revenues to rise as the tower companies expand their tower footprints and further
solidify their relationships with the major wireless carriers. As wireless networks
become denser and more complex, the need for incremental equipment modifications
at the tower will increase. Nonetheless, we believe that network development
services will become a smaller percentage of total revenues, as site leasing revenues
are expected to grow at much higher rates than network development revenues.
Critical mass provides economies of scale
Another fundamental aspect of the tower ownership model is that the direct operating
costs for a tower are low and essentially fixed, regardless of the number of tenants on
the tower. Although tower operating costs vary, they average around $10,000-12,000
per year per site (assuming the tower company leases the land). Substantially all
incremental revenue from the collocation of additional tenants on a tower, therefore,
falls to the tower cash flow line. Key tower operating expenses include the ground
lease, site maintenance, insurance, and utilities, with the cost of the ground lease
usually making up two-thirds or more of the total. Typically, the independent tower
owner does not own the swatch of land on which the tower stands, although all of the

105

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

major independent tower operators are slowly acquiring the land beneath their towers
or extending lease lengths if that makes more sense.
The table below demonstrates the ability to leverage the fixed nature of direct tower
operating expenses and the capacity to accommodate multiple tenants to drive tower
cash flow margins (tower leasing revenues less direct cash tower costs) to as high as
70-85%.
Table 52: Tower Unit Economics

Acquisition multiple
Starting tenant leases
Annual Lease-Up
Tenant target
Ending Annualized Revenue
Operating Cost
Tower Cash Flow
Tower Cash Flow Margin
Avg. New Build or Acquisition Cost per Tower
Monthly Rent per Tenant
Escalator
Terminal Multiple
Implied Leveraged IRR (1)

Anchor Tenant
Only
NA
1.0
0
1.0
$18,000
11,000
7,000
39%
$287,500
$1,500
3.0%
15.0x
3%

Low
Lease-Up
NA
1.0
0.133
2.0
$35,955
11,000
24,955
69%
$287,500
$1,500
3.0%
15.0x
17%

Mature
Tower
NA
1.0
0.266
3.0
$53,910
11,000
42,910
80%
$287,500
$1,500
3.0%
15.0x
24%

Note: We assume 50% leverage at a 6% interest rate.


Source: Company reports and J.P. Morgan estimates.

Similar economies also are achievable across a portfolio of towers because the
marginal overhead expenses related to adding a new tower to a portfolio are minimal.
The consolidation of the industry over the past few years has been driven by the goal
of gaining critical mass to attain economies of scale in operating costs and cost of
capital. Overhead expenses of the major tower operators have increased over the past
few years as these companies have invested in core internal systems, personnel, and
regional offices to accommodate their ongoing rapid growth. We expect these
expenses to continue to grow modestly as the tower operators go through this highgrowth phase and as large acquisitions, both domestic and international, are
integrated into existing systems. In addition, we anticipate that incremental overhead
expenses will be required as these companies augment their network services
businesses.
Tower ownership is critical
The model we have described is exclusively for owned towers, not site management.
Independent tower operators often maintain and manage for collocation towers and
other sites they do not own. In these cases, tower operators share revenues with
property owners, but also avoid the operating costs related to site ownership. The
property owners, which are often commercial building owners renting space on the
rooftop, generally take the lions share of the revenue. Although the revenue split is
negotiated for each site, the site manager generally only gets about 15-40% of the
lease revenue. As a result, the operating cash flow margins to the site manager are
generally much lower, and the revenue growth opportunity for managed sites is
limited. Still, the major tower consolidators frequently enter into tower and rooftop
management agreements with tower and real estate owners for two main reasons:

106

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

1. Adding managed towers to a tower portfolio of owned sites in a market can


increase the marketability and, as a consequence, the lease-up potential for the
entire market footprint, including owned sites.
2. Adding managed sites to a tower portfolio can provide incremental cash flow
with limited upfront capital costs for the sites.
Tower development
Building towers generally is the most capital-efficient way to accumulate towers,
allowing independent tower operators to lower blended capital costs per tower
addition. Below is a breakdown of the capital costs to zone, develop, and construct a
typical multi-tenant, self-supporting lattice tower:
Table 53: Tower Development Costs of a Typical Multi-Tenant Tower
Soft Costs (site acquisition, zoning, etc.)
Materials
Construction
Total Development Costs

US
$50,000-70,000
$60,000-80,000
$140,000-175,000
$250,000-325,000

Source: Company reports and J.P. Morgan estimates.

The tower company should have the requisite knowledge of the local wireless market
and existing coverage by local wireless carriers as well as relationships with wireless
carriers to know whether they would consider collocating on a given tower and, if so,
the timing of those collocations. Meticulous due diligence is required to measure the
potential lease-up of any tower. The greatest risk in the tower business is building a
tower and not being able to lease it up. It is those companies with superior market
knowledge and relationships with wireless carriers that will achieve the highest
returns, in our view. The combination of tower companies avoiding towers in
locations they believe to be uneconomical and negligible churn rates leads us to
believe that the likelihood of an overbuild of towers in the US is very low.

107

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

108

North America Equity Research


13 January 2011

North America Equity Research


13 January 2011

Large Cap Integrated Telecom Operators

Large Cap Integrated Telecom Operators

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

109

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

110

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

AT&T Inc.
Initiate with Overweight Rating and $33 YE11 Price Target
We initiate coverage on AT&T with an Overweight rating and year-end 2011 price
target of $33, implying 14% upside from current levels in addition to an estimated
6.0% dividend yield. In our view, the companys low EPS and free cash flow
multiples and continued improvements in the wireline business as well as wireless
margins offset the risk of a wireless slowdown as AT&T loses exclusivity on the
iPhone in February. In addition to an attractive yield, investors, we believe, could see
more upside in AT&T shares relative to Verizons, which have outperformed in
recent months.

Overweight
AT&T Inc. (T;T US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

28.04
12 Jan 11
30.10 23.78
166.50
Dec
5,938
33.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY

2009A

2010E

2011E

0.53
0.54
0.54
0.51
2.12

0.59A
0.61A
0.55A
0.55A
2.29A

0.62
0.65
0.62
0.61
2.50

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Note: EPS figures based on adjust ed results.

Key Investment Points


Strong cash flow should continue in 2011
AT&T trades at 11.5x our 2011 EPS estimate compared to the S&P 500 and Verizon
multiples of 14.9x and 16.1x, respectively, and 11.8x 2011E free cash flow or a yield
of 8.5%. Our 2011 EPS estimate of $2.50 represents 9.2% growth over 2010E EPS
almost all of it organic and assumes 0.4% consolidated revenue growth and slight
expansion in margins at the wireless and wireline levels.
Wireless opportunity remains despite iPhone challenges
We believe that the major reason for AT&Ts current discount to Verizon is the
companys pending loss of exclusivity for the iconic Apple iPhone. By now though,
in our view, most of the potential price impact is already reflected in AT&T stock
and most investors expect AT&T to lose some amount of postpaid wireless
customers in 2011. Our model assumes that AT&T sees an increase in churn, a drop
in gross add share, and an overall gain of 100K postpaid subs in 2011 vs. an
estimated increase of 2.3 million in 2010. Despite this, we look for wireless revenue
growth of 4% (vs. 9% in 2010E) as postpaid ARPU grows and prepaid subscribers
increase with iPad sales. We note that even with an incremental gain or loss of one
million iPhone customers compared to our assumptions, our revenue forecasts do not
exhibit a large amount of sensitivity. Finally, we assume that wireless margins
increase to 42.6% in 2011 from 41.0% in 2010 due to fewer gross adds and handset
subsidies as well as scaling.

111

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Video rollout, network data demand help to stabilize consumer and enterprise
wireline revenue from here
Revenue in AT&Ts wireline business declined 6.4% in 2009 but is tracking down
only -3.0% by 3Q10. We expect 2011 revenue to decrease 2.4%, but conservatively
model 30 basis points of margin expansion so EBITDA should fall only -1.6%.
While consumer access lines continue to fall at an 11%+ pace, the primary drivers of
AT&Ts revenue stabilization have been the companys success in consumer
broadband and video (consumer revenue began to grow again in 3Q10) and
stabilization in business. Business revenue losses peaked at 8.4% y/y in 3Q09 and
contracted to 3.9% by 3Q10.
Buyback is 0.8% accretive to 2011E EPS; assume another 0.7% in 2012E,
despite spectrum acquisition
AT&T authorized an $8 billion buyback program in December 2010 and we expect
the company to exercise enough of it to maintain leverage at a level of 1.4x
EBITDA. Even with the $1.9 billion spectrum acquisition that AT&T also
announced in December, we expect the company to repurchase $2.5 billion in stock
in 2011 and $4.0 billion in 2012.

Investment Risks
iPhone exclusivity loss comes at same time that network is getting poor
customer satisfaction scores
AT&T will lose its exclusive relationship with Apple for selling the iPhone in early
February. This iconic device was part of what made AT&T one of the top two
wireless carriers for growth in the last four years, and its loss could allow some
iPhone loyalists to leave AT&T and cause others to not subscribe to its service. In
addition, the loss comes at a time when AT&T is under pressure for its poor network
quality (real or perceived) and customer satisfaction (as measured by Consumer
Reports), so there could be a bigger impact to customer stats than we model.
Enterprise and small business revenue remains heavily exposed to any economic
weakness
AT&T has seen the benefit of the economic stabilization in the last year through
slowing revenue losses in its enterprise and small/medium-sized business divisions.
While we do not model a significant economic recovery, if the economy starts to
fade again or unemployment ramps back up, AT&T could see a re-acceleration in
revenue losses and margin compression.
Additional spectrum acquisition is possible; could slow buyback
AT&T recently acquired spectrum from Qualcomm for $1.9 billion, which we expect
to close in 2011. There are more blocks of similar spectrum held by smaller entities
that AT&T could choose to purchase if it becomes available. In addition, we expect
the FCC to eventually find more spectrum to auction, and that AT&T would likely
participate.

112

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Valuation and Price Target Methodology


We are establishing a year-end 2011 price target of $33/share for AT&T. We based
our target price primarily on a discounted cash flow analysis that assumes a 7.5%
WACC and a 0.5% perpetual growth rate. We also give AT&T credit for its
approximately 9% stake in America Movil (AMX, rated Overweight by J.P. Morgan
Latin American Telecom analyst Andre Baggio) and 10% stake in Telmex (TMX,
rated Underweight by J.P. Morgan Latin American Telecom analyst Andre Baggio).
AT&T shares also look attractive on an EV/EBITDA basis. Shares currently trade at
5.1x our 2011 estimate, compared to Verizon at 5.9x. We show our full discounted
cash flow analysis following our earnings and cash flow forecast.

Company Description
AT&T is the largest telecommunications company in the US and one of the largest in
the world, offering wireless and wireline voice, data, and video to consumers and
businesses. The company is the result of the combinations in the last ten years of
AT&T, SBC, BellSouth, AT&T Wireless, and Cingular. It supports over 92 million
wireless customers and its wireline network passes over 52 million homes in the US.
Both businesses cater to consumer and enterprise customers. Of approximately
$124 billion in revenue, wireless and wireline each contribute 48%, with the wireless
piece growing faster than wireline business.
Figure 47: AT&T Revenue Breakdown

Mass Markets,

Figure 48: AT&T EBITDA Breakdown

Adv ertising and

Adv ertising,

Publishing, 3%

Publishing, and

Other, 1%

Other, 4%

1%

Business

Wireless

Solutions, 30%

Serv ice, 43%

Wireline, 48%
Consumer,

Wireless

17%

Equipment, 5%

Source: 3Q10 Company reports.

Wireless, 50%

Source: 3Q10 Company reports.

113

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless Growth Stutter Likely in First Half of 2011


We estimate AT&T added
8.6 million wireless subscribers
in 2010

AT&T Mobility is the second-largest wireless carrier in the US by subscribers and


service revenue, and reported approximately 41% EBITDA margins in the last year.
While the companys growth has been very strong in the last few years, we expect
2011 to come with slower subscriber growth but higher margins than 2010 due to the
loss of iPhone exclusivity.
Figure 49: Wireless Subscriber Breakdown
Connected
Dev ices,
8%
Reseller,
12%
Prepaid,
7%

Postpaid,
73%
Source: 3Q10 Company reports.

Figure 50: AT&T Voice Coverage Map

Source: Company website (used with permission).

114

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Loss of iPhone exclusivity expected to drive subscriber slowdown


Verizon and Apple recently confirmed that the iPhone will be available to Verizon
customers in early February and no longer exlusive to AT&T. Verizon will begin
selling a CDMA version in early February.
While management has noted that it expects to keep churn low even after the
exclusivity periods end, as 80% of its contracted customers are on family plans or
business discount plans, we model an increase in postpaid churn to 1.3% in 1Q11,
1.4% in 2Q11, and 1.3% for full-year 2011, from 1.1% in 3Q10.
In addition, we estimate that the iPhone has enabled AT&T to capture substantially
more gross add share than it would have without the device. As Verizon and
potentially other US carriers start selling the iPhone, we expect AT&Ts share of
postpaid gross additions to decline to 28.5% in 2011 from 30.7% in 3Q10 and a peak
of 33.8% in 3Q09.
Table 54: Big 4 Postpaid Gross Add Share and Churn Trends
1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10E

1Q11E

2Q11E

3Q11E

4Q11E

Gross add share


AT&T Mobility
Verizon Wireless
Sprint Nextel
T-Mobile USA

30.6%
36.5%
12.2%
20.8%

32.7%
35.8%
12.2%
19.3%

33.8%
34.6%
13.8%
17.8%

30.3%
36.1%
16.3%
17.3%

29.5%
33.6%
18.1%
18.7%

28.0%
32.9%
18.2%
21.0%

30.7%
32.4%
18.2%
18.8%

29.9%
32.1%
19.4%
18.7%

27.9%
33.8%
18.7%
19.5%

27.4%
33.7%
19.7%
19.1%

29.4%
32.5%
19.2%
18.9%

29.4%
31.4%
19.7%
19.5%

Churn
AT&T Mobility
Verizon Wireless
Sprint Nextel
T-Mobile USA

1.15%
1.13%
2.25%
2.30%

1.07%
1.00%
2.05%
2.18%

1.14%
1.12%
2.17%
2.44%

1.15%
1.05%
2.11%
2.30%

1.07%
1.05%
2.15%
2.20%

1.01%
0.93%
1.85%
2.20%

1.14%
1.07%
1.93%
2.40%

1.15%
1.00%
1.85%
2.20%

1.30%
0.98%
1.75%
2.20%

1.40%
0.95%
1.75%
2.20%

1.25%
1.00%
1.75%
2.20%

1.20%
1.00%
1.75%
2.20%

Source: Company reports and J.P. Morgan estimates.

Margins expected to widen somewhat in 2011 as iPhone sales slow


When the iPhone 3G was released in June 2008, the terms of Apples contract with
AT&T changed to allow AT&T to subsidize the device for customers signing twoyear contracts. As a result of the revised structure, iPhone prices dropped to the sweet
spot of handset sales and sales jumped upwards, but AT&Ts wireless service
margins took a large hit, falling to just over 35% from 43% the previous quarter.
AT&T aggressively upgraded
customers, and extended
contracts, in 3Q10

As the high-ARPU iPhone base has scaled, margins have generally recovered to over
40.0%, though results ebb and flow depending upon iPhone volumes in any given
quarter and typically are lowest when a new iPhone is released usually in the
third quarter. For example, in 3Q10 the iPhone 4 was launched and in addition to the
typical upgrade cycle, AT&T also more aggressively allowed customers to upgrade
to the device and iPhone activations hit 5.2 million in the quarter.

115

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 51: iPhone Sales vs. Wireless Margins


Sales in thousands

50%

6,000
5,000

44%

43%

40%

4,000
3,000
2,000

44%

43%
38%

40%

43%

43%

41%
38%

39%

44%
42%

41%

35%

45%
40%
35%

1,000
0

30%
1Q08 2Q08 3Q08 4Q08 1Q09 2Q09

3Q09 4Q09 1Q10 2Q10 3Q10 4Q10E 1Q11E 2Q11E 3Q11E 4Q11E
iPhone sales
Wireless serv ice margin %

Source: Company reports and J.P. Morgan estimates.

AT&T is targeting margins in the low-40% range for full-year 2010, and
management continues to target long-term margins in the mid-40s. Although
pressures from subsidies on escalating smartphone sales, operating expenses related
to LTE build plans, and Alltel-related integration costs are likely to provide some
continued pressure, we believe AT&T should be able to attain the companys longterm margin goals.
AT&T needs LTE on new
spectrum to relieve 3G network
congestion

AT&T has accelerated its 4G plans, bringing LTE forward faster


AT&T recently accelerated its 4G LTE plans, we believe to gain access to the
700 MHz and AWS spectrum that it has acquired and reserved for that technology.
The company had previously guided to a meaningful LTE rollout in 2012 with an
aggressive rollout of HSPA 3G software in the meantime. However, in September
AT&T announced plans to launch 4G service across 75 million pops by the end of
2011, and later said the entire LTE network will be upgraded by the end of 2013. In
addition, the company plans to offer two 4G smartphones in 1Q11, five to seven
devices in 1H11, and 20 devices by year-end. We believe this dramatic change in
strategy was driven by the rapidly increasing data usage on its 3G network and its
need to get customers over to new spectrum bands and a more efficient wireless data
technology.
To date, the company has launched HSPA 7.2 in almost all markets and is now
rolling out HSPA+ (version 14.4), which provides double the peak speeds. While
these speeds are slower than the LTE speeds on Verizons new network, the fact is
that there are few (or no) applications on smartphones for which a user can
differentiate between a 2-5 Mbps stream and a 5-10 Mbps stream, and AT&T doesnt
really compete in the laptop card/MiFi device area for which the speeds will be most
evident.
As part of its plans to roll out LTE quickly, AT&T recently signed amendment
contracts with the two biggest US tower vendors, American Tower and Crown
Castle. These contracts will allow AT&T to add LTE and new frequency gear to its
existing towers across the AMT and CCI footprints, rather than having to negotiate
each lease amendment on a one-off basis.

116

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

AT&T supports every wireless


eReader in the US

AT&T so far leads in the emerging devices battle


In addition to traditional handset-based voice and data customers, AT&T has been
aggressively adding other customer types. Devices used by these customers include
tablets, eReaders, alarm monitoring tools, navigation devices, and other automobilebased products. For example, we believe AT&T added approximately 400k 3G
Apple iPads in each of the last two quarters (counted as prepaid subs) at an ARPU of
around $25, and expect an even bigger number in 4Q10. The company has also
activated a significant number of Amazon Kindles (counted as wholesale subs) with
ARPU around $3-4, and currently is the network for every 3G-enabled eReader
selling in the US. Finally, connected devices, such as utility meters and tracking
devices that also generate ARPU of around $3 each, increased by over 4 million in
the last four quarters.
Though the revenue and EBITDA contributions from wholesale and connected
devices are likely to remain modest relative to overall results, we expect this segment
to be a healthy supporter of wireless growth and cash flow.

Table 55: Prepaid, Wholesale, and Connected Device Subscriber Trends


Subscribers in thousands

1Q08
Ending Subscribers
Prepaid
Wholesale
Connected Devices
Net Adds
Prepaid
Wholesale
Connected Devices
Estimated Churn
Prepaid
Wholesale
Connected Devices

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10E

6,085
7,750
1,815

6,139
7,865
2,160

6,112
8,005
2,399

6,106
8,589
2,661

5,961
8,931
2,805

5,558
9,286
3,109

5,386
9,934
3,315

5,350
10,439
4,704

5,377
10,717
5,785

5,881
10,597
6,682

6,209
11,021
7,843

6,659
11,321
9,143

118
280
156

12
109
345

(36)
134
239

(23)
579
263

(155)
337
144

(412)
348
304

(176)
634
235

(58)
484
1,394

24
269
1,052

300
(130)
896

321
406
1,159

450
300
1,300

8.9%
0.5%
0.5%

8.5%
0.5%
0.5%

8.4%
0.5%
0.5%

8.5%
0.5%
0.5%

7.7%
0.5%
0.5%

7.9%
0.5%
0.5%

6.6%
0.5%
0.5%

7.1%
0.5%
0.5%

6.5%
0.5%
0.5%

7.0%
0.5%
0.5%

6.0%
0.4%
0.5%

5.7%
0.3%
0.5%

Source: Company reports and J.P. Morgan estimates.

Wireline
AT&Ts wireline business makes up 48% of the companys consolidated revenue,
and is split between consumer, wholesale, small business, and enterprise businesses.
The company today is the largest provider of local wireline service with
approximately 25 million consumer access lines and the largest global provider of
enterprise telecommunications.

117

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 52: Wireline Revenue Breakdown


%

Mass Markets,
2%

Other, 1%
Consumer,
35%

Business
Solutions, 62%

Source: 3Q10 Company reports.

AT&T and Verizon have an


enterprise duopoly

Enterprise recovery will take time


AT&Ts overall Business Solutions segment continues to face headwinds, driven
primarily by weakness in the global economy. Declining volumes are impacting
voice revenue, while business customers have also generally delayed major new
projects. Customer migrations to lower-cost IP-based services have also had some
impact, but this should be mostly complete and we expect the incremental impact
from here to be small.
All in all, Business Solutions revenue (comprising Enterprise, Wholesale, and Small
Business segments) has declined for the last several quarters. However, the yearover-year reduction has slowed over the past year, and we expect to see further
stabilization in coming quarters. In the third quarter management noted recent new
contracts with orders worth $1 billion over time.

118

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 53: Business Revenue and Growth


Revenue in millions

11,000
10,500
10,000

-4%
-6%

-4%

-4%
-6%

-5%

-6%

9,500

-5%

-4%

-3%

-2%

-3%

-1%

0%
-2%

-4%

-4%

-6%

-6%

-8%

9,000

-8%

8,500

-10%
1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09 1Q10 2Q10 3Q10 4Q10E 1Q11E 2Q11E 3Q11E 4Q11E
AT&T Business Solutions
% change, y -t-y

Source: Company reports and J.P. Morgan estimates.

U-verse gaining scale, with IP now 42% of consumer revenue


After the Business Solutions segment, the Consumer business is the next-largest
contributor to AT&Ts wireline revenue. The Consumer business continues to face
secular challenges, driven by access line disconnects as customers drop wireline
voice connections and instead rely solely on wireless connectivity or move to cable
VoIP lines. In the last year, retail consumer primary lines (including VoIP and legacy
circuit switched) declined by 10.7% only somewhat slower than the 11.7% peak in
1Q09.
Figure 54: Consumer Access Line Trend
Revenue in millions

35,000

-11.4%

30,000

-11.1%

-11.3%

-11.4%

-11.0%
-11.4%

25,000
20,000
15,000

-11.2%
-11.4%

-11.8%

-11.9%

-11.6%
-11.8%

-12.1%

10,000

-12.0%

-12.3%

5,000

-12.2%

-12.4%
4Q08

1Q09

2Q09

3Q09

4Q09

Consumer Access Lines

1Q10

2Q10

3Q10

4Q10E

% change, y -t-y

Source: Company reports and J.P. Morgan estimates.

AT&Ts DSL business has also faced some challenges, particularly in areas in which
the company does not offer U-verse broadband services and is therefore more
exposed to higher-speed cable offers. However, the U-verse business, which relies on
fiber-to-the-node technology to deliver video, broadband, and voice, has become a
significant contributor to the Consumer segment since the service was first rolled out
in late 2006. U-verse and DSL, which represent total wireline consumer IP revenue,
now account more than 42% of total consumer revenue although U-verse broadband
penetration is only approximately 10% vs. the nearly 30% market share we expect
eventually. Management expects U-verse to represent a $5 billion annual revenue
119

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

stream by year-end and to generate positive EBITDA. Given the higher ARPU on
U-verse subscribers, we believe growth in this business will be key to offsetting
some of the secular pressures facing the company.
Figure 55: Broadband Growth Trend
Revenue in millions
600

16%

500
400
300

14%
14%

200

12%
10%

11%

100

8%

8%

(100)

6%
7%
5%

(200)

6%

5%

5%

(300)

4%
4%

(400)
1Q08

2Q08

3Q08

DSL Additions

4Q08

1Q09

2Q09

3Q09

4Q09

U-Verse Additions (estimated)

1Q10

3%

3%

2Q10

3Q10

3%

2%
0%

4Q10E

Broadband % y -t-y change

Source: Company reports and J.P. Morgan estimates.

Industry-best wireline margins, and holding steady


Across large integrated players, AT&T generates the best wireline margins, which
have been consistently in the 31-32% range since 1Q09. This margin level is
approximately 1,000 bps higher than Verizons, though in fairness Verizons
enterprise business started as MCI, which had EBITDA margins of 10% before being
acquired in 2005 vs. far higher legacy margins at AT&T before it was acquired by
SBC.
We expect wireline margins to dip slightly going into the fourth quarter to 31.5%,
from 32.5% in the third quarter, primarily due to seasonal pressures. However, in
2011, we are forecasting margins to expand 30 bps. We believe increasing
contributions from U-verse will be a significant driver. Should employment or
business revenue in the US or globally begin to recover in a meaningful way, it likely
would drive performance above our forecast.

Operating and Financial Outlook


We expect AT&T to experience a sharp slowdown in wireless subscriber growth in
2011, driven by substantially weaker postpaid net additions. We forecast 5.2% total
subscriber growth in 2011 and 4.3% in 2012, with the bulk of net additions coming
from the connected device category. As these low-ARPU customers become a larger
part of the base, overall ARPU is expected to continue falling; postpaid ARPU
should exhibit modest growth as customers continue to migrate to data plans. Our
wireless operating forecasts drive estimated service revenue growth of 5.5% in 2011
and 3.4% in 2012, down from 10.1% in 2010.

120

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

We believe certain wireline trends will improve from 2010, driving a deceleration in
wireline revenue declines. We expect data revenue to exceed voice revenue in 2011,
as U-verse becomes a larger piece of the pie. Improvement in enterprise revenue
growth should also help. As a result, we estimate wireline revenue to decline 2% in
2011 and remain flat in 2012, compared to a 4% decline in 2010.

Earnings and Cash Flow Forecast


As gross additions fall, particularly iPhone adds, AT&Ts wireless margins should
continue to expand toward the mid-40% level over time, in our view. We estimate
160 bps of margin expansion in 2011 to 42.6% and another 30 bps of expansion in
2012 to 42.9%. For wireline, we forecast modest margin expansion in 2011 to just
over 32.2%, as revenue trends improve and more profitable businesses contribute a
greater portion of revenue. We expect margins to gain an incremental 50 bps in 2012,
approaching 33.0%. Our segment forecasts drive consolidated EPS estimates of
$2.50 in 2011 and $2.64 in 2012, representing growth of approximately 9.0% in 2011
and 6% in 2012. Improved margin trends, partially offset by rising wireless capital
spending, should drive robust cash flow. Free cash flow growth of 4.9% in 2011 and
7.4% in 2012 should more than support the companys recently announced share
repurchase program (up to 300 million shares) as well as additional dividend hikes.

121

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 56: AT&T Forecasts and Valuation Analysis


$ in millions
2006

2007

2008

2009

60,962
12.6%
6,892
37.7%
1.8%
$49.09
33,773
10.2%
11,168
24%
33.1%
7,039
20.8%

70,052
14.9%
7,315
6.1%
1.7%
$50.15
38,678
14.5%
14,734
32%
38.1%
3,745
9.7%

77,009
9.9%
6,699
-8.4%
1.7%
$50.40
44,249
14.4%
17,630
20%
39.8%
5,869
13.3%

85,120
10.5%
7,248
8.2%
1.5%
$50.51
48,563
9.7%
19,844
13%
40.9%
5,907
12.2%

30,838
23,484
0%
12,170
23%
3
NM
33,851
70%
11,715

22,811
-2.9%
14,156
27%
231
7600%
41,630
23%
24,075
106%
71,583
5%
25,406
93%
35%
13,972
20%

21,826
-4.3%
15,077
29%
1,045
352%
37,322
-10%
24,415
1%
67,889
-5%
23,848
-6%
35%
13,398
20%

118,928
89%
41,981
108%
35%
21,577
6,253
34%
17,888

$1.89
$1.89
$1.33

$1.95
$1.94
$1.42

NPV of wireless EBITDA 12E -- 20E


NPV of wireless taxes paid 12E -- 20E
NPV of wireless capex 12E -- 20E
NPV of wireless free cash flow 12E -- 20E
NPV of wireless 2020 terminal value @ 3.7x EBITDA
PMV of wireless operations

2011E
201,193
(89,855)
(65,342)
45,996
104,881
150,877

2011E
177,163
(78,843)
(56,602)
41,718
104,881
146,599

NPV of AT&T Inc. EBITDA 12E -- 20E


NPV of AT&T Inc. taxes paid 12E -- 20E
NPV of AT&T Inc. Capex 12E -- 20E
NPV of AT&T Inc. free cash flow 12E -- 20E
NPV of AT&T Inc. 2020 terminal value @ 5.7x EBITDA
PMV of AT&T Inc.

356,643 312,409
(75,309) (67,352)
(140,770) (121,114)
140,563 123,943
150,950 150,950
291,514 274,893

Wireless
Subscribers
% change, y/y
Net additional subscribers
% change, y/y
Churn rate, monthly
ARPU, monthly
Service revenue
% change, y/y
EBITDA (operating cash flow)
% change, y/y
EBITDA margin
Capital Spending
CAPEX/Service revenue
Wireline
Retail consumer access lines
% change, y/y
Retail business access lines
% change, y/y
Total broadband connections
% change, y/y
U-verse video connections
% change, y/y
Voice Revenue
% change, y/y
Data Revenue
% change, y/y
Total Wireline Revenue
% change, y/y
EBITDA (operating cash flow)
% change, y/y
EBITDA margin
Capital Spending
CAPEX/revenue

68,250
13,131
25%
6,499
10%

Consolidated Revenue
% change, y/y
EBITDA
% change, y-t-y
EBITDA margin
Depreciation & Amortization
Taxes
Implied tax rate
Capital expenditures

63,055
20,195
32%
9,907
3,525
32%

Reported EPS - Basic


Reported EPS - Diluted
Dividend Per Share
Private Market Valuation Summary

2010E

2011E

2012E

2013E

2014E

2015E

2016E

2017E

95,336
12.0%
8,635
19.1%
1.3%
$49.48
53,456
10.1%
21,924
10%
41.0%
8,521
15.9%

100,336
5.2%
5,000
-42.1%
1.3%
$48.19
56,386
5.5%
24,030
10%
42.6%
8,740
15.5%

104,669
4.3%
4,333
-13.3%
1.3%
$47.42
58,297
3.4%
25,024
4%
42.9%
8,745
15.0%

108,519
3.7%
3,850
-11.1%
1.3%
$46.95
60,045
3.0%
26,336
5%
43.9%
8,586
14.3%

112,112
3.3%
3,593
-6.7%
1.2%
$46.48
61,521
2.5%
27,234
3%
44.3%
8,613
14.0%

115,078
2.6%
2,966
-17.4%
1.2%
$46.01
62,717
1.9%
28,069
3%
44.8%
8,780
14.0%

117,844
2.4%
2,766
-6.7%
1.1%
$45.55
63,657
1.5%
28,688
2%
45.1%
8,912
14.0%

120,333 122,566
2.1%
1.9%
2,489
2,233
-10.0%
-10.3%
1.1%
1.1%
$45.10
$44.65
64,443
65,064
1.2%
1.0%
28,961
29,137
1%
1%
44.9%
44.8%
9,022
9,109
14.0%
14.0%

27,332
-11.4%
20,106
-7.9%
15,789
30%
2,064
98%
32,324
-13%
25,561
5%
63,514
-6%
20,278
-15%
32%
11,057
17%

24,258
-11.2%
18,763
-6.7%
16,225
31%
2,944
43%
28,308
-12%
27,441
7%
61,186
-4%
19,528
-4%
32%
10,728
18%

22,271
-8.2%
17,835
-4.9%
16,675
32%
3,784
29%
25,531
-10%
28,772
5%
59,578
-3%
19,169
-2%
32%
10,124
17%

21,271
-4.5%
17,515
-1.8%
17,035
32%
4,584
21%
23,570
-8%
29,966
4%
58,810
-1%
19,246
0%
33%
9,410
16%

20,771
-2.4%
17,259
-1.5%
17,395
33%
5,384
17%
21,900
-7%
31,327
5%
58,501
-1%
19,773
3%
34%
9,360
16%

20,471
-3.8%
17,054
-1.2%
17,665
33%
6,184
15%
20,462
-7%
32,510
8%
58,247
-1%
19,979
4%
34%
9,320
16%

20,381
-0.4%
16,890
-1.0%
17,868
34%
6,984
13%
19,204
-6%
33,550
3%
58,029
0%
20,194
1%
35%
9,285
16%

20,341
-0.2%
16,790
-0.6%
18,019
34%
7,784
11%
18,077
-6%
34,575
3%
57,927
0%
20,448
1%
35%
9,268
16%

20,341
0.0%
16,690
-0.6%
18,133
34%
8,184
5%
17,289
-4%
35,437
2%
58,001
0%
20,764
2%
36%
9,280
16%

124,028
4%
42,946
2%
35%
19,883
7,036
35%
19,961

123,018
-1%
41,206
-4%
33%
19,714
6,156
33%
17,151

124,125
1%
42,543
3%
34%
19,483
7,282
35%
19,646

124,628
0%
44,234
4%
35%
19,532
7,957
35%
19,656

125,696
1%
45,253
2%
36%
19,345
8,611
36%
18,946

126,938
1%
47,044
4%
37%
19,139
9,634
37%
18,739

127,963
2%
48,100
6%
38%
19,063
10,122
37%
18,724

128,741
1%
49,107
2%
38%
18,995
10,598
37%
18,857

129,377
0%
49,937
2%
39%
18,953
11,006
37%
18,972

$2.17
$2.16
$1.61

$2.12
$2.12
$1.65

$2.30
$2.29
$1.69

$2.51
$2.50
$1.74

$2.65
$2.64
$1.80

$2.90
$2.88
$1.85

$3.10
$3.08
$1.91

$3.30
$3.28
$1.96

$3.48
$3.46
$2.02

$28.85
$33.21

2009
13.6x
15.7x

2010E
12.6x
14.5x

2011E
11.5x
13.3x

2012E
10.9x
12.6x

227,074
252,980

5.5x
6.1x

5.3x
5.9x

5.1x
5.7x

5.0x
5.6x

Valuation Multiples
Current Price PER
Price Target PER
Current EV/EBITDA
Price Target EV/EBITDA
Wireless Valuation Method:
1
Corporate Valuation Method:
1

Plus (less): investments


Less: net debt, end of period
PMV -- Equity

9,630
(66,073)
235,071

9,630
(65,393)
219,130

Current shares outstanding


Shares to be issued/warrants/options
Fully diluted shares outstanding

5,907
29
5,936

5,909
29
5,938

PMV per share


Private to public discount
Year-end 2011 fair value per share
Plus annual dividend
Total upside to current level

$39.60
10%
$35.64

$36.90
10%
$33.21
$1.74
21.2%

Consolidated Valuation Based on Consolidated Forward P/E Multiple


2009
Consolidated EBITDA
41,206
AT&T Inc. EBITDA
21,362

$33.21
6.5%
7.0%
7.5%
8.0%
8.5%

2011E
44,234
20,204

2012E
45,253
20,229

$2.12
13.6x

$2.29
12.6x

$2.50
11.5x

$2.64
10.9x

Consolidated Free Cash Flow

17,850

14,479

15,359

16,434

AT&T Inc. FCF / diluted share


AT&T Inc. Current FCF yield
Current FCF multiple

$3.01
10.4%
9.6x

$2.44
8.5%
11.8x

$2.61
9.1%
11.0x

$2.85
9.9%
10.1x

AT&T Inc. dividend/share


Dividend yield
Target price dividend yield

$1.65
5.7%
5.0%

$1.69
5.9%
5.1%

$1.74
6.0%
5.3%

$1.80
6.2%
5.4%

Source: Company reports and J.P. Morgan estimates.

122

1 = DCF from above @ 7.5% discount rate (adjust below)


2 = 6.5x forward EBITDA multiple
6.5x

AT&T Valuation Grid - Sensitivity to DCF per Share


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)

2010E
42,543
20,619

Reported EPS, diluted


Consolidated PER

1 = DCF from above @ 7.5% discount rate (adjust below)


2 = 7.2x forward EBITDA multiple
7.2x

-0.5%
36
33
30
28
26

0.0%
38
34
32
29
27

0.5%
40
36
$33.21
30
28

1.0%
43
39
35
32
29

1.5%
46
41
37
34
31

Valuation Grid - year-end value per share


P/E Multiple on Continuing Earnings Per Share
10x
12x
15.0x
16x
18x

2010E
23
27
$34.29
37
41

10x
12x
14.0x
16x
18x

2011E
25
30
$34.99
40
45

2018E

2019E

2020E

124,462
1.5%
1,896
-15.1%
1.1%
$44.31
65,674
0.9%
29,368
1%
44.7%
9,194
14.0%

126,235
1.4%
1,774
-6.4%
1.1%
$43.98
66,150
0.7%
29,481
0%
44.6%
9,261
14.0%

20,341
0.0%
16,690
0.0%
18,219
34%
8,584
5%
16,765
-3%
35,984
2%
58,023
0%
21,062
1%
36%
9,284
16%

20,341
0.0%
16,690
0.0%
18,283
34%
8,784
2%
16,390
-2%
36,490
1%
58,154
0%
21,401
2%
37%
9,305
16%

20,341
0.0%
16,690
0.0%
18,331
35%
8,984
2%
16,182
-1%
36,909
1%
58,365
0%
21,770
2%
37%
9,338
16%

130,114
1%
50,486
1%
39%
18,948
11,301
37%
19,094

130,638
0%
50,922
1%
39%
18,934
11,562
37%
19,185

131,262
0%
51,455
1%
39%
18,943
11,854
37%
19,291

131,837
0%
51,904
1%
39%
18,969
12,112
37%
19,391

$3.64
$3.62
$2.08

$3.79
$3.77
$2.15

$3.96
$3.94
$2.21

$4.12
$4.10
$2.28

CAGR
10E-20E
3%
-15%
-1%
0%
3%
1%

-2%
-1%
1%
12%
-5%
3%
0%
1%
0%
1%
2%

0%
0%
6%
3%

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Verizon Communications
Initiate with Neutral Rating and $38 YE11 Price Target
We initiate coverage on Verizon with a Neutral rating and a year-end 2011 price
target of $38, implying 5.8% upside from current levels in addition to an estimated
dividend yield of 5.4%. In our view, the company already trades at relatively high
multiples of EPS and free cash flow due to upcoming availability of the iPhone on its
network. We believe that the Street already expects strong corresponding subscriber
growth and some offsetting margin compression, and dont foresee the
outperformance likely necessary to drive meaningful incremental upside from current
levels relative to other names under coverage.

Neutral
Verizon Communications (VZ;VZ US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

35.47
12 Jan 11
37.70 24.75
100.38
Dec
2,830
38.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.63
0.63
0.60
0.54
2.40
2.41

0.56A
0.58A
0.56A
0.56A
2.27A
2.21A

0.54
0.50
0.59
0.60
2.23
2.25

Source: Company dat a, Bloomberg, J.P. Morgan estimates. 'Bloomberg' above denotes Bloomberg
consensus estimates .

Key Investment Points


High multiples of cash flow, EPS, and EBITDA
Verizon trades at 16.1x our 2011 EPS estimate compared to S&P 500 and AT&T
multiples of 14.9x and 11.5x, respectively, and 16.5x free cash flow, or a yield of
6.0%. Our 2011 EPS estimate of $2.23 indicates growth of only 4.2% over 2010 and
assumes 3.2% consolidated revenue growth and slight growth in wireline margins
but margin compression at the wireless level due to higher subsidies.
Wireless top line to shine in 2011, but expect margins to compress near term
We believe that the major reason for Verizons current premium to AT&T is the
buildup of expectations that it will sell the iconic Apple iPhone, and that this product
will drive significant outperformance in postpaid adds in 2011 we model
2.3 million postpaid adds for Verizon in 2011. In addition, we and the Street expect
postpaid ARPU to increase as smartphones drive customers to higher-rate plans and
we look for top-line growth of 7% at Verizon Wireless in 2011. Unfortunately
though, the iPhone story has been so heavily reported in the press that we believe the
likely price impact is already baked in, and dont see enough real value being created
by the iPhone to drive incremental multiple expansion from current levels.

123

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wired video rollout drives consumer revenue growth; wholesale and enterprise
stable
Revenue in Verizons wireline business declined 3.6% in the year to 3Q10 but we
expect that to stabilize in the next 12 months or so. The consumer business is already
growing, driven by strong video traction even as access lines continue to drop.
Enterprise is shrinking at a slower rate and should improve with the economy, while
wholesale was down significantly in 3Q10 due to pricing actions but should stabilize
as well on a sequential basis in 2011. Wireline margins also seem to have finally
stabilized, and we model margins of 22.1% in 2011 vs. 21.0% in 3Q10.
Cash flow to improve in 2011 but remain weak given low wireline margins
We estimate Verizon will generate $6.1 billion in cash in 2011 only a 6.0% yield
on the current stock price. At these levels, and compared to AT&T with an estimated
cash flow yield of 8.5%, we dont find Verizon an attractive investment.

Investment Risks
Upside risks
Verizon getting the iPhone could drive a higher share of additions and better
growth than we model
While we already model Verizons churn decreasing and gross add share increasing
once it starts selling the iPhone, the customer impact of the iPhone has often been
underestimated. If churn declines more or gross adds increase more than we expect,
the companys top line could grow faster than we expect.
Enterprise and economic recovery could drive wireline margins higher than we
forecast and increase EPS and possibly the multiple as well
Verizons wireline business is highly exposed to domestic GDP growth and
employment, as well as global economic growth. This business has seen the benefit
of economic stabilization in the last year through slowing revenue losses in enterprise
and small/medium-sized business divisions. While we do not model a significant
economic recovery, if the economy rebounds Verizon could see an acceleration in
revenue growth and margin expansion.
Downside risks
4G handset availability is likely to be weak and prices high through 2011
Verizon has put a large piece of its capital spending and marketing power into
advertising its 4G network, but LTE devices for Verizon remain very limited and we
expect LTE devices to remain limited and expensive through 2011. The higher prices
of LTE handsets, plus the more expensive Apple handsets that we expect to ramp in
1Q11, could impact our wireless margins even more than the 270 basis points that we
already model.
Verizon could pay a premium to purchase parts of Vodafones ownership in
Verizon Wireless
Vodafone owns 45% of Verizon Wireless the most profitable and fastest-growing
of the Verizon entities. We assume that the Verizon/Vodafone relationship remains
static, with no change in Vodafones ownership. However, as Verizon gets closer to
2012 when it has said it may begin paying a wireless dividend, the company may
seek instead to buy Vodafones stake in Verizon Wireless. If Verizon were keen to
effect a transaction, it could pay a significant premium to the current valuation and a
transaction could increase Verizons leverage substantially.
124

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $38/share for Verizon. We based our
target price primarily on a discounted cash flow analysis of the consolidated
company, which assumes a 7.5% WACC and 0.5% perpetual growth rate. We
subtract $96 billion to account for Vodafones 45% stake in Verizon Wireless, but
give Verizon $5 billion of credit for its 23% stake in Italys Omnitel based on 11x
the $440 million annual dividend Verizon received in 2010. In addition, Verizon
appears expensive on an EV/EBITDA basis as well. Shares trade at 5.9x our 2011
estimate, compared to 5.1x for AT&T.

Company Description
Verizon is one of the largest telecommunications companies in the world with assets
that include US landlines (formerly GTE and Bell Atlantic), 55% of the largest US
wireless carrier (Vodafone owns 45%), and a global enterprise business that came
from the MCI acquisition in 2005. The company serves approximately 12.8 million
households of the 26.4 million in its territory with landline service and is currently
selling to 12.1 million of those with its FiOS fiber-optic network, while Verizon
Wireless reports over 93 million wireless customers. Verizon is using its new fiberoptic network to offer high-speed data and video applications to customers. The
companys wireless network is based on CDMA technology, with 3G EVDO Rev A
services. Recently the company began launching a 4G LTE network which currently
covers 130 million people.

125

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 56: Verizon Revenue Breakdown


Global

Figure 57: Verizon EBITDA Breakdown

Other, 1%

Other, 1%

Wholesale,

Wireline, 24%

8%
Global
Enterprise,
15%

Mass

Wireless

Markets, 15%

Serv ice, 53%

Wireless
Equipment

Wireless, 74%

and Other, 8%
Source: 3Q10 Company reports.

Source: 3Q10 Company reports.

Figure 58: Verizon Revenue Breakdown (Excluding Vodafone Stake)


Global

Figure 59: Verizon EBITDA Breakdown (Excluding Vodafone Stake)

Other, 1%
Other, 2%

Wholesale,
11%
Wireline, 37%

Wireless
Global

Serv ice, 41%

Enterprise,
21%

Mass
Markets, 21%

Wireless, 61%

Wireless
Equipment
and Other, 5%

Source: 3Q10 Company reports.

126

Source: 3Q10 Company reports.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless Strong and iPhone in the Wings


Verizon Wireless is the largest wireless company in the US with over 93 million
customers and approximately 46% EBITDA margins the highest in the US wireless
space. The company is 55% owned by Verizon and 45% by Vodafone.
Figure 60: Wireless Subscriber Breakdown
Wholesale,
Prepaid, 5%

7%

Postpaid ,
88%
Source: 3Q10 Company reports.

Figure 61: Verizon Wireless Coverage Map

Source: Company website (used with permission).

127

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 57: Verizon Wireless DCF Analysis


$ in millions

2011E
201,376
(48,031)
(61,557)
91,788
125,036
216,824
926
217,750
(97,987)
119,762

NPV of wireless EBITDA 12E-20E


NPV of wireless taxes paid 12E-20E
NPV of wireless capex 12E-20E
NPV of wireless free cash flow 12E-20E
NPV of wireless 2020 terminal value @ 4x EBITDA
PMV of wireless operations
Plus: Verizon Wireless net cash (debt), end of period
PMV Equity
Less 45% Vodafone Ownership
PMV of Wireless to Verizon Equity
Source: J.P. Morgan estimates.

We expect the majority of


Verizons iPhone sales to be
upgrades, not new customers

iPhone to be focus of attention in first half of 2011


On January 11, Verizon announced that the iPhone would be available on its network
in early February. To date, Verizons smartphone penetration significantly lags that
of AT&T, but recent success with Android-based devices has helped to accelerate
penetration. Given Verizons industry-leading brand, we expect the company to see
strong sales of the iPhone once it becomes available. We believe market share
growth will be fueled by existing iPhone customer defections from AT&T, while
Sprint and T-Mobile could lose some customers to Verizon as well. However, the
bulk of iPhone adopters are likely to be existing Verizon customers upgrading their
handsets. As a result, the primary benefit may be lower churn, not increased gross
additions.
At the same time, we expect the strong iPhone adoption, combined with a continued
move toward more expensive smartphones as a share of handset sales, will pressure
Verizon Wireless margins in the first half of 2011. We model margins falling to
41.9% in 2Q11, which we expect to be the quarter with the biggest impact.

Verizon postpaid ARPU grew an


estimated 1.6% in 2010; could
accelerate in 2011

ARPU to grow with smartphone penetration; set for accelerated revenue growth
Verizons average postpaid revenue per user lags AT&Ts somewhat due to lower
smartphone and 3G penetration. 3G smartphones and integrated devices made up
38.5% of Verizons postpaid base in 3Q10 vs. 51% at AT&T, and smartphones at
Verizon represented only 22.7% vs. AT&T with just iPhone representing nearly 27%
of postpaid customers. As Verizon handsets sales increasingly shift to smartphones,
this percentage should eventually catch up to AT&Ts, and the company should
somewhat close the gap with AT&T on data ARPU.
Table 58: ARPU Breakdown
Wireless ARPU
Postpaid Wireless ARPU
Postpaid Data ARPU ex-Laptop Card
Postpaid Voice ARPU ex-Laptop Card

1Q09
$50.49
$52.11
$13.54
$38.70

2Q09
$50.85
$52.41
$14.14
$38.42

3Q09
$50.77
$52.60
$14.89
$37.87

4Q09
$50.02
$52.22
$15.47
$36.89

1Q10
$49.92
$52.41
$16.31
$36.26

2Q10
$50.35
$53.17
$17.38
$36.01

3Q10
$50.94
$53.66
$18.44
$35.49

Source: Company reports and J.P. Morgan estimates.


Note: Ex-Laptop card ARPU is estimated.

Interestingly, Verizon has had good progress with smartphone sales even without the
iPhone the carrier has been good at selling Blackberry but the focus has turned to
aggressively pushing smartphones based on Googles Android operating system, and
take rates have increased significantly since the launch of the Motorola Droid in

128

4Q10E
$50.39
$53.32
$18.83
$34.75

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

4Q09. In 3Q10, 43% of new sales were smartphones, up from 23% in 3Q10. As a
result, total smartphone penetration of the base was 22.7%, up from 13.4% in 3Q09.
As Verizon grows its smartphone base of users, the company should see accelerating
postpaid ARPU growth and we model growth of 2.3% in 2011 vs. 1.6% in 2010.
Given our estimate for growth of more than 3% at AT&T, there could be upside to
our estimate for Verizon as well.
Figure 62: Device Sales Mix
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%

1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10E

1Q11E

2Q11E

3Q11E

4Q11E

Feature phone

66.0%

56.0%

41.0%

30.0%

33.0%

37.0%

37.0%

23.8%

33.7%

21.2%

22.6%

17.9%

Internet

5.0%

6.0%

6.0%

6.0%

7.0%

6.0%

7.0%

8.5%

8.6%

8.2%

8.4%

9.5%

3G Multimedia

3.0%

17.0%

30.0%

34.0%

24.0%

17.0%

13.0%

15.3%

0.3%

3.4%

0.0%

0.0%

Smartphone (non-iPhone)

26.0%

21.0%

23.0%

30.0%

36.0%

40.0%

43.0%

52.2%

36.1%

35.9%

38.1%

45.4%

Smartphone (iPhone)

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

20.0%

30.0%

30.0%

30.0%

Source: Company reports and J.P. Morgan estimates.

Figure 63: Device ARPU Mix


$100.00

$90

$90
$70

$80.00

$54

$50

$60.00

$34

$40.00
$20.00
$0.00
Smartphone

Smartphone

(iPhone)

(non-iPhone)

3G Multimedia

Internet*

Feature

Total postpaid
base

Source: Company reports and J.P. Morgan estimates.

We expect Verizon to follow


AT&T with smartphone usage
caps

Tiered data pricing on smartphones likely to provide additional fuel to growth


There has been significant discussion by Verizon and other industry players
regarding tiered data pricing. AT&T took the first steps last summer, establishing
$15 and $25 plans for usage of 200 MB and 2 GB, respectively, and eliminating the
unlimited usage plan for new smartphone customers.
More recently, Verizon followed with slightly different plans for smartphone users.
The company maintained an unlimited plan for $30 but also launched a 150 MB plan
for $10. Both carriers believe that the pricing changes will be accretive to ARPU,
with the benefit of users migrating from lower-ARPU devices to the $10 or $15 plans
offsetting migrations down from unlimited plans. In addition, as devices and
129

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

applications continue to advance and network speeds improve, these users are likely
to exceed the introductory data caps and require plans with more bandwidth cushion.
A similar thought process applies to Verizons recently introduced pricing for LTE
data cards. The company is offering 5 GB of monthly data usage for $50, a $10
discount to 3G pricing, and 10 GB for $80. However, with the very fast speeds of
4G, some users have used up their entire 5 GB allotment in less than an hour. We
expect that the company established this $50 plan to attract current 3G users to 4G
(and off the 3G network), but to then up-sell them to the $80 plan over time.
Margins to compress as handset sales mix shifts to iPhone, smartphones
Given the iPhones popularity, we expect high sales volume at Verizon, though not
the ~30-40% share that AT&T has seen over the last year. Regardless of whether
sales are in the form of upgrades or purchases by new customers, the bulk of iPhone
sales will carry a Verizon subsidy a subsidy we estimate will be $100 more than
Verizons already-high ~$250 average smartphone subsidy.
Note that we expect iPhone prices to Verizon to be somewhat higher than those to
AT&T given Verizons expected smaller scale in iPhone handset sales. We estimate
that, given its scale, selling 30%-plus of iPhones in any given quarter, AT&T likely
receives better pricing than any other carrier globally. In addition, we dont expect
Apple to charge AT&T any less than before, despite the ending of the exclusivity
agreement.
Figure 64: Subsidy by Device
$400
$350
$300
$250
$200

$350

$150

$250

$100
$50

$125

$125

3G Multimedia subsidy

Internet subsidy

$0
iPhone subsidy

Smartphone subsidy
(non-iPhone)

Source: Company reports and J.P. Morgan estimates.

In periods of strong sales, AT&T saw significant hits to its margins because of
upfront costs. We expect Verizon to also see some margin compression, though less
than did AT&T given its already-high smartphone sales mix, smaller iPhone share,
and ongoing efforts to reduce costs to offset ramping subsidy expense. The worst of
this compression is expected to occur in 2Q11 and we expect margins to move back
above 45% in 2012.

130

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 65: Verizon Wireless Net Add and Margin Forecasts


$ in millions, subscribers in thousands
800

48.0%

600

46.0%
44.0%

400

42.0%

200

40.0%

38.0%
1Q10

2Q10

3Q10

4Q10E 1Q11E 2Q11E 3Q11E 4Q11E 1Q12E 2Q12E 3Q12E 4Q12E


Postpaid Net Adds

Wireless Serv ice Margins

Source: Company reports and J.P. Morgan estimates.

We expect Verizons 3G network


demand to grow 55% 2010 and
57% in 2011

LTE shift offsets some iPhone network demand; 3G should still grow 57% in
2011
Verizon was the first national carrier to launch LTE with coverage of 110 million
pops in early December 2010 and plans to cover 200 million by mid-2012 and the
entire country by the end of 2013. Speeds on the network early on are bursting above
20 Mbps but as the network loads we expect average speeds of 5-10 Mbps,
substantially higher than the speeds of roughly 1 Mbps on the companys 3G EVDO
Rev A network.
In terms of devices, the company currently only offers two laptop dongles, but will
begin offering four smartphones, two tablets, and two mobile hotspots in 1H11. As
the mix of sales shifts toward 4G from 3G, we expect data demand growth on the
companys 3G network to slow in 2011 and for the network to start to shrink in a few
years.

Table 59: Data Usage by Device and in Total


MB for individual devices, GB for total monthly usage
3G iPhones
3G Smartphones (ex-iPhone)
3G Multimedia
Internet*
Feature
Average postpaid sub
LTE Monthly network data usage (GB)
3G Monthly network data usage (GB)

1Q09
100
100
50
1,500
20
118

2Q09
150
100
50
1,600
20
125

3Q09
200
100
50
1,700
20
130

4Q09
250
150
50
1,800
20
150

1Q10
300
175
50
1,900
20
170

2Q10
325
200
50
2,000
20
189

3Q10
350
200
50
2,000
20
199

9,112

9,799

10,310

12,049

13,748

15,434

16,334

4Q10E
350
225
50
2,100
20
223
127
18,358

1Q11E
350
225
50
2,200
20
244
363
20,077

2Q11E
400
250
50
2,300
20
282
734
23,021

3Q11E
400
275
50
2,400
20
315
1,517
25,171

4Q11E
425
300
50
2,500
20
351
2,543
27,365

Source: J.P. Morgan estimates.

131

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 66: LTE % of Sub Base and Handset Sales


$ in millions, subscribers in thousands

6.0%

19%

5.0%

16%

15.0%

4.0%
3.0%

10.0%

2.0%

6%

1.0%
0.0%

20.0%

1%
4Q10E

5.0%

2%
1Q11E
LTE % of Sub Base

0.0%
2Q11E

3Q11E

4Q11E

LTE % of Handset Sales

Source: Company reports and J.P. Morgan estimates.

Vodafone ownership stake in Verizon Wireless we assume status quo


Vodafone (VOD, rated Overweight by J.P. Morgan European Telecoms analyst Paul
Howard) owns 45% of Verizon Wireless, which it has held since Verizon Wireless
was created in 2000. Despite its ownership, Vodafone exercises little control over the
joint venture and has never received a substantial cash disbursement from the
company. Recently, however, managements of both Verizon and Vodafone have
made public comments that the company could pay a dividend once its debt is fully
paid down in 2012.
Aside from the dividend, we assume that the Verizon/Vodafone relationship remains
static, with no change in Vodafones ownership. While we believe that Verizon
would gladly buy the rest of the company at the right price, Vodafone appears to
have no pressing need for the money, there could be potentially onerous tax
consequences, and we dont believe that Verizon would be so keen as to pay a high
multiple.
Wireless cash flow expected to be re-directed in 2011 and 2012. In the last few
years the majority of Verizon Wireless cash flow has gone to pay down its intercompany debt owed to Verizon which was generated to pay for the $28 billion
acquisition of Alltel in early 2009. That debt has now declined from $20.8 billion in
2Q09 to only $12.5 billion in 3Q10, and we expect it to be repaid in 2011. After the
inter-company debt is fully paid down, Verizon Wireless is likely to instead stream
cash to both parents in the form of a dividend assuming a 90% payout ratio this
could total $16.8 billion in 2012 (at 5% entity tax rate), or $9.2 billion to Verizon and
$7.6 billion to Vodafone.

132

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireline
Consumer wireline revenue now dominated by broadband, video
Verizons consumer business has been plagued by voice line disconnects, driven by
wireless substitution and cable competition. Customers are finding both alternatives
to be more economical. An estimated 25% of all voice customers nationwide use
wireless service exclusively. Cable competitors often benefit from having higher
broadband speeds to bundle with their voice and video offerings, particularly in nonFiOS markets. As a result, Verizon has experienced a 10-12% rate of line loss for
several quarters, contributing to compound annual revenue decline of 10% from 2005
to 2009.
FiOS is moving from buildout to
focus on penetration

FiOS now available for sale in 12.5 million of 17.0 million planned homes, with
build nearly complete. Outside of voice, however, Verizons broadband adds have
remained fairly steady and in recent quarters the companys FiOS business, which
offers both higher-speed data and video, has gained scale. Now available for sale in
12.5 million homes and with 3.9 million Internet and 3.3 million TV customers,
FiOS is a $6.7 billion business and growing at approximately 29% year-over-year.
FiOS now represents 50% of overall consumer revenue.

Table 60: Verizon FiOS Modeling Assumptions


Revenue in $ millions, subscribers in thousands

2009
3,286
805
32%

2010E
4,065
779
24%

2011E
4,727
662
16%

2012E
5,257
530
11%

2013E
5,707
450
9%

2014E
6,090
383
7%

2015E
6,415
325
5%

2016E
6,692
277
4%

FiOS Homes Passed (data)


Net New Homes Passed
FiOS Homes Open for Sale (data)
FiOS Data Penetration

15,400
2,700
12,500
26%

16,750
1,350
13,000
31%

17,750
1,000
14,000
34%

18,750
1,000
18,600
28%

19,000
250
18,620
31%

19,600
600
19,208
32%

19,600
19,208
33%

19,600
19,208
35%

Video
FiOS TV Subscribers
Net FiOS TV Subscribers Adds
% change y/y

2,750
832
43%

3,465
715
26%

4,107
642
19%

4,621
514
13%

5,057
437
9%

5,429
371
7%

5,744
316
6%

6,012
268
5%

FiOS Homes Passed (TV)


Net New Homes Passed
FiOS Homes Open for Sale (TV)
FiOS TV Penetration

14,700
2,000
11,500
24%

16,050
1,350
12,600
27%

17,050
1,000
13,600
30%

19,600
2,550
18,000
26%

19,600
18,000
28%

19,600
18,000
30%

19,600
18,000
32%

19,600
18,000
33%

Triple Play Data/Video ARPU


Triple Play Data/Video Revenue
ARPU - FiOS Internet Only Subs

$122.55
3,432
$94.55

$122.55
4,570
$94.55

$120.40
5,470
$92.40

$116.77
6,115
$88.77

$112.27
6,520
$84.27

$107.92
6,790
$79.92

$103.69
6,951
$75.69

$99.59
7,025
$71.59

FiOS Internet Subscribers


Net FiOS Internet Subscribers Adds
% change y/y

Source: Company reports and J.P. Morgan estimates.

Broadband, video now 50% of Verizons consumer business. Net-net, we


estimate that Verizons consumer revenue has shifted from 86% voice in 2006 to an
estimated 37% in 3Q10. Further, we expect the broadband and video businesses to
continue to grow, and for voice losses to tail off as Verizons weaker markets were
let go in the Frontier spinoff and its core market dynamics are improved by the
presence of FiOS.

133

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

Figure 67: Verizon FiOS

Source: Company reports (used with permission).

134

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Enterprise showing signs of life


Verizons enterprise business has suffered along with the rest of the economy. As
enterprises curtailed telecom spending and pared payrolls, large projects were halted
and voice volumes fell significantly. However, recent trends at Verizon suggest the
business is stabilizing, with the return of some large deals and improved hiring
trends. In addition, CPE spending has picked up as enterprises begin making
necessary equipment purchases.
In the second quarter of 2010, enterprise revenue grew for the first time after steady
declines throughout 2009. We believe trends will continue to depend upon global
macro conditions, and economic improvement may be ahead but is unlikely to
happen quickly, in our view. Nonetheless, stabilizing enterprise trends should help
support revenue stability in the overall wireline business, while stemming expected
pressure on margins.
Figure 68: Verizon Enterprise Revenue and Growth
$ in millions, subscribers in thousands

4,050

4%
2%
0%
-2%
-4%
-6%
-8%
-10%
-12%

4,000
3,950
3,900
3,850
3,800
3,750
1Q09 2Q09 3Q09 4Q09 1Q10 2Q10 3Q10 4Q10E 1Q11E 2Q11E 3Q11E 4Q11E
Enterprise rev enue

y /y change

Source: Company reports and J.P. Morgan estimates.

Margins stabilizing
Verizons wireline margins have declined significantly over the past several quarters,
from 27% in early 2008 to close to a low of 19.4% in 1Q10. However, wireline
margin finally expanded in 2Q10 and then ticked up again in 3Q10, indicating that
the companys massive cost-cutting initiatives in the last couple years are finally
getting expenses under control. Early in 2010, management targeted $1.5-2.0 billion
of potential savings from moves to reduce the cost structure. Verizon has reduced
headcount by approximately 20k since 3Q09, and the company will likely recognize
the full benefit of these reductions in the first quarter of 2010. Though the workforce
is clearly the largest identifiable reduction, other actions including the rationalization
of real estate should also help and we believe management is on track to achieve
savings goals.
In addition, higher-margin businesses such as FiOS and strategic enterprise services
are becoming larger contributors to the top line, while management is also
eliminating certain wholesale streams which are low-margin in nature. These growth
areas are beginning to offset the impact of secularly driven double-digit declines in
access lines. Specifically, for enterprise, which is now almost as large as the mass
markets business, on the whole business trends appear to be stabilizing, though
additional economic strength is likely necessary for revenue growth. We estimate
margins will continue to improve to just over 22% by year-end 2011, however there
135

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

could be upside to our expectations if management can continue to find ways to


reduce the cost structure and if enterprise revenue trends improve more
meaningfully.
Wireline cash flow expected to expand to $2.4 billion in 2011
With declining EBITDA margins and heavy FiOS capital spending, the wireline side
of Verizon has seen operating margins dwindle since late 2007 to negative territory
in 1Q10, and cash flow from the wireline business was minimal in 2010
($351 million). In 2011, however, we expect wireline margins to stabilize and even
grow somewhat, and wireline capital spending to decline from $6.9 billion in 2010
(and $11.0 billion in the peak year of 2007) to $6.1 billion. As a result, we expect the
Verizon wireline business to generate cash flow of $1.2 billion in 2011 and
$1.0 billion in 2012, up from $0.4 billion in 2010.

Operating and Financial Outlook


We expect Verizon to maintain strong postpaid subscriber growth, though a
deceleration in wholesale growth could drive total net additions to fall year-overyear. Success with the iPhone will be critical to achieving our forecasts, which call
for growth of 4.6% in 2011 and 3.3% in 2012. As a result of the strong postpaid
additions, we expect overall ARPU to grow, despite dilution from lower-ARPU
wholesale subscribers. Modest ARPU growth and strong subscriber growth are
expected to drive a 200 bps acceleration in service revenue growth to 7% in 2011,
though this growth should slow as the iPhone benefit wears off in 2012 and beyond.
We believe certain wireline trends will improve from 2010, driving a deceleration in
wireline revenue declines. We expect access line losses to remain high, but
broadband should maintain decent traction. We expect slowing FiOS adds and
slowing DSL disconnects to drive 270k broadband adds, roughly flat with the
expected 2010 level. Similar to our FiOS broadband forecast, we expect a modest
decline in TV adds as well. On the business side, we expect a meaningful bounce
back to positive growth, as recent improvements continue. Collectively, these
individual business trends drive an estimated 1% decline in wireline revenue, a
deceleration from the expected 3% fall in 2010.

Earnings and Cash Flow Forecast


As iPhone sales ramp, we believe Verizons wireless service margins should see
some impact given the high-subsidy nature of the device. Though still industryleading, margins are forecast to fall to 43.9% in 2011, from 46.6% in 2010. However,
given managements focus on maintaining mid- to high-40s margins, we expect a
recovery to 45.8% in 2012. While wireless margins should contract, we believe
wireline margins will continue to improve as Verizon recognizes a full year of force
reductions and high-margin enterprise revenue trends improve. We forecast wireline
margin of 22.1% in 2011, reflecting the first year of margin growth since 2007. We
expect further margin expansion to 22.6% in 2012.
Our segment forecasts drive consolidated EPS estimates of $2.23 in 2011 and $2.45
in 2012. Excluding $0.13 of earnings from divested operations in 1H10, our forecasts
imply 4.2% growth in 2011 (including some dilution from iPhone) and 10.0%
growth in 2012 (as iPhone dilution falls away). Free cash flow comparisons between
2011 and 2010 suffer due to the divestiture of wireline assets to Frontier as well as
the divestiture of certain Alltel properties in 2010. However, our forecast for
136

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

$13.6 billion in free cash flow in 2011 reflects a decline in total capital expenditures
offset by lower wireless profitability due to iPhone pressures.

Table 61: Financial Forecasts


$ in millions except per share and per sub values, subs in thousands
Wireless
Subscribers
% change, y/y
Net additional subscribers
% change, y/y
Churn rate, monthly
ARPU, monthly
Service revenue
% change, y/y
EBITDA (operating cash flow)
% change, y/y
EBITDA margin
Capital Spending
CAPEX/Service revenue
Wireline
Residential Voice Subscribers
% change, y/y
Business Voice Subscribers
% change, y/y
Total Broadband Subscribers
% change, y/y
FiOS TV Subscribers
% change, y/y
Total Wireline Revenue
% change, y/y
EBITDA (operating cash flow)
% change, y/y
EBITDA margin
Capital Spending
CAPEX/revenue
Consolidated Revenue
% change, y/y
EBITDA
% change, y-t-y
EBITDA margin
Depreciation & Amortization
EBIT
Taxes on EBIT @ 32.0%
Implied tax rate
Capital expenditures
EPS - Basic
EPS - Diluted
Dividend Per Share

2006

2007

2008

2009

59,052
15.0%
7,691
3%
1.2%
$49.74
32,796
18%
14,513
20%
44.3%
6,618
20.2%

65,707
11.3%
6,653
-13%
1.2%
$50.92
38,016
15%
16,966
17%
44.6%
6,503
17.1%

85,275
29.8%
7,005
5%
1.4%
$50.84
49,717
33%
22,848
35%
46.0%
6,510
13.1%

89,172
4.6%
5,657
-19%
1.4%
$50.21
52,046
3%
23,668
4%
45.5%
7,152
13.7%

94,590
6.1%
5,304
-6%
1.3%
$50.41
55,627
5%
25,875
9%
46.5%
8,405
15.1%

26,762
-10%
16,476
NM
6,786
32%
207
NM
49,504

23,910
-8%
15,863
-4%
8,013
18%
943
356%
49,129
-1%
13,421
1%
27.3%
10,956
22%

20,956
-10%
14,966
-6%
8,673
8%
1,918
103%
48,214
-2%
12,893
-4%
26.7%
9,797
20%

15,618
-24%
12,540
-16%
8,160
-6%
2,750
43%
42,451
-12%
9,239
-28%
21.8%
8,892
21%

30%
14,264
10,458
3,347
32.0%
17,101

92,375
11%
30,665
24%
33%
14,120
16,545
5,294
32.0%
17,538

97,096
5%
32,572
6%
34%
14,505
18,067
5,781
32.0%
17,238

107,808
11%
35,695
10%
33%
16,215
19,480
6,234
32.0%
17,047

$1.32
$1.31
$1.61

$2.39
$2.39
$1.64

$2.54
$2.54
$1.75

$2.40
$2.40
$1.86

13,325
26.9%
10,259
21%
83,410
24,722

2010E

2012E

2013E

2014E

2015E

2016E

2017E

2018E

2019E

2020E

98,956
4.6%
4,366
-18%
1.3%
$50.66
58,945
7%
25,854
0%
43.9%
8,842
15.0%

102,265
3.3%
3,309
-24%
1.3%
$50.92
61,522
3%
28,152
9%
45.8%
9,228
15.0%

105,116
2.8%
2,850
-14%
1.3%
$51.04
63,545
3%
29,986
7%
47.2%
9,214
14.5%

107,446
2.2%
2,330
-18%
1.3%
$51.17
65,287
3%
31,137
4%
47.7%
9,467
14.5%

109,348
1.8%
1,902
-18%
1.3%
$51.17
66,581
2%
31,762
2%
47.7%
9,654
14.5%

110,897
1.4%
1,549
-19%
1.3%
$51.17
67,636
1%
32,477
2%
48.0%
9,807
14.5%

112,247
1.2%
1,350
-13%
1.3%
$51.17
68,524
1%
33,113
2%
48.3%
9,936
14.5%

113,421
1.0%
1,174
-13%
1.3%
$50.86
68,881
1%
33,317
1%
48.4%
9,988
14.5%

114,439
0.9%
1,017
-13%
1.3%
$50.56
69,131
0%
33,447
0%
48.4%
10,024
14.5%

115,316
0.8%
878
-14%
1.3%
$50.25
69,286
0%
33,513
0%
48.4%
10,047
14.5%

14,112
-8%
11,803
-6%
8,440
3%
3,465
26%
41,184
-3%
8,539
-8%
20.7%
6,942
17%

12,862
-10%
11,213
-5%
8,710
3%
4,107
19%
40,775
-1%
9,021
6%
22.1%
6,116
15%

11,962
-8%
10,672
-5%
8,953
3%
4,621
13%
40,393
-1%
9,139
1%
22.6%
6,059
15%

11,142
-8%
10,247
-4%
9,171
2%
5,057
9%
40,259
0%
9,350
2%
23.2%
6,039
15%

10,522
-6%
9,950
-3%
9,368
2%
5,429
7%
40,152
0%
9,566
2%
23.8%
6,023
15%

10,022
-6%
9,550
-4%
9,545
2%
5,744
6%
39,852
-1%
9,614
1%
24.1%
5,978
15%

9,642
-4%
9,312
-2%
9,704
2%
6,012
5%
39,580
-1%
9,667
1%
24.4%
5,541
14%

9,502
-2%
9,212
-1%
9,848
1%
6,240
4%
39,432
0%
9,749
1%
24.7%
5,323
14%

9,502
0%
9,212
0%
9,977
1%
6,434
3%
39,543
0%
9,896
1%
25.0%
5,141
13%

9,502
0%
9,212
0%
10,093
1%
6,599
3%
39,747
1%
10,066
2%
25.3%
5,167
13%

9,502
0%
9,212
0%
10,197
1%
6,739
2%
39,953
1%
10,238
2%
25.6%
5,194
13%

106,437
-1%
35,420
-1%
33%
16,256
19,164
6,132
32.0%
16,121

108,190
2%
34,714
-2%
32%
16,397
18,317
5,861
32.0%
15,894

110,078
2%
37,140
7%
34%
16,445
20,695
6,622
32.0%
16,223

111,829
2%
39,018
5%
35%
16,713
22,305
7,138
32.0%
16,189

113,890
2%
40,385
4%
35%
16,899
23,485
7,515
32.0%
16,425

114,978
1%
41,058
2%
36%
17,027
24,031
7,690
32.0%
16,568

115,841
1%
41,826
2%
36%
17,125
24,701
7,904
32.0%
16,284

116,651
1%
42,544
2%
36%
17,200
25,344
8,110
32.0%
16,195

117,181
0%
42,894
1%
37%
17,207
25,688
8,220
32.0%
16,064

117,687
0%
43,195
1%
37%
17,199
25,996
8,319
32.0%
16,127

118,094
0%
43,432
1%
37%
17,179
26,253
8,401
32.0%
16,176

$2.27
$2.27
$1.91

2011E

$2.23
$2.23
$1.96

$2.45
$2.45
$1.98

$2.63
$2.63
$2.02

$2.82
$2.82
$2.06

$2.90
$2.90
$2.10

$3.01
$3.01
$2.14

$3.11
$3.11
$2.19

$3.18
$3.18
$2.23

$3.25
$3.25
$2.27

$3.32
$3.32
$2.32

CAGR
10E-20E
2%
-16%
0%
2%
3%

2%

-4%
-2%
2%
7%
0%
2%
-3%
1%
2%

0%
4%
2%

Source: Company reports and J.P. Morgan estimates.

137

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 62: Verizon DCF Analysis


$ in millions except per share and per sub values, subs in thousands
Private Market Valuation Summary
NPV of wireless EBITDA 12E -- 20E
NPV of wireless taxes paid 12E -- 20E
NPV of wireless capex 12E -- 20E
NPV of wireless free cash flow 12E -- 20E
NPV of wireless 2020 terminal value @ 4x EBITDA
PMV of wireless operations
Plus: Verizon Wireless net cash (debt), end of period
PMV -- Equity
Less 45% Vodafone Ownership
PMV of Wireless to Verizon Equity

2010E
149,459
(35,338)
(47,080)
67,040
164,395
231,435
(13,037)
218,398
(98,279)
120,119

2011E
201,376
(48,031)
(61,557)
91,788
125,036
216,824
926
217,750
(97,987)
119,762

NPV of Verizon Corp. EBITDA 12E -- 20E


NPV of Verizon Corp. taxes paid 12E -- 20E
NPV of Verizon Corp. Capex 12E -- 20E
NPV of Verizon Corp. free cash flow 12E -- 20E
NPV of Verizon Corp. 2020 terminal value @ 6.2x EBITDA
PMV of Verizon Corporation
Less 45% Vodafone Ownership of Verizon Wireless
PMV of consolidated operations

195,413
(35,512)
(81,965)
77,936
173,694
251,630
(98,279)
153,351

260,993
(48,906)
(103,728)
108,360
141,194
249,553
(97,987)
151,566

Plus (less): investments


Less: net debt, end of period (2)
PMV -- Equity

4,979
(45,540)
112,791

4,979
(37,502)
119,043

2,832

2,837
0
2,837

$35.93
$37.76

2008
14.2x
14.9x

2009
15.0x
15.8x

2010E
15.8x
16.6x

2011E
16.1x
16.9x

2012E
14.7x
15.4x

134,471
139,662

4.1x
4.3x

5.4x
5.7x

5.7x
5.9x

5.8x
6.1x

5.5x
5.7x

4.9%
4.6%

5.2%
4.9%

5.3%
5.1%

5.4%
5.2%

5.5%
5.2%

Valuation Multiples
Current Price P/E
Target Price P/E
Current EV/EBITDA
Target Price EV/EBITDA
Current Dividend Yield
Target Price Dividend Yield
Wireless Valuation Method:
1

Current shares outstanding


Shares to be issued/warrants/options (1)
Fully diluted shares outstanding

2,832

Corporate Valuation Method:


1

1 = DCF from above @ 7.5% discount rate (adjust below)


2 = 6.0x forward EBITDA multiple
6.0x

Verizon Valuation Grid - Sensitivity to corproate DCF per Share


Terminal Growth Rate (across) on 2020E corporate FCF, Discount Rate (down)
$37.76
6.5%
7.0%
7.5%
8.0%
8.5%

$39.83
10%
$35.84

$41.95
10%
$37.76
$1.96
10.5%

2010E
35,420
23,777

2011E
34,714
23,080

2012E
37,140
24,471

EPS, diluted
Current consolidated P/E

$2.27
15.8x

$2.23
16.1x

$2.45
14.7x

Consolidated Free Cash Flow


Verizon Corp Free Cash Flow (excludes VOD portion of VZW)
Verizon Corp. FCF / diluted share
Verizon Corp. FCF yield (current)
FCF multiple

16,407
9,057
$3.20
8.9%
11.2x

13,591
6,151
$2.17
6.0%
16.6x

12,747
4,323
$1.52
4.2%
23.6x

8.2x
8.9%

9.9x
6.0%

10.5x
4.2%

PMV per share


Private to public discount
Year-end 2011 fair value per share
Plus annual dividend
Total upside to current level

1 = DCF from above @ 7.5% discount rate (adjust below)


2 = 8.0x forward EBITDA multiple
8.0x

-0.5%
43
37
32
27
23

0.0%
47
40
35
30
25

0.5%
51
44
$37.76
32
28

1.0%
57
49
41
35
30

1.5%
63
54
46
39
33

Consolidated Valuation Based on Consolidated Forward P/E Multiple


Consolidated EBITDA (includes 100% of VZW)
Verizon Corp EBITDA (excludes VOD portion of VZW)

EV/FCF
FCF Yield

Valuation Grid - year-end value per share


P/E Multiple on Continuing Earnings Per Share

12x
14x
16.0x
18x
20x

138

10x
12x
14.0x
16x
18x

2011E
22
27
$31.20
36
40

Valuation Grid - year-end value per share


Operational Free Cash Flow Multiple

10x
12x
15.0x
17x
20x

Source: Company reports and J.P. Morgan estimates.

2010E
27
32
$36.30
41
45

2010E
Diluted
32
38
$47.97
54
64

10x
12x
15.0x
17x
20x

2011E
Diluted
32
38
$47.97
54
64

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Sprint Nextel
Initiate with Overweight Rating and $7 YE11 Price Target
We initiate coverage on Sprint with an Overweight rating and year-end 2011 price
target of $7, implying 50% upside from current levels. In our view, the companys
continued operational turnaround offers investors an excellent risk/reward and we see
Sprint as the stock in our coverage with the largest potential upside over the next 13 years. We expect Sprint to add postpaid subscribers for the first time in three years
in 4Q10 and to grow from there, and for margins to stabilize near recent levels for
2011.

Overweight
Sprint Nextel (S;S US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (bn)
Price Target ($)
Price Target End Date

4.41
12 Jan 11
5.31 - 3.10
13.19
Dec
3
7.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

(0.21)
(0.13)
(0.17)
(0.34)
(0.85)
(0.52)

(0.29)A
(0.25)A
(0.30)A
(0.29)A
(1.14)A
(1.14)A

(0.11)
(0.10)
(0.09)
(0.15)
(0.45)
(0.67)

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on GAAP results.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Continued turnaround to drive improvements in subscriber trends
Sprint in the last three years has managed to improve its customer satisfaction scores,
handset offering, and network quality dramatically and we expect this to continue in
2011. The result has been improvements in gross addition share and churn rates, and
the stabilization of ARPU at the $55 level from what had once been mid-single-digit
percentage annual declines. Using a recent Consumer Reports survey as an indicator,
we expect continued improvements in 2011 and beyond.
Top-line improvement leads to margin stability or better
We expect Sprint wireless service revenue to grow by 1.4% in 2011 as subscribers
grow, which should reduce the pressure on margins. We dont model margin
expansion as the cost of handsets and early network transition efforts will keep costs
up, but believe there could be upside to our forecast if postpaid churn declines more
than the 18 basis points we model. While our EBITDA estimate is somewhat below
Street consensus, we see more upside than downside risk to our model from here.
Free cash flow still substantial as EBITDA stabilizes but capex ramps
We expect Sprint to increase capital spending by 37% to $2.8 billion in 2011 and
further to $3.2 billion in 2012. This is offset somewhat by ~$500 million of spectrum
swap spend in 2010 that will not recur, and we model $1.3 billion in cash generation
in 2011 (a 9.1% yield at current levels) and $1.2 billion in 2012 (8.7% yield).

139

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Network modernization offers long-term benefits but near-term financial drag


Sprint will start in 2011 to roll out a three- to five-year network upgrade. Early on we
expect margin headwinds including tower amendment costs but by late 2012 the
company should see improved coverage and the start of roaming savings. Sprints
guidance implies EBITDA margin in the mid-20s to 30% long term, which would
offer substantial upside to our model (20-21% long-term margins).
Expect additional Clearwire investment, eventually
Clearwires recent debt raise takes some of the pressure off, but we eventually expect
Sprint to contribute more cash to fund further 4G buildouts. The relationship,
unfortunately, has been bumpy, and both companies need to be building faster to
keep up with competitive offers and network demands.

Investment Risks
Clearwire issues remain unresolved
Clearwire needs substantial capital, and with Sprint owning 54% of the company, we
think it is unlikely to attract enough from anyone other than Sprint. Until the
relationship is clarified, the risk of substantial further investment and duplicative
capital spending efforts likely will be a drag on both companies stocks. We believe
the worst case could be that Sprint is forced to consolidate Clearwire for accounting
purposes (potentially creating issues with its debt covenants), but not really be able
exert control over the company and its spending.
Verizon takes significant share with iPhone
When Verizon launches the iPhone in early February, it could impact Sprints ability
to generate subscriber growth, as Apple fans that werent willing to go to AT&T
could leave Sprint for Verizon. In addition, we assume that Sprint is able to sell the
iPhone as well eventually, which could prove inaccurate.
Network modernization effort could be delayed or take more capital than
planned
Sprints network modernization effort includes a number of new technologies that
could perform worse than expected or fail entirely. In addition, the network transition
could cost more in terms of Opex and Capex than we and the company expect or
impact network quality and subscriber experience.

Company Description
Sprint is the third-largest wireless company in the US and also owns a global longdistance and internet backbone. Over 87% of revenue comes from wireless. The
company is the result of the merger of Sprint and Nextel in 2005 as well as the
acquisition of a number of affiliates, including Alamosa and iPCS, and Virgin
Mobile which was purchased in 2009. With a national footprint running on CDMA
technology, the company has close to 49 million wireless subscribers and has
stabilized that base from a high-single-digit percentage loss rate in 2008 to a lowsingle-digit rate year to date in 2010. Sprint also holds a non-controlling 54% stake
in Clearwire, which is currently building a 4G network using WiMAX technology.

140

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $7/share for Sprint. We base our target
price primarily on a discounted cash flow analysis that assumes an 8.5% WACC for
the wireless business and 7.5% WACC for wireline. We also use perpetual growth
rates of 1.0% and 0.0% for wireless and wireline, respectively. Finally, we give
Sprint credit for $3.4 billion in NOLs and its 54% stake (pre-conversion of recently
issued out-of-the-money exchangeable notes) in Clearwire based on the current
$5.61 price of CLWR shares. Shares also appear attractive on an EV/EBITDA basis,
trading at 4.6x our 2011E EBITDA, compared to wireless peers at 5.1x and AT&T
and Verizon at an average of 5.9x.

Poised for Continued Postpaid Improvement


Though better, Sprints churn is
still 2x Verizons

Subscriber trends have improved


Sprint has demonstrated steady improvement in its share of industry postpaid gross
additions, which has risen from 12.2% in 2Q09 to 18.2% in 3Q10, and we expect
Sprint to garner share roughly 19% in 4Q10. Unfortunately, however, the company
continues to lose subscribers as its churn of 1.93% in 3Q10, although down from the
mid-2% highs, is still nearly twice that at AT&T and Verizon. Sprint is faced with
the task of continuing to grow gross additions, while reducing churn, and managing
all of this while keeping an eye on EBITDA margins.
Figure 69: Wireless Subscriber Breakdown

Wholesale,
8.5%

Postpaid
iDEN, 12.3%

Prepaid
CDMA,
14.6%

Prepaid
iDEN, 9.2%

Postpaid
CDMA,
55.3%
Source: 3Q10 Company reports.

Table 63: Postpaid Subscriber Metrics


In thousands
Postpaid Gross Adds
Postpaid Disconnects
Postpaid Net Adds

1Q09
1,184
(2,434)
(1,250)

2Q09
1,157
(2,148)
(991)

3Q09
1,415
(2,216)
(801)

4Q09
1,636
(2,140)
(504)

1Q10
1,594
(2,172)
(578)

2Q10
1,619
(1,847)
(228)

3Q10
1,810
(1,917)
(107)

4Q10E
1,860
(1,835)
25

1Q11E
1,711
(1,736)
(25)

2Q11E
1,838
(1,738)
100

3Q11E
1,792
(1,742)
50

4Q11E
1,846
(1,746)
100

Source: Company reports and J.P. Morgan estimates.

141

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 70: Share of Big 4 Postpaid Gross Adds and Disconnects


40.0%

35.3%

35.0%

30.7%
27.8% 27.1% 27.1%
26.2%

30.0%

24.7% 24.6% 25.4%

25.0%

23.4%

21.8% 21.8%

20.2% 19.9% 20.3% 20.5%

20.0%
15.0%

18.5%
15.8%

10.0%

11.8% 12.1% 12.2% 12.2%

13.8%

16.3%

18.1% 18.2% 18.2%

19.4% 18.7% 19.7% 19.2% 19.7%

1Q10

4Q10E 1Q11E 2Q11E 3Q11E 4Q11E

5.0%
0.0%
1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

Postpaid Gross Adds

4Q09

2Q10

3Q10

Postpaid Disconnects

Source: Company reports and J.P. Morgan estimates.

Sprint nearly tied with Verizon


on customer score

Customer satisfaction surging. We believe that Sprints rebound in gross additions


can be attributed to a number of factors including a better handset selection,
improved network performance, and better customer service experience. As a result,
Sprints overall customer score (as measured by Consumer Reports) surged from 67
(out of 100) to 73 in the last year only one point off Verizons and compared to 69
at T-Mobile and 60 at AT&T.
Can share gains continue?
Given the trailing nature of customer satisfaction surveys and Sprints steady
improvement thus far, we believe the company needs to continue to gain gross add
share and reduce churn from here. In addition, with widespread complaints regarding
the AT&T and T-Mobile networks, Sprint should have the chance to benefit as
contracts for customers currently at those two carriers expire, giving dissatisfied
consumers an opportunity to switch.
However, the company may face some headwinds in early 2011. Although Sprint
offers lower smartphone monthly price plans, some customers may be tempted to
make the switch to Verizon and the iPhone, which will be launched in early
February. Some customers who would like an iPhone may have held off making a
switch to AT&T, given perceptions of network issues at that carrier.
iPhone could come to Sprint, perhaps as early as mid-2011
We would not be surprised to see Sprint gain access to the CDMA iPhone, perhaps
as early as mid-2011 after Apple has ramped up Verizon inventory and made sure it
can satisfy demand there (unlike T-Mobile and AT&T, Sprint can use the exact same
phone as Verizon with full functionality). Historically, once a device has moved
beyond exclusivity at one carrier, most carriers have typically also offered the device
within a year. While there is the prospect that Sprint may have to pay somewhat
more than Verizon for the already expensive device, recent high-end handsets sold at
Sprint that have WiMAX as well as CDMA already wholesale in the $550-600 range,
and the average iPhone shouldnt cost much more.

142

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Apple is unlikely to make a WiMAX version, in our view, and Sprint could fall
behind again with Apple if it and Clearwire stick with WiMAX while the global
evolution moves toward LTE.
In the meantime, we believe Sprint needs to continue pushing its new, attractive
handsets. Launches on the Android platform, particularly the EVO with WiMAX,
over the past several months have helped take Sprint up to parity with AT&T and
Verizon (aside from the iPhone of course), and we believe to a somewhat better
position than T-Mobile.
Figure 71: Sprint Nextel Smartphones

HTC EVO 4G $199.99

Samsung Epic 4G $249.99

Motorola i1 $149.99

HTC Hero $149.99

Samsung Intercept $99.99

Samsung Transform $49.99

LG Optimus S $0

Sanyo Zio $0

Source: Company website (used with permission).

Prepaid Less of a Driver Going Forward


Boost iDEN seems to be falling
off

Reinvigorated competitor offerings provide a challenge


When Sprint initially rolled out its Boost Mobile unlimited pay-by-the-month offer in
January 2009, the company had the competitive advantage of offering a nationwide
network. The companys main competitors at the time, Leap Wireless and
MetroPCS, could not match the offer. However, over time that advantage has
dissipated. In addition, through its wholesale relationship with Verizon (and now
AT&T and T-Mobile as well), Tracfone rolled out a nationwide offer (Straight Talk)
at a discount to the Boost plan. Also, MetroPCS and Leap have signed extensive
roaming agreements that effectively give them the ability to market nationwide
coverage. Their offers, as well as Tracfones, are priced at a $5-$10 discount relative
to the $50 price for new Boost customers.

143

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 64: Prepaid Pricing Comparison


LD
X
X
X
X

SMS
X
X
X
X

Data

$35
$45
$55 Android
$55 Blackberry

Talk
X
X
X
X

X
X
X

Taxes
~10%
$35.00
$45.00
$55.00
$55.00

$40
$45
$50
$60
$60 Blackberry

X
X
X
X
X

X
X
X
X
X

X
X
X
X
X

X
X
X
X
X

$40.00
$45.00
$50.00
$60.00
$60.00

$50

$50.00

Talk
$0.10/min
1500 Talk
X
X

LD
X
X
X

Text
X
X
X
X

Data

$15
$30
$50
$70

30 MB data
100 MB data
2 GB data

$25
$40
$60

Talk
300
1200
X

LD
X
X
X

SMS
X
X
X

Web
X
X
X

$27.50
$44.00
$66.00

$30
$45

1000
X

X
X

1000
X

30 MB
X

$33.00
$49.50

$15
$30

200
500

X
X

$16.50
$33.00

$10
$20
$30

50
125
200

X
X
X

$11.00
$22.00
$33.00

$10
$25
$50
$30
$45

100
400
1000
1200
unlimited

X
X
X
X
X

$10.00
$25.00
$50.00
$30.00
$45.00

Leap

PCS

Boost

T-Mobile

Virgin Mobile

Taxes
$16.50
$33.00
$55.00
$77.00

Straight Talk

Net-10

TracFone Value Plans

Page Plus

1200
X

50 MB
20MB

Source: Company websites.

Price war for unlimited pay-by-the-month market on hold for now


After a roughly 18-month period of price competition, the prepaid pricing
environment appears to have stabilized for now. However, share gains across all
competitors could slow, with limited points of differentiation across the market.
Instead, competitors are likely to pursue remaining industry growth in the prepaid
market and use promotions from time to time to win incremental share. We believe
Sprint management prefers to see stable ARPU trends rather than outsized subscriber
gains for what is considered to be a non-core prepaid business.

144

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 72: Prepaid Net Adds


In thousands

1,000
800

777
674

666

600

435

400

471

450

3Q10

4Q10E

348
173

200
0
1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

Source: Company reports and J.P. Morgan estimates.

Sprints other prepaid brands


However, to some degree, Sprint has tried to respond to the competition with
differentiated offerings through its Virgin Mobile and Assurance brands. The Virgin
Mobile offers are priced lower and put more emphasis on unlimited messaging
services and data, while limiting minutes of use. We believe Sprint is using the
Virgin brand for discounted pricing to avoid pricing down its existing base of Boost
customers at the $40-50 price point.
Assurance ARPU is only
~$10/month

Assurance represents the majority of prepaid growth


Sprint is also targeting the very low end of the market with its Assurance offer,
which offers customers who qualify for USF Lifeline service a free handset and
250 minutes per month. Assurance is now available in 15 states after three were
added in 4Q10 and one thus far in 2011, and we believe has roughly 1-2 million
customers. Assurance competes primarily with Tracfones Safe Link, which has had
to increase the number of minutes it offers to keep up with growth in Assurances
offer.

4G Rollout Slowing Creates Competitive Questions


Slower Clearwire build could
create 3G congestion

Clearwire still an overhang, but less so


Sprints investment in Clearwire still represents its 4G strategy. Through its 54%
ownership and wholesale agreement, Sprint utilizes Clearwires spectrum and
network (Clearwire leverages Sprints network as well for roaming), which has
grown from covering 45 million to 120 million people in the past 12 months. Even at
that coverage breadth though, Sprint still needs Clearwire to continue building its
network and adding capacity and coverage to launched markets.
Unfortunately, funding has been and somewhat remains the primary hurdle for
Clearwire adding more markets. Clearwire recently raised $1.3 billion of convertible
debt and secured debt funding, which we estimate will be enough to get to positive
free cash flow if the company stops adding new markets and just increases
penetration in cities in which it has already launched. To grow aggressively again in
2011, Clearwire will need substantial additional funding, which could come in the
form of a spectrum sale, or an equity infusion from Sprint or a new strategic partner.

145

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Sprint may be hesitant to make additional investments


Sprint has the cash to invest in Clearwire, but we believe the two partners disagree on
critical matters, including wholesale pricing for smartphones and Clearwires retail
strategy. Without more control, Sprint seems to be reluctant to continue to fund
Clearwire significantly more than it has already.
Although Sprint does not control Clearwire, it does have >50% ownership and under
the most conservative accounting could be required to consolidate Clearwire
financials for accounting purposes. While its auditor KPMG did not require this in
2009, if this were to change, Sprint could break the 4.5x leverage covenant on its
$2.1 billion revolving credit facility triggering the need to renegotiate the terms or
possibly pay a fee.
Clearwire slowdown could impede Sprints marketing, clog 3G network
After the recent debt offering, Clearwire management (which detailed a goingconcern risk in its third quarter 10-Q) has more breathing room, but still needs more
cash to expand as aggressively as it and Sprint would like.
Without additional funding, Clearwire would have to survive on its latest round of
funding, which would limit expansion. Sprint though is aggressively selling
smartphones that include WiMAX for a large premium, which is wasted on
customers that buy the device outside of a Clearwire market. Even worse, in our
view, in those non-Clearwire markets, Sprint may be spending fairly little on 3G
capacity in anticipation of its network modernization plan, and could start to see
network congestion as its smartphone mix increases.

Network Revamp to Generate Savings and Improve


Performance
Network modernization plan will improve coverage, lower costs
Sprint recently confirmed plans to swap out its 3G network and decommission its
iDEN network over time. The company plans to reduce its iDEN footprint by
eliminating sites in limited use and eventually phasing out all iDEN sites, and
believes it can reduce its total cell site footprint by 20k, from 68k today.
As part of the network modernization (officially called Network Vision), Sprint will
install new equipment that utilizes a single base station for multiple frequency bands;
the multi-modal equipment will work with all of the companys 800 and 1900 MHz
spectrum frequencies (and possibly Clearwires 2500 MHz spectrum eventually). In
addition, with the multi-modal equipment, Sprint can begin offering CDMA services
on the 800 MHz spectrum, achieving better propagation (signal strength) and inbuilding penetration.
Meaningful long-term expense savings after extra upfront spend of $4-5 billion
Today, Sprint needs to use dedicated equipment for each spectrum band, resulting in
duplicative maintenance capital and tower rent expenses. The new multi-modal
equipment, however, will enable site decommissioning and thus reduced cell site
expense. In addition, with the 800 MHz spectrum, Sprint can also reduce its need to
roam on other carriers networks given the signal strength of the frequency we
estimate that Sprint currently spends in the range of $1 billion on roaming annually.

146

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

To achieve this goal, Sprint will need to spend $4-5 billion in incremental capital
expenditure and operating expenses over a three- to five-year period. However, the
company estimates savings of $10-11 billion over the expected seven-year time
horizon to complete the project. While spending on the network is likely to start in
the second half of 2011, we dont expect to see any margin improvement until 2013
when CDMA coverage at 800 MHz and new handsets to access it are substantially
rolled out.
Figure 73: Network Transition Plan Spectrum Reconfiguration

Source: Sprint website (used with permission).

$10-11 billion in savings could translate to low- to mid-20s EBITDA margins


Using Sprints $10-11 billion in expected savings, we estimate that the company
expects to save $1-2 billion annually in the mid- to later years of deployment, which
we would consider 2014 and 2015. On our current $23 billion wireless service
revenue estimate in 2014 that would imply margin expansion of 400-800 basis points
from the roughly 17% levels recently, or a low- to mid-20s margin. Instead, we
conservatively estimate wireless margins of only 20% in that year, and only 21%
peak margins when the project is complete.

147

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 74: Network Transition Plan Base Station Reconfiguration

Source: Sprint website (used with permission).

Implications for existing iDEN customers


We estimate that at year-end 2010 Sprint had 10.1 million iDEN customers, down
from 12.9 million 12 months ago. Although this decline is likely to continue as the
network modernization plan begins, the 6.0 million postpaid iDEN customers, in our
view, are particularly important from a strategic standpoint, as many are public safety
and other government employees, and Sprint will work to ensure that they are served
on CDMA. We expect it to take several years for this transition from iDEN to
CDMA to occur, with the shutdown beginning in 2013 and completion in 2017.
The first Q-Chat was doomed by
weak CDMA coverage

148

Second coming of Q-Chat to enable transition


In order to serve iDEN customers on CDMA, Sprint plans to implement a secondgeneration version of the Q-Chat (developed by Qualcomm) software, which enables
push-to-talk (PTT) capabilities on the CDMA technology. Sprint first launched
Q-Chat in mid-2008, but with the weaker CDMA coverage there was limited
adoption and Sprint eventually stopped selling most of the phones (the Motorola
Renegade V950 was launched in September 2008 and is the last CDMA PTT
available). Many push-to-talk users, given the critical nature of their work, need
ubiquitous coverage and the 1900 MHz spectrum used for CDMA services is not
sufficient in some circumstances. However, once Sprint offers CDMA on 800 MHz,
the company expects the push-to-talk abilities to maintain the same dependability
available currently with iDEN.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Expect free cash flow to decline as margins remain tight, capex ramps
Between continued aggressive handset sales and heavy roaming expense, we model
Sprint wireless EBITDA margins to contract to 17.0% for 2011, from an estimated
17.3% for 2010. At the same time, we model consolidated capital spending
increasing to $2.8 billion in 2011 and $3.2 billion in 2012 as Sprint gets into the
heart of its network upgrade. Net-net, we estimate free cash flow for Sprint will
decline to $1.3 billion in 2011 ($0.43 per share or a 9.1% yield) from $1.8 billion in
2010 and $2.7 billion in 2009.
Table 65: Free Cash Flow Trends
$ millions, except per share data
Free Cash Flow
FCF per share
% y-t-y change

2008
1,496
$0.52
-95.3%

2009
2,697
$0.94
78.9%

2010E
1,799
$0.60
-35.7%

2011E
1,276
$0.43
-29.2%

2012E
1,219
$0.41
-4.5%

Source: Company reports and J.P. Morgan estimates.

Operating and Financial Outlook


We expect Sprint to generate meaningful wireless subscriber growth in 2011,
following a return to positive territory in 2010. Most importantly, while prepaid grew
in 2009 and 2010, we expect postpaid to finally generate positive net additions for
the full year in 2011. Specifically, we forecast 225k postpaid net adds in 2011 and
modest improvement in out years, compared to a loss of 888k in 2010. For prepaid,
we forecast 1,250k adds in 2011, down from 1,442k in 2010. Prepaid should
continue to slow in out years, in line with general industry trends. While subscriber
trends should improve, we do not expect meaningful ARPU growth as data
penetration helps to offset dilution from continued prepaid outperformance. Our
operating forecasts drive estimated service revenue growth of 1% in 2011 and 2% in
2012.
On the wireline side, revenue declines are likely to persist with secular pressures
driving less voice spending. In addition, the gradual reduction in business from Time
Warner Cable will also have an impact. However, improved Internet trends should
help support a deceleration in top-line declines. We forecast wireline revenue to fall
8% in 2011 and 3% in 2012, compared to 12% in 2010.

Earnings and Cash Flow Forecast


We believe Sprint has put a significant amount of effort into trimming its cost
structure. With the return of revenue growth, we expect wireless service margins to
begin to stabilize. The company is unlikely to see flat or expanding margins until the
second half of 2011, as gross adds drive some near-term dilution. As a result, we
estimate 30 bps of service margin pressure in 2011 and then 30 bps of expansion in
2012. In line with revenue pressure, the wireline business is expected to see
continued pressure on margins. Sprint recently announced plans to modernize its
network, which will put upward pressure on capital spending. Continued EBITDA
pressure and approximately $700 million of incremental capital expenditures in 2011
are expected to cause a 42% decline in free cash flow in 2011, though we expect cash
flow to remain flat in 2012 as EBITDA growth offsets further increases in capital
spending.

149

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 66: Sprint Nextel Financial Forecasts and DCF


$ millions except per share and per sub values, subs in thousands
CAGR
08-20E

2007

2008

2009

2010E

2011E

2012E

2013E

2014E

2015E

2016E

2017E

2018E

2019E

2020E

Wireless
Direct ending subscribers
% change, y-t-y
Postpaid
Prepaid
Direct net adds
Postpaid
Prepaid
Postpaid churn rate, monthly
Service revenue
% change, y-t-y
Postpaid ARPU, monthly
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin
Capital expenditures
CAPEX/Service revenue
Pre-tax income
Estimated taxes (paid)/benefit
Tax rate

45,329
-1%
40,751
4,578
(658)
(1,224)
566
2.2%
31,044
0%
$59
9,914
-15%
30.9%
5,314
17%
339
(563)
166%

40,275
-11%
36,678
3,597
(5,054)
(4,073)
(981)
2.2%
27,492
-11%
$56
6,776
-32%
23.8%
2,259
8%
(2,095)
652
31%

44,655
11%
33,967
10,688
(994)
(3,546)
2,552
2.1%
25,286
-8%
$55
5,163
-24%
20.0%
1,161
5%
(3,084)
934
30%

45,160
1%
33,079
12,081
554
(888)
1,442
1.9%
25,623
1%
$54
4,468
-13%
17.3%
1,471
6%
(2,510)
966
39%

46,635
3%
33,304
13,331
1,475
225
1,250
1.7%
25,979
1%
$55
4,465
0%
17.0%
2,350
9%
(873)
336
39%

47,960
3%
33,604
14,356
1,325
300
1,025
1.8%
26,589
2%
$55
4,647
4%
17.3%
2,850
11%
119
(46)
39%

49,360
3%
34,004
15,356
1,400
400
1,000
1.8%
27,136
2%
$55
4,979
7%
18.2%
2,985
11%
904
(348)
39%

50,760
3%
34,504
16,256
1,400
500
900
1.8%
27,750
2%
$55
5,447
9%
19.5%
2,775
10%
1,759
(677)
39%

52,060
3%
35,004
17,056
1,300
500
800
1.8%
28,363
2%
$55
5,901
8%
20.6%
2,553
9%
2,371
(913)
39%

53,160
2%
35,504
17,656
1,100
500
600
1.8%
28,926
2%
$55
6,046
2%
20.7%
2,603
9%
2,695
(1,038)
39%

54,160
2%
36,004
18,156
1,000
500
500
1.8%
29,438
2%
$55
6,170
2%
20.8%
2,649
9%
3,016
(1,161)
39%

55,160
2%
36,504
18,656
1,000
500
500
1.8%
29,934
2%
$55
6,261
1%
20.8%
2,694
9%
3,322
(1,279)
39%

56,160
2%
37,004
19,156
1,000
500
500
1.8%
30,429
2%
$55
6,353
1%
20.7%
2,739
9%
3,646
(1,404)
39%

57,160
2%
37,504
19,656
1,000
500
500
1.8%
30,925
2%
$55
6,447
1%
20.7%
2,783
9%
3,991
(1,537)
39%

Wireline
Revenue
% change, y-t-y
EBITDA
% change, y-t-y
EBITDA Margin
Capital expenditures
CAPEX/Service revenue

6,463
-2%
1,074
10%
17%
632
10%

6,332
-2%
1,175
9%
19%
495
8%

5,629
-11%
1,221
4%
22%
258
5%

4,951
-12%
1,071
-12%
22%
218
4%

4,564
-8%
980
-8%
21%
228
5%

4,424
-3%
929
-5%
21%
177
4%

4,332
-2%
910
-2%
21%
173
4%

4,214
-3%
885
-3%
21%
169
4%

4,156
-1%
873
-1%
21%
166
4%

4,153
0%
872
0%
21%
166
4%

4,163
0%
874
0%
21%
167
4%

4,183
0%
878
0%
21%
167
4%

4,215
1%
885
1%
21%
169
4%

4,256
1%
894
1%
21%
170
4%

Consolidated
Revenue
% change, y-t-y
EBITDA
% change, y-t-y
EBITDA margin of service revenue

40,149
-2%
10,800
-15%
29%

35,635
-11%
7,664
-29%
23%

32,260
-9%
6,407
-16%
21%

32,354
0%
5,551
-13%
18%

32,405
0%
5,420
-2%
18%

32,794
1%
5,670
5%
18%

33,161
1%
5,972
5%
19%

33,670
2%
6,404
7%
20%

34,279
2%
6,835
7%
21%

34,877
2%
6,968
2%
21%

35,435
2%
7,082
2%
21%

35,993
2%
7,167
1%
21%

36,561
2%
7,254
1%
21%

37,140
2%
7,346
1%
21%

0%

GAAP EPS, diluted

($10.27)

($0.98)

($0.85)

($1.14)

($0.45)

($0.08)

$0.15

$0.33

$0.45

$0.52

$0.60

$0.66

$0.74

$0.82

0%

14,008
20,485

2010E
NM
NM

2011E
NM
NM

2012E
NM
NM

2013E
30.7x
30.7x

4.6x
5.7x

4.7x
5.9x

4.5x
5.6x

4.3x
5.3x

12.5%
8.5%

8.8%
6.0%

8.0%
5.5%

9.8%
6.7%

Private Market Valuation Summary


2010E
39,240
(4,783)
(19,166)
15,291
13,071
28,362

2011E
34,775
(5,119)
(16,816)
12,840
14,678
27,519

NPV of Wireline cash EBIT 12E -- 20E


NPV of Wireline taxes paid 12E -- 20E
NPV of Wireline and corporate capex 12E -- 20E
NPV of Wireline free cash flow 12E -- 20E
NPV of Wireline 2020 terminal value @ 7x EBITDA
PMV of Wireline operations (ex corporate capex)
NPV of NOLs
PMV of consolidated operations

6,663
(1,199)
(1,311)
4,154
3,120
7,274
3,046
38,682

5,683
(1,052)
(1,082)
3,548
3,120
6,668
3,476
37,662

Plus (less): investments (1)


Less: net debt, end of period (2)
PMV -- Equity

2,787
(15,443)
22,980

2,787
(14,212)
22,761

NPV of wireless EBITDA 12E -- 20E


NPV of wireless taxes paid 12E -- 20E
NPV of wireless capex 12E -- 20E
NPV of wireless free cash flow 12E -- 20E
NPV of wireless 2020 terminal value @ 5x EBITDA
PMV of wireless operations

Current shares outstanding


Shares to be issued/warrants/options
Fully diluted shares outstanding

2,988
3
2,991

2,990
3
2,993

PMV per share


Private to public discount
Target Price - Year-end 2011
Upside to Target Price

$7.68
10%
$6.92

$7.60
10%
$6.84
46.2%

Valuation Multiples
Current Price P/E
Fair Value P/E

$4.68
$6.84
25,433
31,911

Current EV/EBITDA
Fair value EV/EBITDA
Current FCF yieWireline
Fair value FCF yieWireline
Wireless Valuation Method:

1 = DCF from above @ 8.5% discount rate (adjust below)


2 = 6.0x forward EBITDA multiple
6.0x

1
Wireline Valuation Method:

1 = DCF from above @ 7.5% discount rate (adjust below)


2 = 6.0x forward EBITDA multiple
6.0x

100%

Assign % of Market For Clearwire Stake (0-100%)

Sprint Nextel Valuation Grid - Sensitivity to Wireless DCF per Share


Terminal Growth Rate (across) on 2012E Wireless FCF, Discount Rate (down)
$6.84
7.5%
8.0%
8.5%
9.0%
9.5%

0.0%
7
7
6
6
5

0.5%
8
7
7
6
6

1.0%
8
7
$6.84
6
6

1.5%
9
8
7
7
6

Consolidated Valuation Based on Forward FCF Multiple


Consolidated EBITDA
Consolidated Free Cash Flow
Consolidated FCF / diluted share
Consolidated FCF yield (current)
FCF multiple
EV/FCF

Source: Company reports and J.P. Morgan estimates.

150

2010E
5,551
1,799

2011E
5,420
1,276

2012E
5,670
1,219

$0.60
12.9%
7.8x

$0.43
9.1%
11.0x

$0.41
8.7%
11.5x

14.1x

19.9x

20.9x

Valuation Grid - year-end value per share


Operational Free Cash Flow Multiple

8x
10x
12.0x
15x
20x

2010E
Diluted
5
6
$7.23
9
12

10x
12x
15.0x
20x
25x

2011E
Diluted
4
5
$6.40
9
11

2.0%
9
8
8
7
6

3%

NM

1%
0%
0%

-8%

-3%
-2%

-9%

0%

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Clearwire
Initiating with Neutral Rating and $6 YE11 Price Target
We initiate coverage of Clearwire with a Neutral rating and a December 2011 price
target of $6.00/share, implying approximately 7% upside from current levels. In our
view, the company has a tremendous asset base in its wireless spectrum and
wholesale relationships, which are being diluted by an overly aggressive retail effort
that is actually preparing to ramp further in 2011. This, combined with the
companys continuing funding needs if it wants to continue to build out, forces us to
the sidelines.

Neutral
Clearwire (CLWR;CLWR US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End Date

5.70
12 Jan 11
8.82 - 4.63
5.62
Dec
986
6.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

(0.38)
(0.38)
(0.43)
(0.55)
(1.75)
(1.40)

(0.48)A
(0.13)A
(0.58)A
(0.60)A
(1.79)A
(2.18)A

(0.70)
(0.71)
(0.75)
(0.60)
(2.74)
(1.95)

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on GAAP results.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Clearwires spectrum is great asset and competitive advantage
Clearwire has an average of roughly 140 MHz of spectrum across the United States
the deepest spectrum portfolio of any consumer wireless company. While the
spectrum is in a band in which no other US carrier operates, that band is also a global
standard for 4G services and we think that over time Clearwire has a chance to be
successful using it.
Operations have been a commercial success so far
Clearwire currently has 2.8 million customers split between 1.8 million wholesale
customers (mostly from Sprint) and 1.0 million retail broadband customers, and has
raised guidance each quarter in 2010. While Clearwires retail subs are to be
expected given its substantial build and aggressive retail efforts, the wholesale
subscriber ramp is a testament to how much partners Sprint, Comcast, and Time
Warner Cable (among others) want to market the companys 4G services.
Clearwire needs $7 billion to complete buildout and operate to positive FCF
After raising $1.4 billion in December in a debt and convert offering, Clearwire
currently has roughly $2.0 billion in estimated cash. If it cuts expansion capex and
instead continues to market only its current footprint, Clearwire should be able to
operate for years, but the greater opportunity for the company a national 4G
network with tremendous capacity and large, dedicated wholesale channels likely
would be squandered. Instead, we model that Clearwire will need an additional
$7 billion in cash to build another 105 million pops before reaching positive FCF in
2016.

151

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Sprint relationship needs clarification for both companies to move forward


Sprint owns 54% of Clearwires shares from an asset swap in 2008, but does not
have effective control of the company (this is not counting the conversion of recently
issued out-of-the-money exchangeable notes). This has resulted in a duplicative build
effort for Clearwire and Sprint, which needs to upgrade its network with similar
equipment to that in Clearwires build. In addition, Sprint (and most other strategic
investors) would like Clearwire to focus its spending more on network coverage and
less on retail efforts, and in particular to pull back from launching smartphones. We
believe that Clearwire may need to either raise cash away from Sprint if possible, or
integrate its plans more closely with Sprint to maximize value for both companies
shareholders.

Investment Risks
Upside risks
Sprint could seek full control of Clearwire sooner than expected or at premium
There have been reports that Sprint could seek full control of Clearwire (such as a
September 15, 2010 Financial Times article) as a clean resolution the relationships
issues. However, Sprint has indicated that it has no immediate plans to do so but is
likely to continue to infuse some equity over the coming years and maybe some way
down the road take control of it. Sprint could seek control of Clearwire sooner than
expected, or pay a premium to do so.
Another large wholesale partner could invest in Clearwire
Instead of Sprint, other large potential wholesale partners could choose to invest in
Clearwire to expand the build and get preferential access to the companys network.
If this were to happen it could put a comp on the spectrum value, and assuage
funding concerns as well as worries about the Sprint relationship.
Clearwire could sell spectrum at a premium
Clearwire has stated that it is the process of evaluating the sale of some of its
spectrum. Were that to happen at a premium, it could put a public comp on the
spectrum that indicates substantially more value than the current enterprise value of
$0.22/MHz/pop.
Downside risks
Clearwire may have to stand on its own
If Clearwire cannot find additional near-term funding it will need to halt its buildout
and possibly reduce marketing spending to preserve cash while it works toward
getting to positive FCF. This would impact our growth expectations substantially.
Lightsquared may sign meaningful partnerships, reducing potential wholesale
business for Clearwire
A new carrier called Lightsquared is working to build a national 4G LTE network
using a combination of satellite and terrestrial platforms. If it is successful in
building and attracting wholesale partners, Lightsquared could either win away or
compete down pricing for incremental wholesale partners at Clearwire.

152

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Sprint could choose to build its own 4G network, relying on Clearwire only for
excess capacity
Sprint and Clearwire are currently in a dispute about how much Sprint owes
Clearwire for smartphone wholesale service a dispute that Clearwire could lose.
Furthermore, the mismatched strategies between Sprint and Clearwire could drive
Sprint to put 4G service on its own network as part of its modernization. While
Sprint does not have enough spectrum long term, reducing Clearwire roaming
substantially could meaningfully impact Clearwires revenue stream.

Company Description
Clearwire is an emerging wireless carrier, focused on providing 4G wireless
broadband services. The company currently offers its services to approximately
120 million people with its WiMAX network, up from 45 million 12 months ago. In
addition to operating its own retail business, the company leverages a variety of
wholesale partners to expand distribution. These wholesale partners include equity
investors Sprint Nextel, Comcast, and Time Warner Cable as well as non-investors
Best Buy and cBeyond. Other investors in the company include Intel, Google, and
Bright House Networks. For the third quarter, the company reported a subscriber
base of 2.8 million, including 1.0 million retail subscribers and 1.8 million wholesale
subscribers. Revenue on that base was $147 million, representing year-over-year
growth of 114%.

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $6.00/share for Clearwire. We based our
target price primarily on a discounted cash flow analysis that assumes a 10% WACC
and 5.0% perpetual growth rate. Our $6.00 target price also implies a value of
$0.31/MHz/pop for Clearwires spectrum assuming all cash is spent a substantial
discount to recent transactions and spectrum auctions. Unfortunately, because the
company is forecast encounter heavy EBITDA losses for the next few years at least,
we are unable to evaluate the company on a multiples basis relative to peers.

In Need of CashAgain
Current equity ownership structure still has Sprint >50%
Clearwires investor base consists of a variety of investors including Sprint Nextel,
which holds a 54% stake, and Comcast, Time Warner Cable, Bright House
Networks, Intel, Google, and Eagle River, which collectively hold a 31% stake.
The remaining 15% public float is heavily shorted short interest was 24 million
shares (approximately 10% of float) on November 30, before the recent convert
offering, and rose to over 31 million (13% of the float) as of December 15.
Clearwire can operate existing
markets, but not build, at its
current funding level

Funding remains primary dilemma


Funding continues to be Clearwires biggest near-term hurdle, despite the companys
recent issuance of $1.4 billion in convertible and secured debt in December. While
we believe that cash could get the company through if it stopped building cities, if
management intends to launch new markets in the next couple of years, we expect it
to need more capital. In addition, in order to eliminate the current funding overhang
on the shares, we believe that Clearwire will need substantially more capital.

153

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

We estimate at least $3.0 billion of incremental cash is needed for the company to
comfortably build additional markets (30-50 million pops) through the end of 2012,
though almost $7 billion may be needed to build to our estimated terminal coverage
of 225 million pops from the 120 million covered by year-end 2010 and make it to
FCF breakeven. We model Clearwire building 30 million pops in 2011, 30 million in
2012, 20 million in 2013, 10 million in each 2014 and 2015, and another 5 million in
2016.
Table 67: Clearwire Sources and Uses of Cash
$ millions

EBITDA
Interest expense

2011E
(1,613)
(395)

2012E
(1,388)
(539)

2013E
(678)
(723)

2014E
134
(843)

2015E
1,159
(843)

2016E
1,898
(843)

Cumulative
(490)
(4,188)

Cash from operations

(1,561)

(1,471)

(934)

(237)

797

1,547

(1,859)

Maintenance capex
CPE capex
Buildout capex
Capitalized interest
Total capex

(135)
(166)
(750)
(160)
(1,211)

(230)
(197)
(750)
(160)
(1,337)

(330)
(220)
(500)
(120)
(1,170)

(490)
(221)
(250)
0
(961)

(673)
(217)
(250)
0
(1,140)

(845)
(209)
(125)
0
(1,179)

(2,704)
(1,231)
(2,625)
(440)
(6,999)

FCF
Debt Issuance
Equity Issuance

(2,772)
1,200
1,800

(2,808)
1,200
1,800

(2,104)
0
0

(1,198)
0
0

(344)
0
0

367
0
0

(8,859)
2,400
3,600

Net Cash Flow

228

192

(2,104)

(1,198)

(344)

367

(2,859)

Source: J.P. Morgan estimates.

Additional equity investment from Sprint is one possible solution


We believe Clearwire has a few different potential sources of additional capital. An
additional equity investment may be the most likely, but the likelihood of this
happening in the near term is by no means certain. The most probable investor likely
would be Sprint, given the companys current 54% ownership and public
commitment to using Clearwires 2.5 GHz spectrum for 4G services. Management at
Sprint has stated publicly that it is willing to allocate capital to future investments in
the company.
However, we believe the two partners disagree on critical operational matters, while
agreement on equity valuation also appears to be an ongoing source of conflict.
Without serious competing interest from other potential strategic investors, we do not
expect Sprint to easily agree on valuation with Clearwire management.
Nonetheless, in the long run, we do think Sprint is likely to eventually seek control
of all of Clearwires assets. In order to effectively and efficiently drive its 4G
strategy, Sprint likely would need control, in our view. Currently, Clearwires board
and management can operate independently. Recently, three members of Sprint
management were forced to resign from Clearwires board due to potential conflicts
of interest. Therefore, Sprint now only holds the right to nominate seven of 12
independent board members, but does not have any direct voting rights. Most of the
remaining board members are also independent.

154

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Other strategic investors unlikely for now


Given a mostly independent board, Clearwire could sign a wholesale agreement with
virtually any party it sees fit. Unfortunately, there does not appear to be a surplus of
substantial interest, and Clearwire may have to bide its time for now. The current
cable strategic partners have all indicated an unwillingness to commit further capital
to the joint venture. T-Mobile USA, as confirmed by management and press reports,
has expressed interested, but has also said that it sees no need to partner or acquire
4G capacity for 2-3 years.
Exploring a spectrum sale
Outside of an equity investment, Clearwire is contemplating selling a portion of its
spectrum and has stated that it is in the process of evaluation. Notably, the sale of this
spectrum would not violate any equity-holder rights of the companys strategic
partners because it is not deemed to be critical to offering current or future services.
Clearwire is currently holding an auction process, which we believe is likely entering
the later stages and could be completed by the first quarter of 2011. Management is
open to selling the spectrum in pieces to a variety of partners, which should bring
flexibility to the process.
Overall, though, we would not expect a significant amount of interest in this
spectrum and would not be surprised to see Clearwire eventually pull back from a
sale.
Significant spectrum sale seems unlikely. Management has indicated a willingness
to sell up to $2 billion or up to 20% of its spectrum holdings, representing what
management believes to be excess spectrum. A successful sale could result in a
premium spectrum valuation that could cause shares of Clearwire to be re-rated
higher. However, we do not expect a sale of this magnitude. Though the industry
clearly needs additional spectrum over the long run, current market prices for some
pieces of spectrum remain somewhat depressed relative to historical auction and
transaction values. This seems to be particularly true in second-tier markets which do
not face as many bandwidth constraints.
Furthermore, if no other funding sources are available, a sale could enable a longterm competitor for the company or reduce the companys ability to partner with
additional wholesalers. In addition, a reduced supply of spectrum could increase the
cost of adding additional capacity down the line. Management has already had to add
capacity to the network far earlier than originally expected.
Previous spectrum sales are
poor comps to Clearwires
2.5 GHz spectrum

Table 68: Prior Spectrum Transactions

History of spectrum prices


AT&T/Qualcomm sale
Auction 73 (700 MHz)
AT&T/Aloha
Auction 66 (AWS)
Auction 58 (PCS)
Auction 35 (PCS, cancelled)

Date Frequency Total $ amount ($m)


2010
700 MHz
1,925
2008
700 MHz
14,827
2007
700 MHz
2,500
2006 1.7/2.1 GHz
13,879
2005 1900 MHz
2,250
2001 1900 MHz
16,300

$/MHz/pop
$0.86
$1.29
$1.06
$0.54
$1.06
$2.11

Source: FCC, Company reports, and J.P. Morgan estimates.

155

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Clearwires Retail Initiatives Dilute Wholesale Value


We are generally favorable on the Clearwire 4G buildout and wholesale opportunity,
but believe there is less need for the company to have an aggressive retail effort. In
the retail market, the company must either compete as a wireless carrier against huge
and well-funded incumbents (including Sprint) in a saturated market in which are all
talking about 4G as well, or position itself as a broadband alternative against wellfunded and entrenched incumbents (including its investors Time Warner Cable and
Comcast) in an already saturated broadband market.
Most of Clearwires strategic partners would prefer the company to capitalize on its
opportunity as a carriers carrier for 4G, and keep losses at the retail effort to a
minimum.
Clearwire, Sprint, and cable partners all selling mobile broadband services
Clearwire is currently running a retail business, selling similar services to those sold
by its wholesale partners. These products and services include 3G/4G USB dongle
data devices for mobile broadband use, mobile hotspot devices that can create a
mobile WiFi network, and residential modems for fixed broadband.
In addition to these data-only devices, Sprint also sells smartphones that can access
Clearwires network. The three to date are the HTC EVO and Samsung EPIC, as well
as the recently launched EVO Shift. Clearwires cable partners primarily sell the
mobile broadband services at varying price points with required bundles.
Retail is a drag on cash
Aside from the duplicative services being offered, Clearwire must also build and
support a store and distribution network to support its retail efforts. Thus far, our
store checks reveal little traffic in the companys stores. In our view, Clearwires
investment in its own retail distribution system is unlikely to justify the sales volume
it is likely to see, particularly over the long run. The company is likely to struggle to
grow the retail business in a saturated broadband and wireless voice market against
very strong, established incumbents.
Many of Clearwires wholesale partners share the same sentiment, creating a
significant point of contention among the parties. Managements view is that the
retail business helps justify economic terms for the wholesale business, as the
strategic partners compensate Clearwire for network use by paying retail terms less
some discount. However, the company could also consider setting these terms
through a far less aggressive retail rollout and advertising campaign.
Smartphones seem a move away
from Clearwires competitive
advantage

156

Offering smartphones will increase retail cash drag


Management believes smartphones will be an important component to the retail
business, and 4G generally. Smartphone economics, however, are different relative to
those for a traditional mobile broadband data card or dongle business. Unlike a
dongle which Clearwire can give the customer upfront and charge $45 per month to
recoup its cost, smartphones are heavily competitive and small carriers generally
have little room to raise prices. Instead, smartphones require large upfront subsidies,
which can be as much as $300-400 for national carriers, in order to entice customers
to sign a two-year contract. As a much smaller operator, Clearwire is likely to be
charged higher wholesale prices for devices, relative to those given to national
carriers.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

In addition, we believe most of what remains of Clearwires network speed and


capacity advantage is likely to be lost in the smartphone market, in which few
applications will drive the need for >5 Mbps speeds and unlimited capacity that
Clearwire offers. The company is also likely to see a price disadvantage to
companies such as T-Mobile and AT&T that offer smartphone service on a standard
3G handset which is far less expensive that the CDMA/WiMAX handsets that
Clearwire will need.

Smartphone Wholesale Dispute with Sprint


In the third quarter, Clearwire recognized $4.46 of wholesale ARPU, reflecting a mix
of revenue on wholesale smartphone and data card customers. However,
management noted that ARPU would have been closer to $9.00, reflecting an
incremental $17 million in revenue, if not for a dispute regarding wholesale revenue
received from Sprint. Specifically, the disagreement deals with smartphones, which
drive lower ARPU than data cards but we expect to represent a significant portion of
future wholesale sales. We believe smartphone revenue consists of set payments
Sprint must make per subscriber as well as usage-based payments. However, in this
scenario, it is unclear if it is the fixed pricing or usage component that is driving the
discrepancy.
Significant revenue and margin implications
If the two parties cannot agree on revised wholesale terms, the disagreement is likely
to be resolved through arbitration in the first half of 2011. Given rising smartphone
data usage across the industry, we believe the outcome has significant implications
for Clearwires profitability. Smartphones are likely to represent the majority of the
companys wholesale subscriber base in the long run, and wholesale should represent
the vast majority of the companys total subscribers. Therefore, we think it is critical
that Clearwire extract adequate revenue from Sprint, or future profitability could be
significantly depressed.

WiMAX vs. LTE Not Much Debate Anymore


We expect Clearwire to move to
LTE in the next two years

There are two main flavors of 4G LTE which is likely to be used by the majority of
wireless carriers globally and WiMAX which was a little faster to market but which
we expect to fade over time. Clearwires original expectation was that its choice of
WiMAX technology would allow it to come to market far faster than LTE carriers,
but between the companys delays in getting to market and the faster LTE ramp from
companies like Verizon, we see Clearwire as having only a small advantage in
devices today and expect that to fade by the end of 2011 as the LTE ecosystem grows
to dominate the 4G landscape.
Trialing LTE in Phoenix
Clearwire has rolled out 4G coverage to 120 million people using the WiMAX
technology. However, the company is also currently trialing two flavors of LTE in its
Phoenix market. Specifically, the company has tested the ability for WiMAX and
LTE to coexist under Frequency Division Duplex (FDD) and Time Division Duplex
(TDD) implementations. The FDD configuration, termed LTE 2X, would use
40 MHz of spectrum, consisting of two paired 20 MHz channels, while the TDD
configuration would use just a 20 MHz channel that does both upload and download.

157

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Eventual LTE upgrade is logical; most equipment can be re-used


We believe it is in Clearwires (as well as Sprints) best interests to ultimately pursue
LTE as a replacement for WiMAX. Clearwire may decide to pursue this route, but is
waiting to complete testing and more importantly must resolve funding concerns.
Most of the equipment that Clearwire is rolling out today is either upgradable to LTE
(through an additional card in the case of a base station) or is technology-agnostic
and can be re-used with no changes, e.g., antennas and backhaul.
We expect new coverage deployed with LTE to cost the same as the $25 per pop
Clearwire spends to build a market today, and a fraction of that to upgrade markets
that have already been covered with WiMAX.
LTE may also make it much more feasible for potential wholesale partners to work
with and perhaps invest in Clearwire. Most smaller operators, including T-Mobile
USA, US Cellular, MetroPCS, and Leap Wireless, have already committed to using
LTE for their 4G strategy.

Operating and Financial Outlook


Clearwires subscriber additions and revenue ramp will largely depend on the
companys ability to continue building out its network, as well as its relationship
with its strategic partners. We expect rapid subscriber growth to continue, though at a
modestly slower pace than in 2010. We forecast net adds of 4,600k in 2011 and
4,634k in 2012, with the majority coming from wholesale partners. We expect retail
ARPU to remain relatively stable at approximately $43. However, wholesale ARPU
could have greater variability. We expect Sprint smartphone subscribers to account
for a large part of Clearwires wholesale business, and currently the two partners are
disputing economics on these subscribers. The outcome of those negotiations could
have a large impact on the companys growth trajectory. Currently, we forecast
wholesale ARPU to climb to the $9.50 range in 2001 from $6.59 in 2010. On a
consolidated basis, we forecast revenue growth of 142% in 2011 and 70% in 2012.

Earnings and Cash Flow Forecast


Given heavy operating and capital investments due to startup build and marketing
costs, as well as an expensive retail business, we do not expect Clearwire to generate
positive EBITDA or free cash flow for a few years. We expect 2011 to reflect peak
EBITDA loss for the entire business, though 2012 and 2013 are expected to show
heavy operating investments as well. The business could turn EBITDA positive in
2014, depending on the timing of additional market launches. Management has stated
that markets on average should reach positive EBITDA with 18 months. Positive free
cash flow positive should be generated in 2016, a couple of years after EBITDA
breakeven, as capital spend begins to approach maintenance levels.

158

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 69: Clearwire Financial Forecasts and DCF


$ millions except per share and per sub values, subs in thousands
Covered pops

2005
3,788

2006
5,882

Retail metrics
Ending subscribers
% change, y-t-y
Retail penetration
Net adds
Churn
ARPU

56
184
228%
1.5%
2.1%
56
128
2.0%
1.7%
$31.99 $36.53

2007
13,900

2008
16,800

2009
44,700

350
90%
2.5%
166
1.9%
$40.66

475
36%
2.8%
74
3.7%
$40.71

688
45%
1.5%
170
3.2%
$44.84

Service revenue
% change, y-t-y
Cash EBITDA (operating cash flow)
% change, y-t-y
Cash EBITDA margin
Taxes paid
EPS - diluted
Capital Expenditures and Spectrum
Capital expenditures
CAPEX/service revenue

Less: exchangeable debt (1)


Less: net long-term debt, end of period (2)
PMV of equity
Year-end 2011 shares outstanding
Additional shares from exchangeable debt (1)
Shares to be issued from warrants/options/convert (3)
Fully diluted shares outstanding
DCF Value per Share
Private-public discount of 10%
Price Target
Upside %

CAGR
09-20E
16%

2,013
63%
1.3%
775
3.5%
$42.78

3,013
50%
1.7%
1,000
3.0%
$43.21

4,213
40%
2.1%
1,200
2.7%
$43.63

5,413
28%
2.6%
1,200
2.5%
$43.41

6,573
21%
3.0%
1,160
2.4%
$43.19

7,693
17%
3.4%
1,120
2.3%
$42.97

8,773
14%
3.9%
1,080
2.2%
$42.75

9,813
12%
4.4%
1,040
2.1%
$42.54

10,813
10%
4.8%
1,000
2.0%
$42.32

11,773
9%
5.2%
960
2.0%
$42.11

0%
46
$22.10

3,029
6485%
3%
2,983
$6.59

6,854
126%
5%
3,825
$8.27

10,488
53%
6%
3,634
$9.39

13,758
31%
7%
3,270
$9.50

16,701
21%
8%
2,943
$9.45

19,533
17%
10%
2,832
$9.40

21,965
12%
11%
2,432
$9.35

23,997
9%
13%
2,032
$9.31

25,629
7%
15%
1,632
$9.26

27,061
6%
17%
1,432
$9.21

28,493
5%
19%
1,432
$9.17

734
55%
1.6%
216
192%

4,267
481%
3.6%
3,657
1593%

8,867
108%
5.9%
4,600
26%

13,501
52%
7.5%
4,634
1%

17,971
33%
9.0%
4,470
-4%

22,114
23%
10.5%
4,143
-7%

26,106
18%
11.9%
3,992
-4%

29,658
14%
13.2%
3,552
-11%

32,770
10%
14.6%
3,112
-12%

35,442
8%
15.8%
2,672
-14%

37,874
7%
16.8%
2,432
-9%

40,266
6%
17.9%
2,392
-2%

44%

68
136
216
268
560
101%
59%
24%
109%
700%
(194)
(289)
(387)
(781) (1,264)
-79%
-49%
-34%
-102%
-62%
-287.4% -213.0% -178.9% -291.8% -225.9%
(1)
(3)
(5)
(5)
(1)
(1)

1,352
142%
(1,268)
0%
-93.7%
-

2,305
70%
(1,034)
NM
-44.9%
-

3,299
43%
(315)
NM
-9.5%
-

4,260
29%
507
3728%
11.9%
-

5,179
22%
1,544
148%
29.8%
-

6,036
17%
2,294
45%
38.0%
-

6,819
13%
2,933
22%
43.0%
-

7,529
10%
3,289
15%
43.7%
-

8,179
9%
3,621
13%
44.3%
-

8,790
7%
3,908
11%
44.5%
-

37%

46

62
1.6%
56

206
231%
3.5%
128
128%

350
70%
2.5%
166
29%

475
36%
2.8%
74
-55%

8
NM
(109)

($1.97)

($3.13)

($4.58)

($2.96)

($1.75)

($1.79)

($2.74)

($2.35)

($1.82)

($1.52)

($1.01)

($0.71)

($0.40)

($0.24)

($0.07)

$0.10

133
1570.5%

192
283.7%

389
286.4%

261
120.9%

1,183
442.1%

2,461
439.8%

1,051
77.8%

1,177
51.1%

1,050
31.8%

961
22.6%

1,140
22.0%

1,179
19.5%

1,175
17.2%

1,208
16.0%

1,299
15.9%

1,391
15.8%

Private Market Valuation Summary


NPV of Cash EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of spectrum acquisitions 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value @ 6.7x EBITDA
PMV of operations

2011E
2012E
2013E
2014E
2015E
2016E
2017E
2018E
2019E
2020E
150,000 180,000 200,000 210,000 220,000 225,000 225,000 225,000 225,000 225,000

1,238
80%
1.0%
674
3.3%
$43.43

Wholesale metrics
Ending subscribers
% change, y-t-y
Wholesale penetration
Net adds
ARPU
Consolidated metrics
Ending subscribers
% change, y-t-y
Penetration (of cov. pops)
Net additional subscribers
% change, y-t-y

2010E
120,000

Current Share Price Multiples @


2010E 2011E
6,809
8,077
0
0
0
0
(7,567) (6,515)
(757) 1,562
10,656 10,656
9,898 12,217

(3,406)
6,493
986
3
989
$6.57
10%
$5.91

EV EOY @ current price


Domestic Licensed Pops (mm)
EV/MHz/Pop (dom)
EV/Subscriber
EV/Revenue
EV/EBITDA

2008
6,158
285
$0.14

$5.61
2009
6,869
285
$0.16

2010E
10,099
285
$0.23

2011E
11,071
285
$0.26

2012E
12,079
285
$0.28

2013E
14,183
285
$0.33

2014E
15,381
285
$0.36

2015E
15,725
288
$0.37

2016E
15,358
288
$0.36

$12,965
28.5x
NM

$9,359
25.7x
NM

$2,367
18.0x
NM

$1,249
8.2x
NM

$895
5.2x
NM

$789
4.3x
NM

$696
3.6x
30.3x

$602
3.0x
10.2x

$518
2.5x
6.7x

5,738
$0.13

6,449
$0.15

9,678
$0.23

10,650
$0.25

11,658
$0.27

13,762
$0.32

14,961
$0.35

15,304
$0.36

14,937
$0.35

(729)
(3,112)
8,376

EV EOY @ target price


Target Price EV/MHz/Pop

1,286
0
3
1,289

Valuation Grid - DCF Value per Share


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)

$6.50
10%
$5.85
4%

6
8.5%
9.5%
10.5%
11.5%
12.5%

2.0%
7
5
4
3
2

3.0%
9
7
5
3
2

4.0%
11
8
$5.85
4
3

5.0%
15
10
7
5
4

29%

17%
-1%

37%
-8%

NM

1%

6.0%
22
14
9
6
5

(1) Exchangeable debt convertible at average $7.08 counted as 0 million dilutive shares and $729 million debt.
(2) Debt is net of cash and cash equivalents.
(3) 3.0 million dilutive options/warrants net of 3.5 million shares redeemed (treasury method).

Source: Company reports and J.P. Morgan estimates.

159

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

160

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Leap Wireless International


Initiate with Overweight Rating and $16 YE11 Price Target
We initiate coverage of Leap Wireless with an Overweight rating and a year-end
2011 price target of $16/share, implying 17% upside from current levels. Leap has
executed poorly in the last 12 months but changes started in 3Q and completed in 4Q
should put the company back on a growth path in 1Q11. From a profitability and
cash flow perspective, Leaps margins have been under pressure through the
transition, but as churn comes down and ARPU rebounds, it appears poised for
margin expansion. Free cash flow is likely to ramp meaningfully in 2011 as well, as
capital expenditures decline and move closer to maintenance levels.

Overweight
Leap Wireless International (LEAP;LEAP US)
Company Data
Price ($)
13.19
Date Of Price
12 Jan 11
19.11 - 9.51
52-week Range ($)
Mkt Cap ($ bn)
1.00
Fiscal Year End
Dec
76
Shares O/S (mn)
Price Target ($)
16.00
Price Target End Date 31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

(0.74)
(0.89)
(0.85)
(0.82)
(3.30)
(3.06)

(0.85)A
(0.24)A
(7.06)A
(0.99)A
(9.15)A
(4.44)A

(0.86)
(0.51)
(0.64)
(0.85)
(2.87)
(1.78)

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on GAAP results.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Theres nothing wrong with playing follow the leader
Leap Wireless revamped its price and handset offering in the second half of 2010,
with pricing very similar to changes made by MetroPCS in 1Q10 that were very
successful. This was evidenced somewhat by the 115k net additions in 4Q10 that
Leap announced on January 4, and we expect that between a stronger prepaid
environment in 1Q and the vastly improved handset lineup Leap will be able to add
175k customers in 1Q11 and 400k in 2011.
Better subs, ARPU increases, churn declines, capex declines all indicators of
wireless stock outperformance
Leap has struggled in the last year but the companys traction with new pricing
should increase subscriber growth through both better gross addition traction and
lower real and false churn. In addition, Leap expects ARPU to increase as
customers come in on $50-60 rate plans vs. the ~$30 level at which incremental
customers were joining in 1H10. Finally, capital spending should decline somewhat
in 2011 as Leaps 3G expansion pauses and the company waits to ramp an LTE
overlay until handset prices start to scale.
3G network will be an advantage in 2011 vs. competitors pushing 4G ecosystem
Leap has a CDMA EVDO Rev A network that can offer data speeds in the ~1 Mbps
range a decent speed on a data card and we believe plenty on a smartphone. The
CDMA handset ecosystem is very mature and we believe that handset prices could
come down dramatically for Leap in 2011, allowing the companys customers to
upgrade to an attractive data-capable handset for as little as $50 by the end of 2011.
This should help support ARPU through 2011.
161

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Investment Risks
Prepaid wireless remains highly competitive and heavily levered to employment
and economic levels
The prepaid and pay-in-advance industry has experienced a variety of pricing moves
over the past two years. We believe such moves were brought on by competition, as
players aggressively tried to maintain growth. Leap recently took action to follow the
market by bundling taxes and fees into its rate plans. In addition, employment levels
remain subdued and if employment were to start to fade again Leaps low-credit
demographic could be impacted disproportionately. Should the company need to
make additional pricing changes, profitability and cash flow could be negatively
affected.
Tight spectrum market could create additional costs
Leap is likely to encounter capacity constraints in some markets, as its 3G network is
filled up and it wants to roll out 4G service. The company needs to purchase
additional spectrum, yet we remain concerned that attractive spectrum is likely to be
either unavailable or too expensive. As a result, the company could find itself
overpaying for spectrum resources, investing more capital in its existing network to
increase capacity, or limiting customer usage levels further.
MVNO agreement creates $300 million liability to Sprint
Leap signed an MVNO agreement in 3Q10 to offer service outside its core that
guarantees Sprint a minimum of $300 million over the next five years. Leap does not
expect to ramp this effort until 2H11 and if it is not successful Leap will still owe
Sprint the balance. This agreement could be looked at as a debt-like obligation.
Billing system transition creates risk for cash flow, business model stagnation as
it happens
Leap is in the process of updating its billing system and expects to transition in the
second quarter of 2011. Billing systems transitions create inherent risk in services
companies and have been responsible for high customer service call rates, billing
errors, and higher opex. In addition, Leap will need to freeze its pricing offers for
some time while the transition happens, which could lead to the company being less
than competitive or management distraction in 2011.

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $16/share for Leap Wireless. We based
our target price primarily on a discounted cash flow analysis that assumes a 10%
WACC and 2.5% perpetual growth rate. In addition, Leap currently trades at 5.4x our
2011 EBITDA forecast, in line with the wireless industry average despite offering
higher growth. Leaps closest peer, MetroPCS, trades at 5.9x our 2011 estimate. In
our view, both companies offer attractive growth which we believe justifies higher
multiples.

162

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Company Description
Leap Wireless is one of two companies in our coverage that focus on unlimited
wireless voice and data services on a pay-in-advance basis for $35-60 per month. The
company has historically targeted the low-income, minority, and youth
demographics, though increasingly other customers are also signing on to the service.
While Leap started as a local carrier, it today covers about 94 million pops in markets
including Las Vegas, Phoenix, Denver, Washington, D.C., Baltimore, Philadelphia,
and Chicago and offers free roaming to customers in most of the country through
roaming partnerships. Leap today has 4.8 million voice customers and 600k data
customers who pay an average of $36.71 per month. The company emerged from
bankruptcy in 2004.

Figure 75: Leap Wireless Coverage Map Including Roaming

Source: Company website (used with permission).

163

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Leap Follows Metros Lead on Voice Pricing


In early August at the companys analyst day, Leap management announced new
voice and mobile broadband price plans. The voice plans essentially mimic pricing
established by MetroPCS in early January, with a focus on moving customers upmarket to better phones and services.
The new plan lineup consists of four all inclusive price points ($35, $45, $55, and
$60) that include taxes and fees. The notable exclusion is a $40 price point, which is
a popular price point offered by Metro as well as Tracfone (Straight Talk). In
addition to new pricing, Leap also eliminated offering the first month of service free
but redirected that money to handset subsidies, which effectively equalizes pricing
for new and existing customers. These changes were driven by three motivations:
driving improved market share, lowering churn, and raising ARPU.
Table 70: Pay-in-Advance and Prepaid Price Comparison Summary
LD
X
X
X
X

SMS
X
X
X
X

Data

$35
$45
$55 Android
$55 Blackberry

Talk
X
X
X
X

X
X
X

Taxes
~10%
$35.00
$45.00
$55.00
$55.00

$40
$45
$50
$60
$60 Blackberry

X
X
X
X
X

X
X
X
X
X

X
X
X
X
X

X
X
X
X
X

$40.00
$45.00
$50.00
$60.00
$60.00

$50

$50.00

Talk
$0.10/min
1500 Talk
X
X

LD
X
X
X

Text
X
X
X
X

Data

$15
$30
$50
$70

30 MB data
100 MB data
2 GB data

$25
$40
$60

Talk
300
1200
X

LD
X
X
X

SMS
X
X
X

Web
X
X
X

$27.50
$44.00
$66.00

$30
$45

1000
X

X
X

1000
X

30 MB
X

$33.00
$49.50

$15
$30

200
500

X
X

$16.50
$33.00

$10
$20
$30

50
125
200

X
X
X

$11.00
$22.00
$33.00

$10
$25
$50
$30
$45

100
400
1000
1200
unlimited

X
X
X
X
X

$10.00
$25.00
$50.00
$30.00
$45.00

Leap

PCS

Boost

T-Mobile

Virgin Mobile

Taxes
$16.50
$33.00
$55.00
$77.00

Straight Talk

Net-10

TracFone Value Plans

Page Plus

Source: Company websites.

164

1200
X

50 MB
20MB

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

4Q trends saw turnaround in voice additions


Third quarter results remained weak, with Leap losing 200k subscribers, including
28k broadband and 172k voice customers. However, the new price plans were not
fully rolled out until the middle of the third quarter and handset supply issues in the
second quarter carried into 3Q. Without much traction in the new price plans, ARPU
also remained under pressure, falling 6.4% year-over-year.
However, management has already preannounced 115k net additions for 4Q10,
including 163k voice adds and 48k broadband losses, and noted that late-quarter
trends reflected better performance, especially around churn. Handset supply,
including availability of smartphones, increased and helped drive adoption and take
rates of the higher-priced plans improved. Gross adds remain depressed though, as
customers (including existing) found less incentive to sign up for service or
upgrade (false gross add) without the initial month of free service.
Financials expected to be mixed. ARPU may rebound somewhat given new pricing
but likely will remain depressed as a result of the late-quarter growth, as well as the
acquisition of low-ARPU Pocket subscribers. However, given the better handset
selection and lower prices, we expect handset upgrade rates to rise dramatically to
the double-digit levels (MetroPCS peaked around 12%) and high marketing costs as
well. This is likely to drive 4Q margins substantially below 3Qs 21.8%, and we
model 19.6% in 4Q with a rebound to 22.4% in 1Q.
likely setting Leap up for a strong 1Q
The fourth quarter is usually a good one for prepaid wireless, but the first quarter is
typically Leaps best period, with the target low-income subscriber base having more
discretionary capital due to tax returns which are issued in late January through early
March. With the false churn hangover washed through the system in 4Q and a strong
handset selection, net adds should be significantly higher as gross add volume
improves and churn continues to fall.
We forecast 1Q11 net adds of 175k, driven by a 7% sequential increase in gross adds
to 790k and a 70 bps sequential (and year-over-year) decline in churn to 3.8%. Given
how quickly Leap has achieved high penetration with smartphones, we expect ARPU
to increase year-over-year for the first time since late 2007.
Figure 76: Net Adds and Churn
Subscribers in thousands
500
400
300
200
100
0
(100)

6.00%

5.50%

5.40%

5.50%

5.00%
4.70%

5.00%

4.50%

4.50%

4.40%

4.50%
3.80%

4.00%
3.50%

(200)
(300)

3.00%
2Q09

3Q09

4Q09

1Q10
Net adds

2Q10

3Q10

4Q10E

1Q11E

Reported churn

Source: Company reports and J.P. Morgan estimates.

165

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 77: ARPU and Growth


$42.00

$40.74

$38.00
$36.00

3.1%

$39.24

$40.00

$37.94
$36.43

-3.8%
-5.9%

$34.00

-7.1%
-9.3%

$37.61

-7.7%

$37.31
$36.71

$37.58

-1.7%

-6.4%

$32.00
$30.00
2Q09

3Q09

4Q09
ARPU

1Q10

2Q10

3Q10

4Q10E

6%
4%
2%
0%
-2%
-4%
-6%
-8%
-10%
-12%

1Q11E

% y-t-y change

Source: Company reports and J.P. Morgan estimates.

Competition remains intense; price war truce for now


After a roughly 18-month period of price competition, the prepaid pricing
environment appears to have stabilized for now. However, share gains across all
competitors could slow, with limited points of differentiation across the market,
particularly at the $40 price level. Instead, competitors are likely to pursue remaining
industry growth in the prepaid market and use promotions from time to time to win
incremental share. We believe Leap management prefers to see improved ARPU
trends and lower volatility (churn) rather than outsized subscriber gains.

Broadband to Fade to Background


We expect Leap to de-emphasize
the broadband business

Leap saw significant adoption of mobile broadband since the companys launch of
3G services and topped out at 663k customers, but from 2Q10 through 4Q10, the
company lost 115k customers in total. The company historically received only $35$40 per month, but new broadband plans launched in August have prices of $40, $50,
and $60 with varying usage caps. All three offer a 1.4 Mbps download speed, but
varying usage caps of 2.5 GB, 5 GB, and 7.5 GB, respectively. After passing the cap,
customers are throttled back to 56 Kbps until their next billing cycle or forced to
take a higher plan.
We believe Leap hopes to both reduce network usage and increase profits by forcing
its highest-usage customers off the service to free up capacity for smartphone
customers instead. Management has said that its network can support a maximum of
900k broadband customers at current usage levels, or 2.3 million smartphones at
expected levels. The broadband base stood at approximately 600k as of 3Q10, but we
expect that to fall and would not be surprised to see another price increase. We model
the broadband base remaining flat going forward, as new growth offsets expected
disconnects.
Table 71: Old vs. New Broadband Pricing
Broadband Pricing
Usage Cap
Speed
Source: Company reports.

166

Old
$40
5GB
1.4 Mbps

$40
2.5GB
1.4 Mbps

New
$50
5GB
1.4 Mbps

$60
7.5GB
1.4 Mbps

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 78: Broadband Subscribers and % of Total Base


Subscribers in thousands
11%

600
525

10%

9%

12%
9%

450
375

10%
8%

300
6%

225
150

3%

4%

75

2%

0
2008

2009

2010E

Broadband subscribers

2011E

2012E

% of total

Source: Company reports and J.P. Morgan estimates.

Smartphone Strategy Working Quickly


Leap saw great traction with the
Huawei Ascend in December

Leap introduced several smartphones, including Blackberry and Android devices, in


the third quarter and early fourth quarter. The company started selling the Blackberry
Curve 8530 (originally at $299.99, now $249.99) and Android-based Sanyo ZIO
(originally $249.99, now $199.99) in August, and also the much more inexpensive
Android-based Huawei Ascend ($149.99) in October. Prices are even cheaper with
web-only discounts. Notably, Android devices require the $55 rate plan and
Blackberry the $60 plan, which should help drive ARPU upside.
Take rates on smartphones give us confidence. Leap has stated that 35% of phone
sales (including sales to existing customers) were smartphones through the second
half of October, following the launch of the Ascend, and this percentage was slightly
higher in November and December. These take rates are up from low-double-digit
levels when only the $250 and $300 devices were available. In addition to the
Ascend, Leap has been able offer affordable price points on other devices, with basic
QWERTY devices starting at only $80 ($30 on web) to the Blackberry Curve 8530 at
$230 ($179 web price).
Figure 79: Handset Mix as % of New Sales and Upgrades
Subscribers in thousands

Source: Company reports (used with permission).

167

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

We estimate that subsidies on Leaps highest-end handset are in the $100 range, with
the Huawei device wholesaling for $225-250 vs. a retail price of $150 ($130 web
special). As component prices for handsets like touch screens and NAND ease with
additional capacity in 2011, we expect these prices to come down even farther, and
Leap has said it expects to be able to buy some Android devices for <$150 in 2011.
Figure 80: LEAP Partial Handset Lineup

Samsung Stunt
$40/Free After Rebate

Samsung Messager
Touch
$160/$80

Huawei Ascend
(Android)
$160/$100

Sanyo ZIO by Kyocera


(Android) $250/$130

Blackberry Curve
8530
$300/$180

Source: Company website (used with permission).

MVNO Agreement with Sprint Offers Growth Opportunity


At its August analyst day Leap announced an MVNO agreement with Sprint that
effectively gives the company a nationwide network and the ability to market its
services outside of its footprint. Leap will pay a minimum of $300 million over a
five-year period, providing the right for its customers to use Sprints network for
voice and 3G data services. Leap aims to use the arrangement to grow its subscriber
base beyond covered markets, reduce churn for customers in fringe markets, and
improve the overall competitive positioning of its product.
We expect out-of-network customers to start to grow in the second half of 2011.

4G Will Require Additional Spectrum


Leap management has committed to rolling out 4G LTE services. Specifically, we
expect Leap to build some trial markets in 2011 and fully launch additional markets
in 2012. As a result, we dont expect a significant spike in capital expenditures in
2011. However, more importantly, we believe the companys 4G efforts will be
severely limited until management acquires additional spectrum. The company may
consider any spectrum Clearwire decides to sell, but we do not sense much interest in
bidding aggressively.

168

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 72: LEAP Spectrum Holdings


Pops in thousands
Market

Pops

PCS

AWS

700 MHz

Total

Chicago, IL

8,886

10

Philadelphia, PA

5,995

20

20

Houston, TX

4,975

10

10

20

Washington, DC

4,686

20

20

Phoenix, AZ

3,387

10

10

20

Seattle, WA

3,245

30

30

Minneapolis, MN

3,212

30

30

San Diego, CA

2,959

10

20

30

St Louis, MO

2,862

30

30

Denver, CO

2,629

10

10

20

Baltimore, MD

2,578

20

20

Pittsburgh, PA

2,446

10

10

20

Portland, OR

2,098

20

20

Kansas City, MO

2,056

10

20

30

Charlotte, NC

2,034

10

10

20

San Antonio, TX

1,861

10

10

20

Milwaukee, WI

1,851

20

20

Norfolk, VA

1,751

10

10

Nashville, TN

1,727

15

10

25

Salt Lake City, UT

1,593

15

10

25

10

Source: FCC and J.P. Morgan estimates.

Operating and Financial Outlook


Leap has suffered subscriber losses in recent quarters, however operations reversed
course in 4Q10 as recent pricing changes improved retention and drew in new
customers. After adding 249k subscribers in 2010, Leap is expected to generate 400k
net additions in 2011 and another 400k in 2012. Though significantly lower than
historical levels, ARPU preservation and growth is more critical at this stage for
Leap, in our view. We expect ARPU to return to growth in 2011 after several
quarters of declining trends. Our operating forecasts drive estimated service revenue
growth of 10% in 2011 and 8% in 2012, versus 9% in 2010.

Earnings and Cash Flow Forecast


As a result of recent subscriber woes and ARPU compression, Leaps margins have
expanded much less than expected. The company was also working through dilution
from recently launched markets and its relatively new broadband business. We
expect margin expansion to pick up, as the company gains scale and top-line growth
accelerates. Our margin assumptions of 23.8% and 26.0% in 2011 and 2012
represent expansion of 90 bps and 220 bps, respectively. In 2010, Leap generated
positive free cash flow for the first time, though in only two of the quarters. We
expect the company to report positive free cash for the full year 2011, though free
cash may turn negative in some quarters depending on the timing of capital
investments. Free cash generation should continue into 2012 and beyond, though we
will closely monitor the companys LTE build plans and resulting capital needs.

169

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 73: Leap Wireless Financial Forecasts and DCF


$ in millions except per share and per sub values, subs in thousands
Ending subscribers
% change, y-t-y
Penetration (of cov. pops)
Net additional subscribers
% change, y-t-y
Churn rate, monthly
Total revenue
% change, y-t-y
ARPU, monthly
% change, y-t-y
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin
Adjusted EBITDA (ex. non-cash comp.)
% change, y-t-y
EBITDA margin
Taxes (paid)/benefit
Implied tax rate
Capital Expenditures
EPS - Reported
Cash EPS - diluted

2006
2,260
35%
6.1%
623
431%
3.8%
1,137
24%
$41.27
5.0%
256
NM
26.4%
276
1%
28.4%
(9)
-66%
595
($0.07)
$0.28

2007
3,001
33%
6.2%
741
19%
3.9%
1,622
43%
$44.47
7.7%
363
42%
25.9%
392
42%
27.9%
(37)
-97%
505
($1.13)
($0.87)

2008
4,018
34%
5.8%
1,049
42%
3.8%
1,973
22%
$38.28
-13.9%
392
8%
22.8%
428
9%
24.8%
(39)
-41%
796
($1.97)
($1.65)

2009
4,954
23%
5.4%
1,110
6%
4.5%
2,383
21%
$32.66
-14.7%
443
13%
20.7%
486
14%
22.7%
(41)
-20%
807
($3.30)
($2.94)

Private Market Valuation Summary


NPV of EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value@ 14x FCF
PMV of operations
Plus (less): NPV of NOLs
Less: assumed spectrum spending needs for continued growth
Less: net long-term debt, end of period (1)
PMV of equity
Fully diluted shares outstanding
DCF Value per Share
Private to public discount
Price Target
Upside to price target

Source: Company reports and J.P. Morgan estimates.

170

2010E
5,523
11%
5.9%
249
-78%
4.7%
2,521
6%
$37.24
14.0%
497
12%
21.2%
533
10%
22.8%
(30)
-4%
418
($9.15)
($8.86)

2011E
5,923
7%
6.3%
400
61%
4.1%
2,847
13%
$37.44
0.5%
568
14%
22.1%
608
14%
23.6%
(44)
-25%
450
($2.87)
($2.55)

2012E
6,323
7%
6.6%
400
0%
4.0%
3,066
8%
$37.75
0.8%
676
19%
24.4%
716
18%
25.8%
(44)
-74%
555
($1.36)
($1.04)

2013E
6,723
6%
7.0%
400
0%
4.0%
3,256
6%
$37.64
-0.3%
727
8%
24.7%
767
7%
26.0%
(44)
-563%
530
($0.68)
($0.36)

2014E
7,043
5%
7.3%
320
-20%
4.0%
3,421
5%
$37.53
-0.3%
787
8%
25.4%
827
8%
26.7%
(44)
84%
465
$0.11
$0.43

Current Price Multiples @


2011E
4,989
(253)
(2,781)
1,955
1,945
3,900
146
(471)
(2,242)
1,333
75.9
$17.57
10%
$15.81
15%

$13.72
EV ($b)
3.14

Current EV/Cash EBITDA


Current PER
Price Target EV/Cash EBITDA

Spectrum needs estimate


Price per pop
MHz needed

2015E
7,363
5%
7.5%
320
0%
4.0%
3,570
4%
$37.41
-0.3%
824
5%
25.5%
864
5%
26.7%
(44)
49%
420
$0.61
$0.93

3.77

2016E
7,663
4%
7.8%
300
-6%
4.0%
3,711
4%
$37.30
-0.3%
860
4%
25.6%
900
4%
26.8%
(44)
35%
437
$1.08
$1.40

2017E
7,963
4%
8.0%
300
0%
4.0%
3,848
4%
$37.19
-0.3%
894
4%
25.6%
934
4%
26.8%
(44)
27%
453
$1.54
$1.86

2018E
8,263
4%
8.3%
300
0%
4.0%
3,985
4%
$37.08
-0.3%
928
4%
25.7%
968
4%
26.8%
(44)
22%
469
$2.00
$2.32

2019E
8,463
2%
8.4%
200
-33%
4.0%
4,090
3%
$36.97
-0.3%
973
5%
26.2%
1,013
5%
27.3%
(44)
18%
482
$2.59
$2.91

2020E
8,663
2%
8.5%
200
0%
4.0%
4,176
2%
$36.86
-0.3%
989
2%
26.1%
1,029
2%
27.2%
(44)
17%
492
$2.81
$3.13

2009
6.5x
NM

2010E
5.9x
NM

2011E
5.2x
NM

2012E
4.4x
NM

2013E
4.1x
NM

7.8x

7.1x

6.2x

5.3x

4.9x

$0.50
10

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
16
1.5%
2.0%
2.5%
3.0%
3.5%
8.0%
26
29
32
36
41
9.0%
18
21
23
26
29
10.0%
13
14
$15.81
18
20
11.0%
8
9
10
12
13
12.0%
4
5
6
7
8

CAGR
08-20E
7%

-13%

6%
0%
8%
8%

-4%
NM
NM

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

MetroPCS
Initiate with Overweight Rating and $16 YE11 Price Target
We initiate coverage of PCS with an Overweight rating and a year-end 2011 price
target of $16/share, implying 24% upside from current levels. MetroPCS has
executed well in the last 12 months despite broad secular concerns regarding
competition. The company has managed to improve its churn and ARPU profile,
while growth has slowed but remained fairly robust. From a profitability and cash
flow perspective, Metro is in an attractive position as pricing should remain stable,
supporting EBITDA growth. Free cash flow, however, is likely to ramp more
meaningfully as capital expenditures decline and move closer to maintenance levels.
Despite outperforming its closest peer, Leap Wireless, and the wireless group in
2010, shares of MetroPCS have additional upside in 2011, in our view.

Overweight
MetroPCS (PCS;PCS US)
Company Data
Price ($)
12.97
Date Of Price
12 Jan 11
14.40 - 5.52
52-week Range ($)
Mkt Cap ($ mn)
4,591.43
Fiscal Year End
Dec
354
Shares O/S (mn)
Price Target ($)
16.00
Price Target End Date 31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.12
0.11
0.21
0.12
0.55
0.45

0.06A
0.23A
0.22A
(0.17)A
0.34A
0.68A

0.16
0.25
0.23
0.22
0.86
0.96

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on GAAP results.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Good execution to continue despite competitive environment
MetroPCS has exhibited strong performance over the past few quarters, after seeing
macro and competitor-driven weakness in 2009. We expect the company to continue
performing relatively well, and expect good growth in 1Q. However, we believe
Street estimates for 2011 may be somewhat aggressive. We expect net additions to
slow as general prepaid growth experiences a secular slowdown, and we forecast net
adds of 800k adds for 2011, down from 1.5 million in 2010. Despite the slower
subscriber growth, ARPU and churn appear to be on good trajectories, which should
help overall profitability.
4G improves data availability but will be a small part of handset mix in 2011
Metro was the first carrier in the US to launch 4G LTE services, and to date has
rolled out the service to nine markets. In addition, Metro also launched the first LTE
smartphone, the Samsung Craft, late in the third quarter and we expect additional
LTE smartphones in 2011. Handsets will be expensive for some time until the
technology gains some scale as other carriers launch their own offerings. However,
having a 4G smartphone may help the carrier improve its marketing capabilities in an
increasingly data-centric environment.

171

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Significant cash cushion to use for spectrum purchases


We expect Metro to end the year with approximately $1 billion in cash. With no
significant market buildout plans on the horizon and LTE not far from completion,
Metro has excess capital at its disposal, by our estimates. A share repurchase at this
point is unlikely, in our view. We believe the company may seek to purchase certain
small regional operators and spectrum assets to support a future ramp in 4G data
traffic, particularly in its top markets. The source of potential spectrum, however,
remains a question mark. The company could purchase spectrum through Clearwires
auction process or possibly buy Terrestars spectrum as was detailed in recent press
reports (WSJ MetroPCS Mulls Deal, 12/17/10). However, we do not believe either
is very attractive to management. For now, the company should be fine in most
markets, but over time spectrum will become an increasingly greater issue. We
include an estimated NPV of approximately $500 million for assumed spectrum
purchases in our valuation.
Turning into an FCF story; model 8.5% yield in 2011 and 12.7% in 2012
MetroPCS is poised to generate significant cash flow over the next several years. The
company has completed most of its expansion efforts, with only smaller markets
remaining at this point. In addition, its LTE build will mostly be completed in 1H11.
With more stable operating performance, profitability should improve. This, in
conjunction with declining capital investments, should result in substantial free cash
flow.

Investment Risks
Pricing pressure
The prepaid and pay-in-advance industry has experienced a variety of pricing moves
over the past two years. We believe such moves were brought on by competition, as
players aggressively tried to maintain growth. MetroPCS, in particular, took action
by bundling taxes and fees into its rate plans. Should the company need to make
additional pricing changes, profitability and cash flow could be negatively affected.
Tight spectrum market could create additional costs
MetroPCS is likely to encounter capacity constraints in some markets, especially
Philadelphia where it has only 10 MHz of spectrum. As data traffic grows on the
companys 4G network and penetration rises on its legacy network, Metro is likely to
see bandwidth shrink further. The company needs to purchase additional spectrum,
yet we remain concerned that attractive spectrum is likely to be either unavailable or
too expensive. As a result, the company could find itself overpaying for spectrum
resources, investing more capital in its existing network to increase capacity, or
limiting customer usage levels.
LTE ecosystem takes time to scale
We believe Metro could eventually derive a significant competitive advantage from
having a 4G network. However, until LTE smartphones become more affordable for
Metros target demographic, the company likely will have difficulty driving high
rates of adoption. Our concern is that handset pricing may not fall quickly enough
before competitors launch their own LTE services, eliminating any lead Metro
currently has.

172

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Company Description
MetroPCS is one of two companies in our coverage that focus on unlimited wireless
voice and data services on a pay-in-advance basis for $35-60 per month. The
company has historically targeted the low-income, minority, and youth
demographics, though increasingly other customers are also signing on to the service.
While MetroPCS started as a local carrier, it today covers about 125 million pops in
markets including Las Vegas, Miami, Dallas, Detroit, Philadelphia, Los Angeles, and
New York, and offers free roaming to customers in most of the country through
roaming partnerships. MetroPCS today has 7.9 million voice and data customers who
pay an average of $39.70 per month. The company went public in 2007.

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $16/share for MetroPCS. We based our
target price primarily on a discounted cash flow analysis that assumes a 9% WACC
and 2.5% perpetual growth rate. In addition, MetroPCS currently trades at 5.9x our
2011 EBITDA forecast, compared to the wireless industry average of 5.5x. Metros
closest peer, Leap Wireless, also trades at 5.4x our 2011 estimate. In our view, both
companies offer attractive growth which we believe justifies higher multiples than
the industry average.

Potential Leader in Competitive Prepaid Market


Coverage improved in 4Q
MetroPCS has exhibited strong performance over the past few quarters, after seeing
some demand and competitor-driven weakness in 2009. Despite lacking a nationwide
footprint through the third quarter of 2010, Metros share of new customers is not far
behind those of Tracfones Straight Talk (which is offered through multiple national
carriers) and Sprints Boost Unlimited, by our estimates.
Figure 81: MetroPCS Quarterly Net Adds
In thousands
800
600

692
435
303

400

205

200

223

298

13

0
(54)

(200)
1Q09

2Q09

3Q09

4Q09

1Q10

2Q10

3Q10

4Q10

Source: Company reports and J.P. Morgan estimates.

In November 2010, Metro broadened its coverage network for customers to more
than 90% of the US, which should improve sales in what had been marginal coverage
areas.

173

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 82: Coverage Map, Including Roaming

Source: Company website (used with permission).

Distribution focused on dedicated stores rather than big-box retailers


Interestingly, Metro focuses on generating most of its growth from smaller, indirect
distribution points and avoiding national, big-box chains. Therefore, there is no headto-head competition with Straight Talk, which is distributed exclusively through
Walmart. As a result, Boost Unlimited may be Metros biggest competitor; however,
we note that Sprint appears to be taking emphasis off that product line and focusing
on other lower-priced offerings.
Coming into the sweet spot of prepaid growth
We expect Metro to continue performing relatively well, and see good growth in 1Q,
after preannouncing net adds of 298k for 4Q, followed by seasonal weakness in 2Q
and 3Q. While 4Q was weaker than many expected, we are not surprised given
reports of high-end handset shortages in December. While handset inventory still
sounds tight, we expect Metro to have fixed that problem by the time the business
really ramps up in late January.
In total, our net add estimate of 800k in 2011 is somewhat more conservative than
consensus of over 1 million net adds, as we expect some level of incremental
competition for data plans and devices in Metros demographic. In an extremely
competitive prepaid market, competition for Metros highest-ARPU customers is
coming from low-end postpaid offerings and competition for the companys low-end
customers is coming from tradition pay-by-the-minute or pay-by-the-day offerings.

174

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

4G Offers Limited Differentiation in 2011


4G handsets will remain limited
and expensive in 2011

Metro has lagged the industry on the technology front, opting to skip launching 3G
services, while the companys closest peer Leap Wireless aggressively rolled out
EVDO and EVDO Rev A. However, Metro was the first carrier to launch 4G LTE
services. Although we do not believe the companys iteration of the service is as
robust as Verizons, the service offers a new point of differentiation that no other
prepaid operator can match for the time being. Metro has launched LTE in nine
markets, including Boston, New York, Sacramento, Dallas-Fort Worth, Detroit, Las
Vegas, Los Angeles, Philadelphia, and San Francisco, and plans to expand coverage
to its entire footprint by the end of 2011.
Metro launched the first LTE smartphone, the Samsung Craft, late in the third
quarter; the device is priced at $299 (likely $400-450 wholesale) and requires a $55
or $60 price plan. However, it remains early stages on the device front for LTE.
Handsets should remain expensive for some time until the technology gains some
scale, as other carriers launch their own offerings. We expect Metro to offer
additional LTE smartphones in 2011. These are likely to remain at a $100+ premium
to similar 3G devices well into 2012, so for the time being those that can offer 3G
smartphones are likely to have the advantage.
In addition, while Metro can offer nationwide voice, text, and data roaming on
partners 1XRTT 2G networks, no one else has an LTE network like Metros, so
nationwide LTE coverage is unlikely for some time.
Figure 83: MetroPCS Partial Handset Lineup

Huawei M228 $10


After Instant Savings

LG Select

$59

LG Baanter Touch
$149

Samsung Craft $299


After Instant Savings

Source: Company reports (used with permission).

Significant Cash Pile, but Limited Spend Ahead


We expect Metro to end 2010 with approximately $1 billion in cash. With no
significant market buildout plans on the horizon and LTE capital needs ramping
down in 2011, Metro has excess capital at its disposal, by our estimates. A share
repurchase at this point is unlikely, in our view. We think it is more likely that the
company continues to stockpile funds for potential tuck-in transactions and potential
spectrum acquisitions.
Pursuing spectrum where it is available
We believe MetroPCS will need to add spectrum depth to support a future ramp in
4G data traffic, especially in some markets like Philadelphia where it has only
10 MHz. In July and August, Metro purchased or exchanged for an additional
10 MHz of spectrum in the Orlando and Tampa markets, where it had also been at
only 10 MHz. In exchange, it had to give up spectrum in non-core markets Salt Lake
City and Portland. Metro also purchased 10 MHz of AWS spectrum in the Northeast
metropolitan area in the fourth quarter. Specific markets in which the company may
need additional spectrum include areas, such as Boston and New York, in which
spectrum has been very expensive in the past.
175

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

The source of potential spectrum, however, remains a question mark. The company
could consider purchasing 2.5 GHz spectrum through Clearwires auction process;
however, the LTE ecosystem is at an early stage and it will take some time before
compatible handsets are available. Another option could be to consider purchasing
Terrestars S-band spectrum; however, that asset may be highly encumbered and the
ecosystem for satellite-compatible LTE devices is at an even earlier stage in
development. The FCCs planned Auction 92 in 2011 does not provide for spectrum
in top metropolitan areas, and major progress on pulling additional broadcast
spectrum is likely to take several years. Therefore, we believe Metro could face
significant hurdles in finding additional capacity. For now, the company should be
fine in most markets, but as smartphone penetration ramps in 2011 and into 2012,
spectrum should become an increasingly greater issue.

Operating and Financial Outlook


We forecast Metro to maintain healthy subscriber growth. We currently estimate
subscriber growth of 10% in 2011 and 8% in 2012. ARPU stability, in our view, is
more critical to the companys success. As customers continue to migrate to higher
price points introduced in early 2010 and adoption of smartphone and Blackberry
plans increases, we expect ARPU to be stable in 2011. We forecast a modest decline
in ARPU in out years, though we believe this could be conservative as the subscriber
base is likely to transition to smartphones well into 2012 and 2013. Our operating
metric forecasts drive estimated top-line growth of 13% in 2011 and 8% in 2012.

Earnings and Cash Flow Forecast


As Metro re-priced its subscriber base and altered the way it subsidized customers in
2010, margin expansion was somewhat limited. However, we expect pricing to
remain stable for 2011 and LTE operating expenditures to begin tapering off, paving
the way for accelerated margin expansion in 2011. We forecast almost 200 bps of
improvement in 2011 and roughly 100 bps of additional growth in 2012. Improved
EBITDA trends, as well as a decline in capital expenditures, should drive significant
free cash flow generation, by our estimates. We forecast negative cash flow in 2010,
despite positive contributions in 1Q and 2Q, yet we expect levered free cash to jump
to $394 million in 2011 (an 8.5% yield on the current share price) and $585 million
in 2012 (12.7% yield).

176

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 74: MetroPCS Financial Forecasts and DCF


$ millions except per share and per sub values, subs in thousands
CAGR
08-20E
8%

Ending subcribers
% change, y-t-y
Penetration (of cov. pops)
Net additional subscribers
% change, y-t-y
Churn rate, monthly

2007
3,963
35%
5.7%
1,022
-16%
4.7%

2008
5,367
35%
8.4%
1,404
37%
4.7%

2009
6,640
24%
7.6%
1,272
-9%
5.4%

2010E
8,155
23%
8.0%
1,516
19%
3.6%

2011E
8,955
10%
8.4%
800
-47%
3.7%

2012E
9,685
8%
9.1%
730
-9%
3.7%

2013E
10,342
7%
9.7%
657
-10%
3.7%

2014E
10,933
6%
10.3%
591
-10%
3.7%

2015E
11,466
5%
10.8%
532
-10%
3.7%

2016E
11,945
4%
11.2%
479
-10%
3.7%

2017E
12,376
4%
11.6%
431
-10%
3.7%

2018E
12,764
3%
12.0%
388
-10%
3.7%

2019E
13,113
3%
12.3%
349
-10%
3.7%

2020E
13,427
2%
12.6%
314
-10%
3.7%

Revenue
ARPU, monthly
% change, y-t-y
Service revenue
% change, y-t-y
Total revenue
% change, y-t-y

$42.89
0.2%
1,918
49%
1,918
49%

$41.30
-3.7%
2,437
27%
2,437
27%

$40.02
-3.1%
3,130
28%
3,130
28%

$39.07
-2.4%
3,688
18%
3,688
18%

$39.41
0.9%
4,172
13%
4,172
13%

$39.21
-0.5%
4,503
8%
4,503
8%

$39.02
-0.5%
4,810
7%
4,810
7%

$38.82
-0.5%
5,081
6%
5,081
6%

$38.63
-0.5%
5,319
5%
5,319
5%

$38.44
-0.5%
5,529
4%
5,529
4%

$38.24
-0.5%
5,714
3%
5,714
3%

$38.05
-0.5%
5,875
3%
5,875
3%

$37.86
-0.5%
6,016
2%
6,016
2%

$37.67
-0.5%
6,138
2%
6,138
2%

-1%

Adjusted EBITDA (operating cash flow)


% change, y-t-y
Cash EBITDA margin
Depreciation
Tax expense

667
69%
34.8%
178
(123)

783
17%
32.1%
255
(117)

956
22%
30.5%
378
(87)

1,177
23%
31.9%
450
(162)

1,349
15%
32.3%
525
(197)

1,515
12%
33.6%
547
(252)

1,627
7%
33.8%
547
(297)

1,725
6%
34.0%
547
(336)

1,810
5%
34.0%
547
(369)

1,883
4%
34.1%
547
(398)

1,945
3%
34.0%
547
(423)

1,999
3%
34.0%
547
(444)

2,044
2%
34.0%
547
(462)

2,081
2%
33.9%
547
(477)

8%

Capital Expenditures
Capital expenditures/ service revenue
% change, y-t-y

768
40%
44%

1,201
49%
56%

832
27%
-31%

810
22%
-3%

698
17%
-14%

676
15%
-3%

721
15%
7%

711
14%
-1%

745
14%
5%

774
14%
4%

800
14%
3%

822
14%
3%

842
14%
2%

859
14%
2%

$0.46

$0.55

$0.86

$1.10

$1.29

$1.46

$1.61

$1.73

$1.84

$1.93

$2.01

$2.08

2009
7.9x
23.1x

2010E
6.4x
38.3x

2011E
5.6x
14.9x

2012E
5.0x
11.7x

9.0x
28.4x

7.3x
47.0x

6.4x
18.3x

5.7x
14.3x

Earnings per share (diluted)

$0.28

$0.34

Private Market Valuation Summary

Price Multiples @

NPV of Cash EBITDA 12E -- 20E


NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value@ 6x EBITDA
PMV of operations

2009
9,229
(1,634)
(4,388)
3,208
7,105
10,313

2010E
7,947
(1,457)
(3,505)
2,984
7,105
10,089

2011E
10,835
(2,207)
(4,552)
4,076
5,548
9,624

Less: NPV of potential spectrum spending


Plus: NPV of NOL
Less: net debt, end of period (1)
PMV of equity

497
(2,716)
8,093

330
(3,289)
7,130

(486)
133
(2,895)
6,375

Current shares outstanding


Shares to be issued from warrants/options (2)
Fully diluted shares outstanding

349
7
356.2

353
7
361

354
7
361

DCF Value per Share


Discount
Price Target per share
Upside to price target

$22.72
10%
$20.45

$19.77
10%
$17.80

2010E
1,177
(12)
($0.03)

2011E
1,349
394
$1.09

-0.3%
-624.9x
-0.2%

8.4%
19.2x
6.9%

$12.95
EV ($b)
7.57

Current EV/Cash EBITDA


Current PER
Price Target EV/Cash EBITDA
Current PER
Spectrum acquisition
Price per pop
MHz needed

8.63

-12%

8%
8%

7%
NM
-3%

13%

$0.50
10

MetroPCS Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
16
7.0%
8.0%
9.0%
10.0%
11.0%

$17.66
10%
$15.89
23%

1.5%
22
17
14
11
9

2.0%
24
19
15
12
10

2.5%
27
20
$15.89
13
10

3.0%
30
22
17
14
11

3.5%
34
25
19
15
12

Valuation Based on Forward FCF Multiple


EBITDA
Levered Free Cash Flow
FCF / diluted share

2009
956
88
$0.24

Current FCF Yield


EV/FCF
Target FCF Yield

2012E
1,515
585
$1.62
12.5%
12.9x
10.2%

2013E
1,627
653
$1.81
14.0%
11.6x
11.4%

2014E
1,725
427
$1.18
9.1%
17.7x
7.4%

* 2014 is first year at full cash tax rate.

Valuation Grid - year-end value per share


Free Cash Flow Multiple

10x
11x
12.0x
13x
15x

2011
11
12
$13.09
14
16

2012
16
18
$19.44
21
24

Valuation Based on Forward P/E Multiple


EBITDA
EPS - Diluted

2009
956
$0.55

2010E
1,177
$0.34

2011E
1,349
$0.86

2012E
1,515
$1.10

2013E
1,627
$1.29

2014E
1,725
$1.46

Current PER
Target PER

23.4x
28.7x

38.5x
47.2x

15.0x
18.4x

11.8x
14.5x

10.0x
12.3x

8.9x
10.9x

Valuation Grid - year-end value per share


P/E Multiple

10x
12x
15.0x
17x
20x

2011
9
10
$12.97
15
17

2012
11
13
$16.50
19
22

Source: Company reports and J.P. Morgan estimates.

177

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

178

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

NTELOS Holdings Corp.


Initiate with Neutral Rating and $22 YE11 Price Target
We initiate coverage of NTELOS with a Neutral rating and a year-end 2011 price
target of $22/share, implying 13% upside from current levels. Improved execution
combined with accretive acquisitions should drive moderately better financial trends.
In addition, management recently announced plans to separate its wireless and
wireline assets, both of which we believe are attractive strategic assets. We believe
the two businesses are likely to be worth more separately than together. Although
shares provide a nice yield, upside is likely to be limited near term, as operating
trends are unlikely to provide upside surprise, in our view. We believe investors will
find the potential for better upside with other telecom equities.

Neutral
NTelos Holdings Corporation (NTLS;NTLS US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

19.68
12 Jan 11
20.35 15.84
0.82
Dec
42
22.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.41
0.41
0.34
0.34
1.50
1.41

0.30A
0.27A
0.26A
0.22A
1.07A
1.13A

0.39
0.40
0.45
0.44
1.70
1.35

Source: Company dat a, Bloomberg, J.P. Morgan estimates.


Note: EPS figures based on G AAP results. 'Bloomberg' above denotes Bloomberg cons ensus estim ates .

Key Investment Points


Subscriber trends improving in the wireless business
NTELOS wireless business returned to positive postpaid subscriber growth in the
second quarter after losing subscribers for four straight quarters. The improvement is
the result of lower churn, while gross additions remain depressed. We believe the
improvement in churn is sustainable and gross additions could begin to grow in 4Q
given the companys more attractive handset lineup and competitive pricing plans.
On the other hand, we remain concerned about the lower-value prepaid business,
which has lost a meaningful number of subscribers in the past two quarters. In
addition, ARPU has declined significantly for several quarters. Though management
is addressing the issues, we do not expect meaningful improvement to take place
quickly. We forecast only 3.2% wireless growth in 2011.
Sprint wholesale volumes increasing; could drive upside to margins in 2011
Sprint and NTELOS have a contract in place that permits Sprint CDMA customers to
roam onto the NTELOS network. In exchange, Sprint pays NTELOS for the minutes
and bandwidth used. While the contract stipulates that NTELOS receive monthly
minimum revenue of $9 million, October was the first time in several quarters that
Sprints volumes were high enough that NTELOS received more than the minimum.
We attribute the increase to subscriber stabilization and increased data use at Sprint.
In line with our belief that Sprints subscriber trends will continue to improve, while
data traffic in general grows organically, we expect wholesale revenue from
NTELOS to continue growing. We expect wholesale to drive the bulk of wireless

179

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

growth in 2011, with margins on that revenue tracking significantly higher than
overall wireless margins.
Both wireline and wireless pieces are attractive strategic assets
In December, NTELOS management made the decision to spin its wireline assets
into a separate public entity with current shareholders receiving shares in the new
company a split that we expect to be completed in 2H11. We believe both pieces
are attractive strategic assets, the value of which may become more apparent after the
split occurs. Sprint has a wholesale relationship with NTELOS in the wireline
business, and both Frontier and Windstream operate in adjacent markets to those that
the companys wireline segment serves.
Return-of-capital story remains positive after split
We expect dividend strategies at the entities to diverge from what is in place today.
Given that pure-play wireless peers typically dont pay a dividend, we would expect
the wireless business to pay at best a nominal yield. NTELOS wireline business has
transformed from a mainly rural operator to a majority competitive carrier, with
heavy emphasis on fiber-based services for enterprise customers. While many
competitive carriers do not pay dividends, NTELOS has low leverage and excellent
cash flow so we expect its high payout ratio likely will continue. Windstream has
implemented a similar strategy, yet the company maintains a very attractive 6.1%
yield.

Investment Risks
Upside risks
NTELOS steals significant share
As NTELOS improves its execution, the company could be better positioned to take
share from competitors. To the extent that management can meaningfully
differentiate the companys offerings and draw customers from competitors, there
could be a significant ramp in gross additions, driving accelerated revenue growth.
Split takes place sooner
Should management receive necessary regulatory approvals sooner than expected,
NTELOS could split the two businesses earlier than the 2H11 target. The expected
value of separating the businesses would be achieved sooner, and could open the
door for a revaluation of these two strategic assets.
Downside risks
Sprint turnaround reverses course
We expect increased revenue from the wholesale agreement with Sprint to be a
growth driver going forward. Should Sprint hit major roadblocks in its turnaround,
revenue to NTELOS could fall back below the monthly minimum of $9 million. As a
result, wireless growth for NTELOS could be closer to flat, limiting interest in the
story relative to other opportunities in wireless.
Competition intensifies
We are forecasting modestly improved wireless operating trends beyond 2010. To
the extent that management is unable to sustain recent improvements in churn, our
forecasts could be too optimistic. Continued losses of postpaid customers likely
would drive negative wireless revenue growth.

180

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Integration of FiberNet hits obstacles


NTELOS recent acquisition, FiberNet, is expected to be a large contributor to
overall wireline revenue going forward. Management expects cost synergies of $69 million once FiberNet is fully integrated. In addition, the companys fiber-based
offerings are expected to drive most of the growth for the wireline business. As a
result, delays in integrating FiberNet or realizing synergy potential would negatively
affect revenue and profitability trends for the wireline business.

Company Description
NTELOS is a diversified regional telecommunications carrier that operates in
Virginia and West Virginia. Nearly three-fourths of the companys revenue is
derived from wireless sales, with the company serving nearly 434k postpaid and
prepaid customers in Virginia and West Virginia paying an average of $50.45 per
month. Its wireless network covers 5 million of the 8.7 million people under its
licenses. Within the wireline business, the company supports approximately 36k
access lines through its rural local exchange carrier (RLEC) business and an
additional 49k lines through its CLEC subsidiary, which predominantly targets
businesses. NTELOS came public in 2006 and private equity company Quadrangle
still owns 27% of the company. NTELOS in December 2010 announced a plan to
split its wireless and wireline divisions and expects to close that deal in the second
half of 2011.

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $22/share for NTELOS. We based our
target price primarily on a discounted cash flow analysis that assumes an 8.5%
WACC and 0.5% perpetual growth rate. In addition, NTELOS currently trades at
5.9x our 2011 EBITDA forecast, compared to an average of 5.2x for other integrated
carriers. Finally, we use a sum-of-the-parts analysis to support our DCF analysis.
Assuming a multiple of 6.5x 2011E EBITDA for wireless and 5.0x for wireline, we
arrive at a similar valuation.

Wireless/Wireline Split Has Finally Come


NTELOS management and investors have long discussed the possibility of splitting
the companys wireless and wireline businesses. As public entities, the two sides
offer different cash flow and growth profiles and therefore are likely more suitable
for different types of investors. In addition, the telecom space continues to
consolidate, and the assets could prove attractive strategically, given scarcity value.
In December, NTELOS management made the decision to spin its wireline assets
into a separate public entity, with current shareholders receiving shares in the new
company. The transaction is expected to take place in the second half of 2011,
following regulatory approvals. For two years following the agreement, the wireline
business will have access to certain key corporate management members, including
the CEO and CFO.

181

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Implications for the operating performance of the businesses


We do not expect any major impact to the way the two businesses operate after the
split occurs. Financially, both pieces could lose some benefit compared to being an
integrated entity. For example, there are certain systems, support functions, and other
administrative aspects of the business shared by both wireless and wireline today. In
addition, NTELOS has integrated selling of the services in some respects.
However, we believe that, from a revenue and cost perspective, the financial impact
will be minimal. The two businesses operate fairly independently, and the transaction
should give both an opportunity to expand upon that independence. Specifically,
management teams of both will have the autonomy to make capital allocation
decisions without concern about how that fits into a consolidated wireless/wireline
budget.
Both wireline and wireless pieces are attractive strategic assets
We believe both pieces are attractive strategic assets, the value of which may become
more apparent after the split occurs. Sprint has a wholesale relationship with
NTELOS in the wireline business, and both Frontier and Windstream operate in
adjacent markets to those that the companys wireline segment serves.
Return-of-capital implications
NTELOS currently supports an annual dividend of $50 million annual ($1.12/share),
representing a 5.7% yield and 62% payout ratio. In addition, the company currently
has a $40 million share repurchase program which has bee in place since August
2009; $17 million of the plan was completed in 2009 but the company has not
repurchased any shares since then. However, as separate companies, the growth
profiles and peer groups of the wireless and wireline businesses could promote
different return-of-capital strategies. Management is targeting leverage of 3.25x or
lower on an LTM basis for the wireless business, as per covenant requirements, and
leverage of 3.00x for the wireline business.
For the purposes of our analysis, we use 2011E EBITDA given expected timing for
the transaction. Assuming 3.00x net leverage for the wireline business, we estimate
the segment could support $290 million of debt and the resulting wireless leverage
would be 2.6x a very reasonable level, in our view.
Given that pure-play wireless peers typically dont pay a dividend, we would expect
the wireless business to pay a nominal yield, but for the wireline business to pay out
a large percentage of cash flow, with the remainder left to use on small acquisitions.
Dividend policies of wireline peers vary significantly, with payout ratios typically in
the 60-80% range and yields in the range of 6-8% for Windstream, CenturyLink, and
Frontier, while competitive carriers often do not pay any dividends. NTELOS
wireline business has transformed from a mainly rural operator to a majority
competitive carrier, with heavy emphasis on fiber-based services for enterprise
customers. The wireline business is expected to continue investing aggressively. That
being said, Windstream has implemented a similar strategy, yet the company
maintains a very attractive 7.4% yield. We would expect NTELOS management to
follow a similar strategy, given the cash flow potential for the business.

182

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Price target of $22 still implies attractive yields post-split; limited share upside
In the table below, we outline how we see the company paying out cash after the
split. We assume a current value for the wireless business of 5.5x EBITDA a
reasonable level given current wireless comps resulting in an implied valuation of
6.9x EBITDA for the wireline business. Our target price is commensurate with the
wireless valuation expanding modestly to 6.0x, and the wireline valuation remaining
fairly constant at 7.0x. Assuming a 3x net debt-to-EBITDA ratio for the wireline
business implies an 8.3% dividend yield at a 78% payout ratio.
Table 75: Post-Split Leverage and Return of Capital at Current and Target Prices
In millions
Current Price
Stock price
Equity value
EBITDA
Net debt
Net debt/EBITDA
EV
EV/EBITDA

2011 Target Price


Consolidated
$19.51
822
251

Wireless
$10.81
455
158

Wireline
$8.70
366
93

693
2.8x

413
2.6x

280
3.0x

Net debt
Net debt/EBITDA

1,514
6.0x

869
5.5x

646
6.9x

EV
EV/EBITDA

Stock price
Equity value
EBITDA

Consolidated
$21.54
907
251

Wireless
$12.69
534
158

Wireline
$8.85
373
93

693
2.8x

413
2.6x

280
3.0x

1,600
6.4x

948
6.0x

652
7.0x

Dividend ($m)
Dividend yield

50
6.1%

19
4.2%

31
8.5%

Dividend ($m)
Dividend yield

50
5.5%

19
3.6%

31
8.3%

Free cash flow


Payout ratio

75
66.4%

36
53.5%

40
77.9%

Free cash flow


Payout ratio

75
66.3%

36
53.5%

40
77.7%

Source: J.P. Morgan estimates.

Operating and Financial Outlook


We forecast NTELOS to drive improved subscriber growth. We currently estimate
net additions of 9.5k in 2011 and 11.5k in 2012, driven by better postpaid trends. In
addition, ARPU trends should also improve. We forecast flat postpaid ARPU,
compared to declines in recent quarters. Prepaid ARPU declines have been
significant over the past several quarters, and we expect a sharp deceleration as
management places emphasis on higher-ARPU plans. Our operating metric forecasts
drive estimated wireless top-line growth of 3.2% and 3.5% in 2011 and 2012,
respectively. On the wireline side, we expect the competitive carrier side of the
business to drive most of the growth. We forecast low- to mid-single digit growth,
although our estimates could prove conservative should NTELOS aggressively win
new business.

Earnings and Cash Flow Forecast


NTELOS management has demonstrated an ability to preserve and expand margins
despite operating headwinds. As wireless top-line growth returns, we expect service
margins to move back to the low-40% range over the next couple of years. We
expect wireline margins to move from 56% in 2010 to 50% in 2011 and 51% in
2012, driven by the acquisition of the lower-margin FiberNet business. However,
wireless margin expansion should help offset this dilution, driving modest
consolidated margin growth. In addition, NTELOS has historically generated
attractive free cash flow, supporting its dividend, and we would expect double-digit
growth to continue over the next couple of years. We do not expect any meaningful
changes to the companys capital investment program, with the only major new
initiative being spend related to the integration of FiberNet.

183

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 76: NTELOS Financial Forecasts and DCF


$ millions except per share and per sub values, subs in thousands
2007

367
9%
4.2%
31
-10%
0.6%
3.2%
$53
1%
322
15%
112
37.2%
25%

407
11%
4.7%
40
28%
0.8%
2.8%
$56
5%
378
17%
142
40.0%
27%

435
7%
5.0%
28
-29%
0.6%
2.9%
$55
-1%
416
10%
163
41.8%
15%

439
1%
5.0%
4
-88%
0.1%
3.3%
$53
-3%
425
2%
162
40.6%
-1%

440
0%
5.1%
1
-69%
0.0%
3.1%
$50
-6%
405
-5%
147
38.8%
-9%

449
2%
5.2%
10
780%
0.2%
3.0%
$50
0%
417
3%
158
40.3%
7%

461
3%
5.3%
12
21%
0.2%
2.9%
$50
0%
430
3%
170
41.8%
7%

471
2%
5.4%
10
-13%
0.2%
2.8%
$50
0%
438
2%
171
41.3%
1%

479
2%
5.5%
8
-20%
0.2%
2.6%
$50
0%
445
2%
175
41.5%
2%

485
1%
5.6%
6
-25%
0.1%
2.5%
$50
0%
451
1%
179
41.6%
2%

489
1%
5.6%
4
-33%
0.1%
2.4%
$50
0%
455
1%
181
41.8%
2%

493
1%
5.7%
4
0%
0.1%
2.3%
$50
0%
458
1%
184
41.9%
1%

497
1%
5.7%
4
0%
0.1%
2.2%
$50
0%
462
1%
186
42.0%
1%

501
1%
5.8%
4
0%
0.1%
2.0%
$50
0%
464
1%
187
42.0%
1%

505
1%
5.8%
4
0%
0.1%
1.9%
$50
0%
467
1%
189
42.1%
1%

Wireline revenue
ILEC access lines
CLEC access lines
EBITDA
% margin
% change, y-t-y

117
45
47
65
56%
5%

122
44
49
65
54%
0%

123
41
50
69
56%
6%

125
38
50
72
57%
4%

140
45
140
78
56%
9%

219
43
142
106
48%
35%

226
41
143
116
51%
10%

227
39
144
118
52%
1%

229
37
145
119
52%
1%

229
35
146
121
53%
1%

230
33
148
121
52%
0%

230
31
149
120
52%
0%

230
29
150
119
52%
-1%

229
27
151
118
51%
-1%

229
25
152
117
51%
-1%

5%
-6%
1%
4%

Consolidated revenue
EBITDA
% change, y-t-y
EBITDA margin

440
159
8.5%
36.2%

500
198
24.4%
39.6%

540
223
12.7%
41.4%

550
222
-0.6%
40.4%

546
213
-4.2%
39.0%

636
251
18.0%
39.5%

656
273
8.8%
41.6%

665
273
0.0%
41.0%

675
279
2.2%
41.4%

681
284
1.8%
41.7%

685
286
0.8%
41.8%

689
288
0.6%
41.8%

692
289
0.3%
41.8%

694
290
0.2%
41.7%

696
290
0.1%
41.7%

2%
3%

(6)
87
19.7%
($0.18)
($0.66)
$0.00

(16)
110
21.9%
$0.77
$0.87
$0.00

(24)
132
24.5%
$1.12
$1.20
$0.70

(29)
99
18.1%
$1.50
$1.62
$1.06

(24)
92
16.9%
$1.07
$1.19
$1.12

(29)
99
15.6%
$1.70
$1.79
$1.20

(43)
97
14.8%
$2.06
$2.15
$1.24

(43)
99
14.9%
$2.06
$2.13
$1.27

(45)
95
14.0%
$2.16
$2.22
$1.31

(46)
96
14.1%
$2.24
$2.30
$1.35

(47)
92
13.4%
$2.27
$2.34
$1.39

(48)
92
13.4%
$2.30
$2.36
$1.43

(48)
88
12.7%
$2.31
$2.38
$1.48

(48)
88
12.7%
$2.32
$2.39
$1.52

(48)
88
12.7%
$2.33
$2.39
$1.57

Cash taxes paid


Capital Expenditures
CAPEX/revenue
Earnings per diluted share
Cash EPS
Dividend per share

2008

Private Market Valuation Summary


NPV of EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value @ 6x EBITDA
PMV of operations

2011E
1,764
(286)
(584)
894
826
1,721

Less: net long-term debt, end of period (1)


PMV of equity

(693)
1,028

Shares outstanding
Shares to be issued from warrants/options (2)
Fully diluted shares outstanding

2010E

2011E

Multiples @
Current EV/EBITDA
Current P/E

2012E

2013E

$19.51

Target EV/EBITDA
Target P/E

2014E

2015E

2016E

Current
2010
7.1x
5.5x
7.6x

$24.41
10%
$21.97
1.12
18%

2011
6.0x
5.5x
6.1x

2010E
7.1x
16.4x

2011E
6.0x
10.9x

2012E
5.5x
9.1x

7.3x
13.5x

7.6x
18.5x

6.4x
12.2x

5.9x
10.2x

Consolidated multiple
Estimated wireline multiple
Implied wireless multiple

Price Target
2010
2011
7.6x
6.4x
6.0x
6.0x
8.1x
6.4x

Valuation Based on Forward P/E Multiple


Cash EBITDA
Cash EPS - Diluted

2010E
213
$1.19

2011E
251
$1.79

2012E
273
$2.15

Tax-affected NOL's remaining per share


EOY 2009E cash levels per share
Current stock value ex. cash and NOL's
Adjusted price / diluted EPS

0.44
$0.75
$18.32
15.4x

0.22
$1.35
$17.94
10.0x

0.17
$0.00
$19.34
9.0x

2010E
31
58
$1.39

2011E
57
75
$1.81

2012E
91
86
$2.06

Valuation Grid - year-end value per share


P/E Multiple on 2011E EPS
Normalized
Diluted
+NOL's
12x
22
10x
18
14x
25
12x
22
15.0x $26.90
15.0x
$27.12
16x
29
16x
29
18x
32
18x
33

Valuation Based on Forward FCF Multiple


Cash EBITDA
Free Cash Flow
FCF / diluted share

Present value of tax-affected NOL's remaining per s $0.44


EOY 2009E cash levels per share
$0.75
Current stock value ex cash and NOL's
$18.32
Adjusted price / FCF
13.2x
FCF Yield
7.1%

$0.22
$1.35
$17.94
9.9x
9.3%

Source: Company reports and J.P. Morgan estimates.

184

$0.17
$0.00
$19.34
9.4x
10.6%

Valuation Grid - year-end value per share


Free Cash Flow Multiple on 2011 FCF

21x
22x
23x
24x
25x
26x

2011E
38
40
$41.55
43
45
47

2017E

2009
6.8x
12.0x

Valuation Grid - DCF Value per Share


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
0.0%
0.5%
1.0%
1.5%
22 -0.5%
6.5%
27
29
31
33
36
7.5%
22
23
24
26
28
8.0%
20
21
$21.97
23
25
9.0%
16
17
18
19
20
10.0%
13
14
15
15
16

41.7
0.4
42.1

DCF Value per Share


Private-public discount of 10%
Price target
Plus annual dividend
Upside to target

2009

21x
22x
23x
24x
25x
26x

Normalized
+NOL's
38
40
$41.77
44
45
47

2018E

2019E

2020E

CAGR
10E-20E

2006
Wireless
Ending subscribers
% change, y-t-y
Penetration (of cov. pops)
Net additional subscribers
% change, y-t-y
Penetration gain (of cov. pops)
Churn rate, monthly
LocaL ARPU, monthly
% change, y-t-y
Total revenue
% change, y-t-y
EBITDA
% margin
% change, y-t-y

1%

14%

0%
1%
3%

0%
7%

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

US Cellular Corp
Initiate with Underweight Rating and $45 YE11 Price Target
We initiate coverage of US Cellular with an Underweight rating and a year-end 2011
price target of $45/share, implying 9% downside from current levels. We expect the
business to continue to struggle to grow; while recent moves to improve subscriber
and margin performance should help to some degree, we do not expect a meaningful
change. We believe US Cellulars positioning in the industry provides the company
with limited protection against price competition, and we think many potential
customers could opt for a pay-in-advance operator or postpaid competitor. In
addition, shares trade at a premium valuation, free cash flow is weak, there is no
dividend, and the Carlson family controls US Cellular, all of which we believe justify
our Underweight rating.

Underweight
US Cellular (USM;USM US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ mn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

50.30
12 Jan 11
50.85 33.84
4,347.33
Dec
86
45.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
P/E FY

2009A

2010E

2011E

0.97
0.96
0.41
0.14
2.48
20.3

0.55A
0.47A
0.43A
0.23A
1.69A
29.8A

0.51
0.36
0.47
0.59
1.93
26.1

Source: Company dat a, Bloomberg, J.P. Morgan estimates.

Key Investment Points


Belief Project benefits likely to be modest
US Cellulars postpaid subscriber growth has been negative through the first three
quarters of 2010. At the beginning of the fourth quarter, management launched the
Belief Project to address these trends. The new campaign introduces new price plans
and handsets, as well as features to alleviate customer complaints. Though the
offerings seem to have improved retention efforts in 4Q, we are cautious on the
ability for US Cellular to gain much traction in gross additions. Given its positioning
in the market, the company faces intense competition from high-end national carriers
as well as low-end prepaid players, some of which have superior brands.
Low smartphone penetration creates opportunity for existing base
Only 20% of US Cellulars postpaid customers have smartphones or other devices
that require data plans, well behind similar metrics at AT&T and Verizon. The
companys smaller scale, and thus late adoption of the devices, is the primary reason
behind the lower penetration. Recent take rates, driven by Android device adoption,
are encouraging; the company currently has five Android devices for sale. As
additional devices are rolled out in 2011, take rates could grow, which would also
support ARPU as customers purchase data plans. However, we believe this will
primarily be driven by existing customer upgrades, rather than new customers. In
addition, a Verizon iPhone could have a large impact on US Cellular, in particular.
As a result, we forecast only modest ARPU growth in 2011.

185

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless margin improvement will be difficult


Wireless EBITDA margins fell 450 bps from 2007 to 2009, and we expect another
220 bps of incremental decline in 2010. Although there could be some stabilization
in subscriber trends and margins in 2011, we believe handset subsidies will continue
to weigh on margins. As management aims to push smartphones into customers
hands, it will need to subsidize these more than lower-end devices. In addition, as
part of its loyalty program, the company is likely to make new handsets available
earlier over the life of a contract than it has previously. As a result, we forecast
limited margin improvement in 2011.

US Cellular is controlled by the


Carlson family

Public shareholders have limited influence on direction


US Cellular is 83% owned by Telephone & Data Systems, while TDS is controlled
through super-voting shares by the founding Carlson family. Although US Cellular
has struggled for relevance and shareholders have at times called for a sale of the
company, we see no evidence that the family wants to sell in the near term.
Premium valuation tough to justify
Though US Cellular could have some success in coming quarters as execution
improves with a new CEO and with continued emphasis on the Belief Project, we are
skeptical that a turnaround can come quickly. As a result, we see limited upside to
our forecasts and anticipate growth well below peer levels. With shares trading at
6.0x our 2011E EBITDA, we believe wireless investors will find more interesting
upside stories in Sprint Nextel, Leap Wireless, and MetroPCS, which trade at 4.6x,
5.4x, and 5.9x, respectively. If investors do want exposure to US Cellular, we believe
it would be better to invest in Telephone & Data Systems which trades at 4.0x total
EBITDA (and an implied 1.1x wireline EBITDA excluding its USM stake) despite
generating approximately 84% of its revenue from US Cellular.

Investment Risks
Return of wholesale bodes well for wireless growth and margins
In the third quarter, US Cellular reported growth in roaming revenue, after six
quarters of declines on a year-over-year basis. The prior declines were caused by
reduced business as a result of the Verizon/Alltel merger, but that impact has now
cycled through and growth could continue going forward as data traffic continues to
accelerate. Improved roaming results will be a critical component of revenue growth
and profitability for US Cellular in coming quarters, and results could exceed our
expectations.
Belief Project and ramping smartphone penetration drive subscriber revenue
US Cellulars new Belief Project marketing campaign introduces new handset
pricing and more consumer-friendly contract structures. The Belief Project seems to
have stemmed churn somewhat in 4Q10 and if it succeeds in driving gross add share
as well could increase wireless top-line revenue trends. In addition, smartphone
penetration at US Cellular is very low and could grow more quickly than we assume,
increasing ARPU and potentially margins.
US Cellular owns valuable 850 MHz wireless licenses
US Cellular is 83% owned by Telephone & Data Systems, while TDS is controlled
by the founding Carlson family. While the company has faced customer defections
and significant margin pressure over the past several quarters, US Cellular does own
valuable 850 MHz wireless licenses, which are strategically important and attractive.
186

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $45/share for US Cellular. We base our
target price primarily on a discounted cash flow analysis that assumes an 8.5%
WACC and 1.0% perpetual growth rate. In addition, US Cellular currently trades at
6.0x our 2011 EBITDA forecast, compared to an average of 5.5x for other wireless
carriers. Given our forecasts for limited growth, we believe investors will find better
opportunities elsewhere in the wireless space. We prefer Sprint Nextel, Leap
Wireless, and MetroPCS.

Company Description
US Cellular is a regional wireless carrier and the sixth-largest operator in the country.
The company operates a CDMA network, also enabled for EVDO 3G services,
covering 26 states. The company currently has 6.1 million customers, including
5.2 million postpaid and 500k prepaid customers, resulting in a 13.1% penetration
rate across its markets. These customers spend on average $47.13 a month.
US Cellular is an 83%-owned subsidiary of Telephone & Data Systems, which is
controlled by the Carlson family.
Table 77: USM and TDS Ownership
Name

Position with TDS, USM and Principal Occupation

LeRoy T. Carlson

Chairman and Director Emeritus of TDS


Director of USM
President and Chief Executive Officer of TDS
Director of TDS
Director and Chairman of the Board of USM

LeRoy T. (Ted) Carlson, Jr.

Age

Director Since
TDS
USM

93

NA

1987

63

1968

1984

Letitia G. Carlson, M.D.

Director of TDS.
Physician and Associate Clinical Professor at George
Washington University Medical Center

49

1996

NA

Prudence E. Carlson
Walter C.D. Carlson

Director of TDS and Private Investor


Director and non executive Chairman of the Board of TDS
Director of USM
Partner, Sidley Austin LLP, Chicago, Illinois

58
56

2008
1981

NA
1989

Source: Company reports and J.P. Morgan.

Table 78: USM and TDS Shares and Voting Rights


USM

Shares

Votes

Adj. Votes

Series A

33.0

10

330.1

86.1%

Common
Shares

53.4

53.4

13.9%

61.8%

86.4

383.5

100.0%

100.0%

Float

15.6

TDS ownership
Series A

70.8

Common

37.8
Shares

TDS

33.0

Voting %

Economic
Ownership %
38.2%

15.6

4.1%

18.1%

367.8

95.9%

81.9%

10
1

330.1

86.1%

38.2%

100.0%

37.8

9.9%

43.7%

70.7%

Votes

Adj. Votes

Voting %

% of
class

114.7

100.0%

Economic
Ownership %
100.0%

Common

49.8

49.8

43.4%

47.2%

Special

49.3

0.0%

46.7%

Series A

6.5

10

64.9

56.6%

6.1%

64.6

56.3%

13.8%

Shares

% of
class

105.6

29.3%

Carlson Voting

14.6

Common

0.7

0.7

0.6%

0.7%

1.5%

Special

7.5

0.0%

7.1%

15.2%

Series A

6.4

10

63.9

55.7%

6.1%

98.4%

Source: Company reports and J.P. Morgan.

187

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Operating and Financial Outlook


We forecast US Cellular to show modestly improved subscriber growth in 2011 and
2012, relative to subscriber losses in 2009 and 2010. We currently estimate net
additions of 25k in 2011 and 20k in 2012, driven by better retail execution through
implementation of the Belief Project. In addition, ARPU trends should also improve
as smartphone adoption accelerates, and we forecast ARPU growth of 1% in 2011
and 2% in 2012. Our operating metric forecasts drive estimated wireless top-line
growth of 1% in 2011 and 2% in 2012.

Earnings and Cash Flow Forecast


Limited subscriber and ARPU growth, offset somewhat by smartphone costs, drive
our wireless service margin forecasts of 21.5% in 2011 and 22.6% in 2012, up
slightly from 21.0% in 2010. Timing and magnitude of an LTE launch could cause
downside to our forecasts, however. We factored in a slow LTE rollout, and expect a
modest increase in wireless capital spend in 2011 and 2012. As a result, cash flow
should stay relatively flat in 2011, though EBITDA growth could drive a nice boost
in 2012.

188

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 79: US Cellular Financial Forecasts and DCF


$ millions except per share and per sub values, subs in thousands
Subscribers
% change, y-t-y
Penetration
Net additional subscribers
% change, y-t-y
Penetration gain
Churn rate, monthly

2006
5,815
6%
12.6%

2007
6,102
5%
13.0%

2008
6,196
2%
13.0%

2009
6,141
-1%
12.8%

2010E
6,123
0%
12.6%

2011E
6,148
0%
12.6%

2012E
6,168
0%
12.5%

2013E
6,186
0%
12.5%

2014E
6,203
0%
12.5%

2015E
6,219
0%
12.5%

2016E
6,233
0%
12.5%

2017E
6,246
0%
12.5%

2018E
6,258
0%
12.5%

2019E
6,268
0%
12.5%

2020E
6,277
0%
12.5%

310
-35%
0.7%

281
-9%
0.6%

94
-67%
0.2%

(55)
-159%
-0.1%

(18)
-67%
0.0%

25
-239%
0.1%

20
-20%
0.0%

18
-10%
0.0%

17
-6%
0.0%

16
-7%
0.0%

14
-8%
0.0%

13
-9%
0.0%

12
-10%
0.0%

10
-11%
0.0%

9
-11%
0.0%

CAGR 10-20E
0%
0%
SM

1.8%

2.1%

1.9%

2.1%

1.9%

1.9%

1.9%

1.8%

1.8%

1.8%

1.8%

1.8%

1.8%

1.8%

1.8%

3,473
14%
$41.39
4%

3,946
14%
$44.25
7%

4,243
8%
$46.51
5%

4,215
-1%
$46.95
1%

4,183
-1%
$47.12
0%

4,229
1%
$47.59
1%

4,300
2%
$48.07
1%

4,366
2%
$48.55
1%

4,392
1%
$48.55
0%

4,418
1%
$48.55
0%

4,444
1%
$48.55
0%

4,466
0%
$48.55
0%

4,487
0%
$48.55
0%

4,507
0%
$48.55
0%

4,527
0%
$48.55
0%

1%

865
24%
26.9%

1,033
19%
28.1%

1,015
-2%
25.8%

926
-9%
23.6%

825
-11%
21.0%

858
4%
21.5%

919
7%
22.6%

1,043
13%
25.2%

1,111
7%
26.7%

1,136
2%
27.1%

1,159
2%
27.5%

1,171
1%
27.6%

1,183
1%
27.8%

1,194
1%
27.9%

1,206
1%
28.1%

4%

EBIT
% change, y-t-y

290
56%

443
53%

28
-94%

326
1078%

229
-30%

261
14%

317
22%

416
31%

472
14%

485
3%

496
2%

495
0%

495
0%

494
0%

492
0%

8%

Depreciation & Amortization


Depreciation
% change, y-t-y
Amortization

539
12%
36

554
3%
36

541
-2%
36

535
-1%
36

561
5%
36

560
0%
37

563
0%
40

583
4%
45

594
2%
45

606
2%
45

618
2%
45

631
2%
45

643
2%
45

656
2%
45

669
2%
45

2%

Taxes Paid
% change, y-t-y

(120)
-36%

(217)
-80%

(8)
96%

(114)
-1317%

(101)
12%

(112)
-11%

(136)
-22%

(175)
-29%

(203)
-16%

(216)
-6%

(230)
-6%

(240)
-5%

(252)
-5%

(266)
-5%

(281)
-6%

11%

Capital Expenditures
Capital expenditures/service revenue

584
18.2%

566
15.4%

586
14.9%

547
13.9%

589
15.0%

620
15.6%

620
15.3%

620
15.0%

600
14.4%

600
14.3%

580
13.8%

560
13.2%

540
12.7%

520
12.2%

500
11.6%

-2%

EPS

$2.00

$3.02

$0.36

$2.48

$1.69

$1.93

$2.34

$3.02

$3.50

$3.73

$3.96

$4.15

$4.36

$4.59

$4.85

2010E
6.0x
29.4x

2011E
5.7x
25.7x

2012E
5.3x
21.1x

2013E
4.7x
16.4x

2014E
4.4x
14.1x

2015E
4.3x
13.3x

5.5x
26.6x

5.3x
23.3x

4.9x
19.1x

4.3x
14.9x

4.1x
12.8x

4.0x
12.0x

Revenue
% change, y-t-y
Local ARPU, monthly
% change, y-t-y
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin

Private Market Valuation Summary

Valuation Multiples @

NPV of cash EBITDA 12E -- 20E


NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value
PMV of consolidated operations

2011E
6,780
(1,305)
(3,546)
1,930
2,859
4,788

PMV -- minority stakes


Total PMV of Cellular Operations

695
5,483

Less: net long-term debt, end of period (1)


PMV -- Equity

(637)
4,846

Current shares outstanding


Shares to be issued/warrants/options (2)
Fully diluted shares outstanding

EBITDA
Free Cash Flow
FCF / diluted share

Current EV/EBITDA
Current PER

EBITDA
EPS - Normalized diluted
% change, y-t-y

2%

Valuation Grid - 2011E DCF Value per Share


Terminal Growth Rate (across) on 2020 EBITDA, Discount Rate (down)
45
6.5%
7.5%
8.5%
9.5%
10.5%

0.0%
56
47
41
36
33

0.5%
59
50
43
38
34

1.0%
63
52
$44.87
39
35

1.5%
68
56
47
41
36

2.0%
73
59
50
42
37

$56.09
20%
$44.87
2010E
825
56
$0.64

2011E
858
51
$0.59

2012E
919
92
$1.07

2013E
1,043
177
$2.04

Valuation Grid - year-end value per share


Free Cash Flow Multiple on FCF
12x
15x
20.0x
22x
25x

Valuation Based on Forward P/E Multiple

4,514

Price Target EV/EBITDA


Target PER

86.0
0.4
86.4

PMV per share


Private to public discount
12-month DCF value per share
Valuation Based on Forward FCF Multiple

$49.52
EV
4,915

0%

2010E
825
$1.69

Tax-affected NOL's remaining per share

2011E
858
$1.93

2012E
919
$2.34

2013E
1,043
$3.02

$0.00

$0.00

$0.00

2011
7
9
$11.75
13
15

2012
13
16
$21.31
23
27

Valuation Grid - year-end value per share


P/E Multiple on fully-taxed EPS
$0.00

12x
15x
20.0x
22x
25x

2011
23
29
$38.52
42
48

2012
28
35
$46.87
52
59

Source: Company reports and J.P. Morgan estimates.

189

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

190

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Telephone & Data Systems


Initiate with Neutral Rating and $42 YE11 Price Target
We initiate coverage of Telephone & Data Systems with a Neutral rating and a yearend 2011 price target of $42/share, implying 17% upside from current levels.
US Cellular represents TDSs wireless business and main financial driver, and we
expect the business to continue to struggle. New initiatives to improve subscriber and
margin performance should help to some degree, but we do not expect meaningful
change. In addition, we expect TDSs wireline business result to remain at current
levels, as broadband penetration offsets traditional voice pressures. Our Neutral
rating for TDS primarily reflects a more attractive valuation relative to USM,
although we believe there are better opportunities elsewhere in the telecom sector.

Neutral
Telephone & Data Systems (TDS;TDS US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ mn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

36.28
12 Jan 11
37.91 28.84
3,820.21
Dec
105
42.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
P/E FY

2009A

2010E

2011E

0.66
0.66
0.33
0.15
1.80
20.2

0.46A
0.38A
0.39A
0.20A
1.44A
25.2A

0.43
0.34
0.40
0.49
1.66
21.9

Source: Company dat a, Bloomberg, J.P. Morgan estimates.

Key Investment Points


Belief Project benefits likely to be modest
US Cellulars postpaid subscriber growth has been negative through the first three
quarters of 2010. At the beginning of the fourth quarter, management launched the
Belief Project to address these trends. The new campaign introduces new pricing
plans and handsets, as well as features, to alleviate customer complaints. Though the
offerings could improve retention efforts and drive share gains, to some extent, we
are cautious on the ability for US Cellular to gain much traction. Given its
positioning in the market, the company faces intense competition from high-end
national carriers as well as low-end prepaid players, some of which have superior
brands.
Low smartphone penetration creates opportunity for existing base
Only 20% of US Cellulars postpaid customers have smartphones or other devices
that require data plans, well behind similar metrics for AT&T and Verizon. The
companys smaller scale, and thus late adoption of the devices, is the primary reason
behind the lower penetration. Recent take rates, driven by Android device adoption,
are encouraging; the company currently has five Android devices for sale. As
additional devices are rolled out in 2011, take rates could grow, which would also
support ARPU as customers purchase data plans. However, we believe this will
primarily be driven by existing customer upgrades, rather than new customers. In
addition, the Verizon iPhone could have a large impact on US Cellular, in particular.
As a result, we forecast only modest ARPU growth in 2011.

191

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Wireless margin improvement will be difficult


Wireless EBITDA margins fell 450 bps from 2007 to 2009, and we expect another
220 bps of incremental decline in 2010. Though there could be some stabilization in
subscriber trends in 2011, we believe handset subsidies will continue to weigh on
margins. As the company aims to push smartphones into customers hands, it will
need to subsidize these more than lower-end devices. In addition, as part of its
loyalty program, the company is likely to make new handsets available earlier over
the life of a contract than it has previously. As a result, we forecast limited margin
improvement in 2011.
Wireline results to reflect access line drag offset by broadband
We expect the TDS wireline business to reflect little, if any, growth. Modest
broadband growth is offsetting mid-single-digit declines in access lines. The
broadband business has also been helped by customers taking higher-speed, higherARPU offerings, but with the business at 43% penetration we are somewhat
concerned that growth will taper off. In addition, the CLEC business, which is
typically a growth driver at other wireline operators, continues to report declining
revenue trends. This is largely because of the businesss embedded residential base
of subscribers which continue to disconnect. Going forward, we expect management
to invest further in its broadband network to roll out higher speeds, which could limit
free cash flow from the business in 2011.
Public shareholders have limited influence on direction
TDS is controlled through super-voting shares by the founding Carlson family.
Though US Cellular has struggled for relevance and shareholders have at times
called for a sale of the company, we see no evidence that the family wants to sell in
the near term.

Investment Risks
Upside risks
Return of wholesale bodes well for wireless growth and margins
In the third quarter, US Cellular reported growth in roaming revenue, after six
quarters of declines on a year-over-year basis. The prior declines were caused by
reduced business as a result of the Verizon/Alltel merger, but that impact has now
cycled through and growth could continue going forward as data traffic continues to
accelerate. Improved roaming results will be a critical component of revenue growth
and profitability for US Cellular in coming quarters, and results could exceed our
expectations.
Belief Project and ramping smartphone penetration drive subscriber revenue
US Cellulars new Belief Project marketing campaign introduces new handset
pricing and more consumer-friendly contract structures. The Belief Project seems to
have stemmed churn somewhat in 4Q10 and if it succeeds in driving gross add share
as well could increase wireless top-line revenue trends. In addition, smartphone
penetration at US Cellular is very low and could grow more quickly than we assume,
increasing ARPU and potentially margins.

192

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

US Cellular owns valuable 850 MHz wireless licenses


US Cellular is 83% owned by Telephone & Data Systems, while TDS is controlled
by the founding Carlson family. While the company has faced customer defections
and significant margin pressure over the past several quarters, US Cellular does own
valuable 850 MHz wireless licenses, which are strategically important and attractive.
Downside risks
Wireless competition intensifies
The US wireless industry is very competitive, with arguably a surplus of players in
any given market. As growth continues to decelerate, there is potential for
competitors to aggressively pursue share by lowering price. As a regional operator,
TDSs wireless business is disproportionately exposed to national carrier
competition. Should additional pricing pressures occur, we would expect downside
risk to our current wireless forecasts.
Free cash flow impacts return of capital
TDS currently pays roughly $48 million in dividends each year, and is also
repurchasing shares under a $250 million program that was established in November
2009 and expires in November 2012. Though current free cash generation is ample,
should the companys wireless and wireline businesses deteriorate significantly, we
believe management would need to alter its return-of-capital strategy. Potential
capital investment plans could also have an impact, as management intends to
expand availability of higher broadband speeds on the wireline side and an LTE
rollout in late 2011 or early 2012 is also likely.

Valuation and Price Target Methodology


TDS trades at less than 2x
EBITDA excluding its USM
ownership

We establish a year-end 2011 price target of $42/share. We use a discounted cash


flow analysis to arrive at our target and assume an 8.5% WACC and 0.5% perpetual
growth rate. TDS trades at 4.0x our 2011 EBITDA estimate compared to other
integrated carriers at 5.2x, while the company only pays a 1.3% dividend yield
compared to 5.7% for peers. Though this represents a sharp valuation gap, we believe
most of it is attributable to a control discount. In addition, most of the potential
upside in TDS shares likely would come from its wireless business, and we do not
currently hold a positive view on USM.

Company Description
TDS is a regional integrated telecom operator providing service predominantly in the
Midwest. The vast majority of approximately $5.0 billion in revenue comes from the
companys 83% ownership stake in publicly traded subsidiary US Cellular.
US Cellular is a regional wireless operator with over 6.0 million customers. The
balance of revenue, approximately 16%, comes from the companys wireline
business. Wireline, including both the ILEC and CLEC businesses, have almost
1.1 million access line customers and approximately 270k broadband customers.
TDS is controlled by the Carlson family.

193

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 80: USM and TDS Ownership


Name

Position with TDS, USM and Principal Occupation

LeRoy T. Carlson

Chairman and Director Emeritus of TDS


Director of USM
President and Chief Executive Officer of TDS
Director of TDS
Director and Chairman of the Board of USM

LeRoy T. (Ted) Carlson, Jr.

Age

Director Since
TDS
USM

93

NA

1987

63

1968

1984

Letitia G. Carlson, M.D.

Director of TDS.
Physician and Associate Clinical Professor at George
Washington University Medical Center

49

1996

NA

Prudence E. Carlson
Walter C.D. Carlson

Director of TDS and Private Investor


Director and non executive Chairman of the Board of TDS
Director of USM
Partner, Sidley Austin LLP, Chicago, Illinois

58
56

2008
1981

NA
1989

Source: Company reports and J.P. Morgan.

Table 81: USM and TDS Shares and Voting Rights


USM

Shares

Votes

Adj. Votes

Series A

33.0

10

330.1

86.1%

Common
Shares

53.4

53.4

13.9%

61.8%

86.4

383.5

100.0%

100.0%

Float

15.6

TDS ownership
Series A

70.8

Common

37.8
Shares

TDS

33.0

Voting %

Economic
Ownership %
38.2%

15.6

4.1%

18.1%

367.8

95.9%

81.9%

10
1

330.1

86.1%

38.2%

100.0%

37.8

9.9%

43.7%

70.7%

Votes

Adj. Votes

Voting %

% of
class

114.7

100.0%

Economic
Ownership %
100.0%

Common

49.8

49.8

43.4%

47.2%

Special

49.3

0.0%

46.7%

Series A

6.5

10

64.9

56.6%

6.1%

64.6

56.3%

13.8%

Shares

% of
class

105.6

29.3%

Carlson Voting

14.6

Common

0.7

0.7

0.6%

0.7%

1.5%

Special

7.5

0.0%

7.1%

15.2%

Series A

6.4

10

63.9

55.7%

6.1%

98.4%

Source: Company reports and J.P. Morgan.

Operating and Financial Outlook


We forecast US Cellular to show modestly improved subscriber growth in 2011 and
2012, relative to subscriber losses in 2009 and 2010. We currently estimate net
additions of 25k in 2011 and 20k in 2012, driven by better retail execution through
implementation of the Belief Project. In addition, ARPU trends should also improve
as smartphone adoption accelerates, and we forecast ARPU growth of 1% in 2011
and 2% in 2012. Our operating metric forecasts drive estimated wireless top-line
growth of 1% in 2011 and 2% in 2012.
On the wireline side, we expect the ILEC side of the business to drive most of the
growth, while CLEC weakness continues. We forecast 1% ILEC revenue growth in
2011 and 2012, while we estimate continued mid-single-digit declines for the CLEC
business. Our US Cellular and TDS Telecom forecasts drive estimated total TDS
revenue growth of approximately 1% in each of 2011 and 2012.

194

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Earnings and Cash Flow Forecast


Limited subscriber and ARPU growth, offset somewhat by smartphone costs, drive
our wireless service margin forecasts of 21.5% in 2011 and 22.6% in 2012, up
slightly from 21.0% in 2010. Timing and magnitude of an LTE launch could cause
downside risk to our forecasts, however. We factored in a slow LTE rollout, and
expect a modest increase in wireless capital spend in 2011 and 2012. As a result,
cash flow should stay relatively flat in 2011, though EBITDA growth could drive a
nice boost in 2012. We forecast modest wireline margin expansion in 2011 and 2012,
as expense reductions help to offset top-line pressure. Our US Cellular and TDS
Telecom forecasts drive estimated total TDS EBITDA margins of 22.6% and 23.5%
in 2011 and 2012, respectively, up from 22.1% in 2010. We expect consolidated
capital expenditures to remain relatively flat in 2011 and 2012, resulting in estimated
free cash flow growth of 8% and 21%, respectively.

195

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 82: TDS Forecasts and DCF


$ millions except per share and per sub values, subs in thousands
2006
WIRELESS
Subscribers
% change, y-t-y
Penetration
Net additional subscribers
% change, y-t-y
Penetration gain
Churn rate, monthly

2007

2008

2009

2010E

2011E

2012E

2013E

2014E

2015E

2016E

2017E

2018E

2019E

2020E

CAGR 1020E

5,815
6%
12.6%

6,102
5%
13.0%

6,196
2%
13.0%

6,141
-1%
12.8%

6,123
0%
12.6%

6,148
0%
12.6%

6,168
0%
12.5%

6,186
0%
12.5%

6,203
0%
12.5%

6,219
0%
12.5%

6,233
0%
12.5%

6,246
0%
12.5%

6,258
0%
12.5%

6,268
0%
12.5%

6,277
0%
12.5%

0%

310
-35%
0.67%

281
-9%
0.60%

94
-67%
0.20%

(55)
-159%
-0.11%

(18)
-67%
-0.04%

25
-239%
0.05%

20
-20%
0.04%

18
-10%
0.04%

17
-6%
0.03%

16
-7%
0.03%

14
-8%
0.03%

13
-9%
0.03%

12
-10%
0.02%

10
-11%
0.02%

9
-11%
0.02%

SM

1.8%

2.1%

1.9%

2.1%

1.9%

1.9%

1.9%

1.8%

1.8%

1.8%

1.8%

1.8%

1.8%

1.8%

1.8%

-1%

3,473
14%
$41
4%

3,946
14%
$44
7%

4,243
8%
$47
5%

4,215
-1%
$47
1%

4,183
-1%
$47
0%

4,229
1%
$48
1%

4,300
2%
$48
1%

4,366
2%
$49
1%

4,392
1%
$49
0%

4,418
1%
$49
0%

4,444
1%
$49
0%

4,466
0%
$49
0%

4,487
0%
$49
0%

4,507
0%
$49
0%

4,527
0%
$49
0%

1%

865
24%
26.9%

1,033
19%
28.1%

1,015
-2%
25.8%

926
-9%
23.6%

825
-11%
21.0%

858
4%
21.5%

919
7%
22.6%

1,043
13%
25.2%

1,111
7%
26.7%

1,136
2%
27.1%

1,159
2%
27.5%

1,171
1%
27.6%

1,183
1%
27.8%

1,194
1%
27.9%

1,206
1%
28.1%

4%

456
2%

435
-5%

393
-10%

356
-9%

331
-7%

308
-7%

286
-7%

269
-6%

256
-5%

245
-4%

236
-4%

229
-3%

222
-3%

215
-3%

209
-3%

-5%

646
-3%

668
4%

611
-9%

600
-2%

615
3%

621
1%

629
1%

636
1%

643
1%

651
1%

656
1%

657
0%

655
0%

654
0%

652
0%

1%

265
-14%
41.1%

299
13%
44.8%

260
-13%
42.5%

234
-10%
39.1%

244
4%
39.7%

250
2%
40.2%

255
2%
40.6%

258
1%
40.6%

261
1%
40.6%

264
1%
40.6%

266
1%
40.6%

266
0%
40.6%

266
0%
40.6%

265
0%
40.6%

265
0%
40.6%

1%

456
2%

435
-5%

393
-10%

356
-9%

331
-7%

308
-7%

286
-7%

269
-6%

256
-5%

245
-4%

236
-4%

229
-3%

222
-3%

215
-3%

209
-3%

-5%

Revenue
% change, y-t-y

236
2%

237
0%

220
-7%

199
-9%

188
-6%

177
-6%

166
-6%

158
-5%

152
-4%

148
-3%

144
-3%

141
-2%

138
-2%

136
-2%

133
-2%

-3%

EBITDA (operating cash flow)


% change, y-t-y
EBITDA margin

23
87%
9.8%

38
64%
16.0%

42
10%
18.9%

28
-32%
14.1%

27
-4%
14.3%

26
-2%
15.0%

25
-6%
15.0%

24
-5%
15.0%

23
-4%
15.0%

22
-3%
15.0%

22
-3%
15.0%

21
-2%
15.0%

21
-2%
15.0%

20
-2%
15.0%

20
-2%
15.0%

-3%

CONSOLIDATED
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin

1,150
11%
26.4%

1,373
19%
28.2%

1,317
-4%
25.9%

1,194
-9%
23.8%

1,101
-8%
22.1%

1,136
3%
22.6%

1,198
5%
23.5%

1,323
10%
25.6%

1,393
5%
26.8%

1,420
2%
27.2%

1,445
2%
27.5%

1,457
1%
27.7%

1,468
1%
27.8%

1,478
1%
27.9%

1,489
1%
28.0%

3%

10%

Revenue
% change, y-t-y
Local ARPU, monthly
% change, y-t-y
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin
ILEC
Access lines
% change, y-t-y
Revenue
% change, y-t-y
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin
CLEC
Access line equivalents
% change, y-t-y

Taxes Paid
% change, y-t-y

0%

(116)
-17%

(269)
131%

(30)
-89%

(135)
348%

(115)
-14%

(135)
17%

(155)
15%

(198)
28%

(227)
15%

(241)
6%

(255)
6%

(265)
4%

(277)
4%

(290)
5%

(304)
5%

Capital Expenditures

727

700

732

671

743

738

732

732

712

713

693

673

652

631

609

-2%

EPS
Diluted EPS, reported

$1.39
$1.39

$3.25
$3.23

$0.81
$0.80

$1.80
$1.80

$1.44
$1.44

$1.67
$1.66

$1.92
$1.92

$2.50
$2.49

$2.92
$2.91

$3.16
$3.15

$3.41
$3.40

$3.62
$3.61

$3.86
$3.84

$4.12
$4.10

$4.42
$4.40

12%
12%

2009
3.7x
20.0x

2010E
4.0x
25.0x

2011E
3.9x
21.7x

2012E
3.7x
18.8x

2013E
3.3x
14.4x

4.2x
0.0x

4.6x
0.0x

4.5x
0.0x

4.2x
0.0x

3.8x
0.0x

Current
$49.52
86.4
71.6
3,544
637
4,181

Targets
$44.87
86.4
71.6
3,212
637
3,848

4,421
240
0.9x

5,061
1,212
4.5x

Private Market Valuation Summary

Valuation Multiples @
2011E
8,510
(1,447)
(4,231)
2,832
3,306
6,137

NPV of EBITDA 12E -- 20E


NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value
PMV of consolidated operations
PMV -- minority stakes
Less: US Cellular public shareholder stake
Less: net long-term debt, end of period (1)
PMV -- Equity

695
(665)
(637)
5,530

Current TDS shares outstanding


Current TDS.S shares outstanding
Fully diluted shares outstanding

57
48
105
$52.52
20%
$42.01
$0.47
18%

PMV per share


Private to public discount
12-month DCF value per share
Plus annual dividend
Upside to target

Valuation Based on Forward FCF Multiple


EBITDA
Free Cash Flow
FCF / diluted share
EV/FCF
FCF yield

Valuation Based on Forward P/E Multiple


EBITDA
EPS - Continuing
% change, y-t-y

2010E
1,101
153
$1.45
28.9
4.0%

2011E
1,136
165
$1.57
26.7
4.4%

2012E
1,198
200
$1.90
22.1
5.3%

2013E
1,323
279
$2.65
15.9
7.4%

2010E
1,101
$1.44
-20%

2011E
1,136
$1.66
15%

2012E
1,198
$1.92
15%

2013E
1,323
$2.49
30%

Note: Dollars in millions, subscribers and pops in thousands except where noted.
(1) We assume that funding needs are met with debt. Debt is net of cash and cash equivalents.

Source: Company reports and J.P. Morgan estimates.

196

Current EV/EBITDA
Current PER
Price Target EV/EBITDA
Target PER

$35.94
EV
4,421

5,061

TDS Value Check @ USM Price:


USM diluted shares outstanding
USM shares owned by TDS
USM equity value in TDS
USM-level net debt
USM EV in TDS
TDS EV
Wireline EV
Wireline EV/EBITDA

Valuation Grid - 2011E DCF Value per Share


Terminal Growth Rate (across) on 2020 EBITDA, Discount Rate (down)
42 -0.5%
0.0%
0.5%
1.0%
1.5%
6.5%
52
55
59
63
67
7.5%
45
47
49
52
55
8.5%
39
40
$42.01
44
46
9.5%
34
35
37
38
40
10.5%
30
31
32
33
34
Valuation Grid - year-end value per share
Free Cash Flow Multiple on FCF
10x
15x
20.0x
22x
25x

2011
16
24
$31.42
35
39

2012
19
28
$37.97
42
47

Valuation Grid - year-end value per share


P/E Multiple on fully-taxed EPS
10x
15x
20.0x
22x
25x

2011
17
25
$33.19
37
41

2012
19
29
$38.32
42
48

North America Equity Research


13 January 2011

Wireline Operators

Wireline Operators

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

197

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

198

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Frontier Communications Corp


Initiate with Overweight Rating and $11 YE11 Price Target
We prefer Frontier within our RLEC coverage, as we find the 7.8% dividend yield,
the highest within our telecom coverage and the highest in the S&P 500, attractive
and believe it is sustainable. The company is undergoing a major transition to
integrate the acquired Verizon assets, which increases its risk profile, though early
signs from initial integration milestones have been comforting. Nevertheless, we
expect Frontier to put in the necessary capital investment as well as implement its
local engagement strategies to begin turning around these markets as early as 1H11.
The roadmap to an operational turnaround, likely upward revisions on synergy
guidance, as well as its attractive dividend make Frontier our top pick within RLECs.

Overweight
Frontier Communications (FTR;FTR US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End Date

9.43
12 Jan 11
9.84 - 6.96
9.33
Dec
989
11.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.12
0.09
0.17
0.01
0.38
0.55

0.14A
0.11A
0.03A
0.09A
0.36A
0.42A

0.10
0.10
0.09
0.09
0.38
0.37

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on Adjusted Proforma.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Long road of integration ahead
Frontiers recent acquisition of Verizon wireline assets has initiated a long and
potentially cumbersome integration period though we are getting more comfortable
as the company meets timeline requirements and beats synergy targets. The company
has successfully completed the integration of its first market, West Virginia, with
minimal disruption. While the integration of the remaining 13 states will be a major
effort, and a large portion of the synergies will be realized upon full conversion, we
look for upward revisions on the synergy number as 2011 progresses and
management finds additional pockets of savings.
A turnaround likely in acquired markets
While commentary around performance in acquired markets has been promising, we
have yet to see a meaningful operational improvement driven by Frontiers execution
and local marketing approach. We expect to see top-line synergies come from FTRs
broadband rollout in these regions and look for access line trends to improve
meaningfully in 2011.
Frontiers management team has demonstrated strong execution
The legacy Frontier markets have demonstrated some of the lowest access line loss
rates within RLECs, of ~6%, and generated industry-high margins of ~55% in 2009.
Frontier is now in the process of system conversions as well as implementation of its
local engagement approach at the recently acquired Verizon assets, which present
significant opportunity for improvement given the much worse access line losses of
~11% y/y as of 3Q10.
199

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Moving towards investment-grade metrics


Frontiers balance sheet has improved significantly since the close of the transaction,
with leverage decreasing from 4x trailing EBITDA to ~2.9x currently. While we
expect a high ~80% payout ratio for 4Q10, we believe the $0.75 dividend is
sustainable and we model that equating to a payout ratio of ~65% in 2013, as
synergies come through and capex overhang diminishes.
Pause on acquisitions, for a while
Frontier will have limited ability to engage in further consolidation over the next
18 months, as management has its hands full with the integration process, and the
Reverse Morris Trust structure of the deal doesnt allow Verizon shareholders
ownership to fall below 50% of the company. We expect management could start
evaluating new acquisition opportunities as early as late 2011.

Investment Risks
Large transaction; integration could be cumbersome
The Verizon transaction, which closed on July 1, has tripled the size of Frontier. The
company has delivered a smooth transition in its West Virginia market, representing
~13% of its total acquired access lines, which it committed to integrate prior to the
close of the deal. The conversion of the remaining 13 states is expected to start in
2011, and any operational disruptions could be costly from a financial or customer
retention perspective.
Further deterioration in economic fundamentals could impact revenue
Further deterioration in economic fundamentals poses risk to access line and revenue
trends as consumer markets could continue to deteriorate at current high levels and
the expected recovery in business markets likely would be pushed further out. This
could decrease the traction the company could get with its strategic broadband and
business growth initiatives.
Regulatory revenue declines could be significant
Roughly 12% of Frontiers pro forma revenues (including surcharges) rely on
regulatory streams, with 6% of revenues coming from switched access and 4% of
revenues from USF. Year to date regulatory revenues are down 9% y/y, and we
expect this revenue stream to continue to diminish gradually. While concerns around
a sudden disruption to regulatory revenues have abated somewhat, policy reform that
would decrease USF revenue faster than expected or significant access rate cuts
could still significantly affect the top line.

200

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Company Description
Frontier is the fifth-largest wireline provider based on access lines. The companys
footprint is mostly rural, with access line density of 24 lines (13 lines before Verizon
transaction). Serving ~6 million access lines and 1.7 million HSI connections in 27
states, the company is in a phase of transformation. Legacy Frontier had operations
in 24 states, serving ~2 million access lines, 647k Internet subs, and 176k video subs.
The company acquired wireline assets from Verizon in a $3.2 billion deal that closed
on July 1, 2010, which tripled the size of the company. Frontier was called Citizens
Communications until July 31, 2008.
The Verizon deal: Legacy Frontier acquired ~4 million access lines from Verizon in
a stock deal that closed at the end of 2Q10. The acquired properties also had
1.1 million HSI and 117k FiOSTV customers as well as 200k+ DirecTV wholesale
subscribers. These landline operations pass ~5.1 million homes in Arizona, Idaho,
Illinois, Indiana, Michigan, Nevada, N. Carolina, Ohio, Oregon, S. Carolina,
Washington, W. Virginia, and Wisconsin, as well as portions of California bordering
Arizona, Nevada, and Oregon.
Figure 84: Frontier Footprint

Source: Company reports (used with permission).

Valuation and Price Target Methodology


We establish a year-end 2011 price target of $11. We reach our price target using our
discounted cash flow analysis, which assumes a terminal growth rate of -1% and a
discount rate of 6.7%. Our price target, together with the annual dividend of $0.75,
implies an expected total return of 22% over the next 12 months. Our price target of
$11 implies an EV/EBITDA multiple of 7xour 2011 adjusted EBITDA estimate.

201

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 83: Frontier Pro Forma Access Line Breakdown by State


Access line trends in the
acquired territories
should recover in 2011

West Virginia
Indiana
New York
Illinois
Ohio
Washington
Michigan
Pennsylvania
Wisconsin
Oregon
North Carolina
Minnesota
California

10.8%
10.2%
9.7%
9.5%
9.0%
8.2%
7.5%
6.1%
4.9%
4.6%
3.7%
3.0%
2.4%

Arizona
Idaho
South Carolina
Tennessee
Nevada
Iowa
Nebraska
Alabama
Utah
Georgia
New Mexico
Montana
Mississippi
Florida

2.2%
1.9%
1.8%
1.1%
0.8%
0.6%
0.6%
0.4%
0.3%
0.3%
0.1%
0.1%
0.1%
0.1%

Source: Company reports.

Legacy Territories Tracking Well


Legacy Frontiers access line loss trends are towards the lower end of the wireline
peer group, given the rural profile of 13 access lines per square mile. The company
has seen a gradual improvement in legacy access line loss trends over the past year,
exiting 3Q10 at a loss rate of 6% vs. 6.1% in 2009 and 7.2% in 2008. In 3Q10, y/y
residential line losses came in at 7.4%, the same as in 3Q09, but better sequentially
compared to the 2Q10 level. Increased bundled sales and price protection plans, as
well as a low number of housing moves in the region, have helped keep disconnects
under control. Business line loss trends have also been on a trajectory of gradual
improvements in y/y trends, and after seven consecutive quarters of reduced y/y
losses, the loss rate was only 3.5% exiting 3Q10.
We look for ongoing gradual improvement in both residential and business access
line trends in legacy markets, and look for loss rates of 6.4% and 2.3%, respectively,
exiting 2011.
Figure 86: Y/Y Access Line Losses at Legacy FTR
change y/y

Figure 85: Frontier Access Line Breakdown

Business
36%

0.0%
-2.0%
-4.0%
-6.0%
-8.0%
-10.0%

64%

2Q
08
3Q
08
4Q
08
1Q
09
2Q
09
3Q
09
4Q
09
1Q
10
2Q
10
3Q
10
4Q
10
E

Residential

Access lines
Source: Company reports.

202

Source: Company reports.

Residential

Business

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Executing Towards an Inflection Point in Acquired Markets


The relatively more urban assets acquired from Verizon, with 37 lines per square
mile, are tracking at a line loss rate of 10.9%, faster than legacy Frontier (6%) as well
as the more urban Verizon territories (8.5%). Frontiers management believes it can
drive 200-300 bps of improvement in access line loss trends in acquired territories in
the near term, by increasing broadband reach and implementing its local market
approach. The absolute line losses in these territories have shown some level of
improvement after the deal closed. There was a meaningful improvement in
disconnects upon closing of the transaction, though gross adds came in light in 3Q.
We expect to see more meaningful improvement in subscriber trends in 2011, when
Frontiers marketing push gains traction and investment starts to bear fruit. We
expect the access line loss trends in these territories to come in at ~9% in 2011, and
decline to ~8% in 2012.

Increasing Broadband Reach Should Lead to Positive


Trends
Frontier legacy markets have ~92% broadband reach. Broadband availability at the
acquired Verizon territories, however, is well below those levels. Those markets had
~63% reach at the time of acquisition, and Frontier is committed to raising reach to
85%, which should drive penetration levels much higher. Management plans to
expand broadband reach in the acquired territories to near levels in the companys
legacy markets over a three- to five-year time frame.
In the first quarter after the closing of the deal, Frontier increased broadband
availability by 27k households, and expects this to increase to 300k (a ~6%
improvement in reach in acquired regions) by the end of 2010. We expect the
company to continue aggressively deploying broadband in its acquired footprint in
2011.

Verizon had built but not


marketed DSL territories

Spinco saw tough FiOS trends


Internet trends in the acquired territories have seen significant deterioration, with
FiOS adds slowing significantly and broadband incurring net sub losses over the past
few quarters, as Verizon neglected marketing in those territories and cable was
aggressive. The acquired territories have FIOS built in Indiana (early build), Oregon,
and Washington (later builds).
We expect broadband trends to improve significantly in 2011 as Frontiers local
engagement approach and increased broadband reach bear fruit. We model
broadband penetration of access lines in legacy markets to reach 38% in 2012 from
30% in 2009, and the penetration in Spinco markets to reach 36% from 25% in 2009.
We look for 35k HSI adds in Spinco territories in 2011, providing 3%+ growth y/y,
but believe our estimate could be too conservative. We model terminal penetration of
access lines reaching 50%+ in both legacy and acquired territories.

203

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 87: HSI Trends in Legacy FTR Markets


140
120
100
80
60
40
20
0

Figure 88: HSI Trends in Acquired Territories


50%
40%

150

50%
40%

100

30%
20%
10%
0%
2007

2008

2009 2010E 2011E 2012E 2013E 2014E


HSI adds

30%

50

20%

10%
0%

-50
2007

2008

HSI penetration

Source: Company reports and J.P. Morgan estimates.

2009 2010E 2011E 2012E 2013E 2014E


HSI adds

HSI penetration

Source: Company reports and J.P. Morgan estimates.

Spinco Head-to-Head with DOCSIS 3.0-Enabled Cable


In Frontiers legacy footprint, the company mainly competes with Time Warner
Cable (~35% overlap), followed by Comcast at ~15%, Charter at 10%, and other at
<10%. The acquired territories have a much heavier overlap with Comcast; as a
result, the company now faces competition from Comcast in 32% of its overall
footprint, followed by TWC and Charter at 23% and 14%, respectively.
The high overlap with Comcast exposes acquired properties to a head-to-head
competition with DOCSIS 3.0-enabled cable lines. However, the highest exposure to
DOCSIS 3.0 is observed in Washington, Oregon, and Indiana, which are Frontiers
FIOS markets, as well as Ohio. The acquired territories only have 50% cable VoIP
overlap (vs. legacy markets at ~73%), which provides opportunity for higher
penetration levels through bundling in these regions.

Failures of Verizons Prior Spinouts Keep Skeptics Bearish


Fairpoints 2008 acquisition of Verizon access lines and the sale of Verizons lines in
Hawaii to private equity firm Carlyle (Hawaii Telecom) in 2005 both resulted in the
spinout company filing for bankruptcy. In our view, the main issue in both of these
deals was the integration or spinout of access lines, which was executed poorly, with
customers experiencing significant disruption, leading to rising churn rates.
Execution issues, combined with very high leverage, contributed eventually to filing
for bankruptcy. In 2006 Verizon spun out its yellow pages business, Idearc, which
also filed for bankruptcy in 2009.
Table 84: Verizon Has Had Three Spin-Off/Asset Sales that Resulted in Chapter 11 Filings
Time

Acquirer

Acquired asset

2008

Fairpoint

Verizon's Maine, New Hampshire and Vermont


properties:1.6m access lines, ~230k DSL subs

2005

Carlyle (Hawaii Telecom)

Verizon's Hawaii assets: 715,000 access lines

2006

Idearc

Yellow pages business

Source: Company reports.

204

Payment
US$2.715bn- $1.015 bn of
FairPoint common stock, $1.7 bn
cash& debt

Subsequent events
Stock traded as low as -20%, closed -12% on the day.
Major service and tech issues. Filed Chapter 11 in
October 2009
Transition issues, financial/customer service problems. In
US$1.65bn in cash
3 years the company lost 21% of customer base. Filed
Chapter 11 in December 2008.
The spun-off company was loaded Filed Chapter 11 in March 2009. Reemerged as
with ~US$9bn debt
SuperMedia in December 2009.

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Integration risk is far smaller for Frontier


We view Frontiers transaction to be significantly different from Fairpoints, placing
Frontiers a lot more favorably. On the systems conversion front, Frontier had to
convert only the West Virginia state prior to the closing of the deal, which it was able
to accomplish with little disruption on the customer end.
For the other 13 states that were acquired, Frontier will owe Verizon $94 million
annually (a flat rate; it does not scale down until the end) to manage the customer
service and billing systems as it gradually transfers customers to its own systems.
The company has up to five years to complete the conversion, which should provide
ample time for transition, but we expect the process to be completed by the end of
2012. Fairpoints outsourcing agreement was much more expensive and had a rising
cost structure over time, while Frontiers only increase will be CPI adjustments.
On the leverage front, the Verizon transaction was a significantly deleveraging one
for Frontier, allowing the company to attain investment-grade-like metrics post close.
Frontiers leverage of ~4x trailing EBITDA decreased to ~2.9x post close. By
contrast, the Fairpoint transaction was a levering one that limited the companys
financial flexibility in a significant way.

Financial Outlook
We expect synergy guidance to
steadily increase through 2011

Expect further upside to synergies


Frontier reported that it has already reached $252 million of annual run-rate
synergies as of the end of 3Q10, and management increased its annual Opex synergy
target from $500 million to $550 million, and reiterated that the companys internal
targets are much higher. We estimate ~$750 million in annual synergies eventually.
Note that this represents Opex only, and does not include any revenue improvements
that Frontier can make to the Verizon properties, which could drive significant
upside for the company.
By the end of 2010 Frontier expects to reach $300 million in annual run-rate
synergies and, of the rest ($250 million remaining), $94 million will go away on the
last day of conversion when the payment to Verizon for the 13 states finishes. The
company expects near-term synergies from circuits and traffic moving from the
Verizon network to owned backbone, and switching long-distance calling from MCI
to another provider, which represent total annual costs of ~$60-70 million. The 1,700
contractors who came over from Verizon should see salary reductions that total a run
rate of ~$86 million. On top of that, vendor and lease rationalization should drive
further savings. The company expects to reach $400 million in run-rate synergies in
2011, mainly from offloading of traffic, which should occur on a straight-line basis
through the year, representing roughly an incremental $6 million per quarter.
We expect to see additional cost synergies from further personnel cost rationalization
that is not assumed in current synergy guidance. Frontier has committed to
maintaining some IT personnel such as fixed technicians and cost service
representatives for an 18-month period after the closing of the deal, and these
employees will be redundant once the conversion period ends, providing opportunity
for further Opex rationalization.

205

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

While we do not model for revenue to recover, we expect to see some level of topline synergies as well. Frontier has committed to increase its broadband reach in the
Verizon properties from 63% to 85% and we expect reach could eventually expand
to ~92%, comparable to the level at legacy Frontier. This should help Frontier
achieve a major improvement in line loss rates, from the current run rate of ~11% to
~8% by 2012E, and drive customer ARPU well above current levels.
Leverage improving
The acquisition of Verizon lines was a deleveraging transaction for Frontier, taking
leverage from 4x trailing EBITDA before the transaction to ~2.9x at the end of
3Q10. However, we estimate the transaction raises the payout ratio to a high 75%+
for 4Q, despite the 25% dividend cut from $1.00 to $0.75. Despite the dividend cut,
which lowered the yield from 13% to 7.8%, the yield still remains the highest in
S&P 500. We model payout ratio declining to mid-60% as synergies come through.

Earnings and Cash Flow Forecast


Capex should ramp up near to medium term
Verizon spent only ~9% of run-rate revenue on maintenance capex in the Spinco
territories, compared to Frontiers 12% in legacy territories. Frontier will need to
make significant upfront investments to help improve the access line performance in
the underinvested Verizon markets and increase broadband reach. Moreover, we
expect some FiOS investment in acquired territories in states in which Verizon has
already rolled out service.
Management expects 12% of revenue plus $100 million annually will be sufficient
for the networks, FiOS, and Frontiers ongoing investments for the next three years,
or roughly $700 million annually, excluding integration capex. Thereafter,
management believes the capex level could fall below 12%, possibly to 10% of
revenue. We model capex eventually declining to 11% of service revenue.

206

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 85: Frontier Valuation Model


2009

2010E

2011E

2012E

2013E

2014E

2015E

2016E

2017E

2018E

2019E

2020E

Access lines
% change, y-t-y
Additional lines

6,315
NM

5,744
-9.1%
(572)

5,304
-7.7%
(440)

4,938
-6.9%
(366)

4,710
-4.6%
(228)

4,526
-3.9%
(183)

4,396
-2.9%
(130)

4,296
-2.3%
(101)

4,227
-1.6%
(69)

4,171
-1.3%
(56)

4,124
-1.1%
(47)

4,078
-1.1%
(46)

HSI connections
% change, y-t-y
Additional lines

1,696
NM

1,696
0.0%
(1)

1,749
3.1%
53

1,802
3.0%
53

1,848
2.6%
46

1,888
2.2%
40

1,921
1.7%
32

1,946
1.3%
26

1,967
1.1%
21

1,984
0.8%
17

1,997
0.7%
13

2,007
0.5%
11

Voice Revenue
% change, y-t-y

3,119
NM

2,807
-10.0%

2,560
-8.8%

2,362
-7.7%

2,215
-6.2%

2,110
-4.7%

2,029
-3.9%

1,966
-3.1%

1,918
-2.5%

1,881
-1.9%

1,849
-1.7%

1,819
-1.6%

Data and Internet Revenue


% change, y-t-y

1,844
NM

1,841
-0.2%

1,839
-0.1%

1,881
2.3%

1,928
2.4%

1,967
2.1%

1,999
1.6%

2,023
1.2%

2,041
0.8%

2,052
0.6%

2,059
0.3%

2,062
0.2%

768
NM

682
-11.2%

627
-8.0%

590
-6.0%

554
-6.0%

521
-6.0%

490
-6.0%

460
-6.0%

433
-6.0%

415
-4.0%

399
-4.0%

383
-4.0%

Consolidated Revenue
% change, y-t-y

6,071
NM

5,685
-6.4%

5,383
-5.3%

5,192
-3.6%

5,058
-2.6%

4,962
-1.9%

4,884
-1.6%

4,819
-1.3%

4,764
-1.2%

4,724
-0.8%

4,685
-0.8%

4,647
-0.8%

EBITDA (operating cash flow)


% change, y-t-y
EBITDA margin

2,878
NM
47.4%

2,532
-12.0%
44.5%

2,546
0.5%
47.3%

2,524
-0.9%
48.6%

2,554
1.2%
50.5%

2,496
-2.3%
50.3%

2,447
-2.0%
50.1%

2,410
-1.5%
50.0%

2,377
-1.4%
49.9%

2,353
-1.0%
49.8%

2,329
-1.0%
49.7%

2,305
-1.0%
49.6%

Adj. EBITDA (operating cash flow)


% change, y-t-y
Adj. EBITDA margin

2,968
NM
48.9%

2,703
-8.9%
47.6%

2,646
-2.1%
49.2%

2,624
-0.8%
50.5%

2,634
0.4%
52.1%

2,576
-2.2%
51.9%

2,527
-1.9%
51.7%

2,490
-1.5%
51.7%

2,457
-1.3%
51.6%

2,433
-1.0%
51.5%

2,409
-1.0%
51.4%

2,385
-1.0%
51.3%

Capital Expenditures
Capital expenditures/revenue

790
13.0%

720
12.7%

715
13.3%

692
13.3%

625
12.4%

551
11.1%

507
10.4%

499
10.4%

492
10.3%

487
10.3%

482
10.3%

477
10.3%

Regulatory Revenue
% change, y-t-y

EPS
Dividend per share

$
$

0.44 $
1.00 $

0.36 $
0.88 $

0.38 $
0.75 $

0.42 $
0.75 $

2011E
13,399
(1,586)
(3,045)
8,768
9,802
18,570

2012E
11,034
(1,375)
(2,397)
7,262
9,802
17,063

2011E
16,618
(2,014)
(3,599)
11,005
8,803
19,809

Less: net long-term debt, end of period


PMV -- Equity

(6,959)
11,611

(8,336)
8,727

(7,684)
12,125

989
0
989

987
0
987

989
0
989

Current shares outstanding


Shares to be issued from warrants/options
Fully diluted shares outstanding

$11.74
10%
$10.57

PMV per share


Private to public discount
12-month DCF value per share
Plus dividend
12 month return
2011E

Free Cash Flow


FCF / diluted share
FCF yield
Dividend
Dividend yield
Payout ratio

0.52 $
0.75 $

0.55 $
0.75 $

Valuation Multiples @

Private Market Valuation Summary


NPV of EBITDA 11E -- 20E
NPV of taxes paid 11E -- 20E
NPV of capex 11E -- 20E
NPV of free cash flow 11E -- 20E
NPV of 2020 terminal value @ 7x EBITDA
PMV of consolidated operations

Valuation Based on Forward FCF Multiple

0.50 $
0.75 $

2012E

1,064
$1.08
11%
$

$8.84
10%
$7.96
0.75

2013E

1,067
$1.08
11%

Current EV/Adj EBITDA


Current PER
Current Dividend yield
Free cashflow yield
Price Target EV/Adj EBITDA
Target PER
Expected Dividend yield
Free cashflow yield

2014E

$9.68
EV
17,257

18,596

0.63 $
0.75 $

2011E
6.5x
25.2x
7.7%
11.1%
7.0x
27.5x
6.8%
10%

0.67 $
0.75 $

0.69 $
0.75 $

2012E
2013E
6.6x
6.6x
22.9x
19.3x
7.7%
7.7%
11.1%
11.8%
7.1x
25.0x
6.8%
10%

7.1x
21.1x
6.8%
10%

0.71
0.75

2014E
6.7x
18.5x
7.7%
12.2%
7.2x
20.2x
6.8%
11%

Valuation Grid - 2011E DCF Value per Share


Terminal Growth Rate (across) on 2020 EBITDA, Discount Rate (down)
11 -3.0%
-2.0%
-1.0%
0.0%
1.0%
4.7%
13
15
17
20
24
5.7%
11
12
13
15
18
6.7%
9
10
$11.03
12
14
7.7%
8
8
9
10
11
8.7%
7
7
8
8
9

$12.26
10%
$11.03
0.75
22%

1,131
$1.14
12%

0.59 $
0.75 $

2015E

1,167
$1.18
12%

2016E

1,182
$1.20
12%

2017E

1,168
$1.18
12%

2018E

1,155
$1.17
12%

1,152
$1.16
12%

0.75 $
7.7%

0.75 $
7.7%

0.75 $
7.7%

0.75 $
7.7%

0.75 $
7.7%

0.75 $
7.7%

0.75 $
7.7%

0.75
7.7%

70%

70%

66%

64%

63%

64%

64%

64%

Valuation Grid - year-end value per share


Free Cash Flow Multiple on FCF
8x
9x
10x
11x
12x

2011
9
10
$10.76
12
13

2018
9
10
$11.65
13
14

Source: Company reports and J.P. Morgan estimates.

207

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

208

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Windstream Communications
Initiate with Neutral Rating and $14 YE11 Price Target
Access line losses have
shown improvement and
are lowest among peers

We initiate coverage of Windstream with a Neutral rating and a year-end 2011 price
price of $14 per share, implying expected total return of 13% including the
companys US$1/share annual dividend. Windstream has a rural footprint with the
lowest access line density within our RLEC coverage, which has enabled the
company to deliver better operating metrics than its peers. Moreover, management
has been shifting its revenue stream away from the declining consumer voice
business and towards business and data revenues, which positions the company well
for top-line stabilization when macro trends improve. Windstream has been one of
the best-performing stocks within telecom over the past year, with a total return of
37% including its dividend (vs. total return of 15% for the S&P 500). While the
current dividend yield of 7.4% remains one of the highest within telecom, the
companys higher leverage and EV/EBITDA multiple compared to those of wireline
peers, as well as ongoing acquisition risk, keep us sidelined.

Neutral
Windstream Corp (WIN;WIN US)
Company Data
Price ($)
13.37
Date Of Price
12 Jan 11
14.40 - 6.02
52-week Range ($)
Mkt Cap ($ bn)
6.41
Fiscal Year End
Dec
480
Shares O/S (mn)
Price Target ($)
14.00
Price Target End Date 31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.20
0.21
0.18
0.17
0.76
0.82

0.17A
0.17A
0.18A
0.21A
0.73A
0.77A

0.21
0.20
0.21
0.22
0.84
0.80

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on adjusted proforma
res ults. 'Bloomberg' above denotes Bloomberg c onsensus estimates.

Key Investment Points


Rural footprint enables best-in-class metrics
With an access line density of 17 pops per square mile, Windstream has the most
rural footprint in our coverage, which has resulted in industry-low access line losses
of 3.7%, as well as HSI penetration of 39% and video penetration of 13% of total
access lines, both at the high end of the peer group. The increase in broadband
penetration, coupled with tuck-in acquisitions of RLECs and CLECs, has helped
keep revenue losses at a minimum.
Acquisitions targeted to business
For the past couple of years, while its peers were engaging in transformative (in
terms of size) RLEC consolidation, Windstream adopted an acquisition strategy more
focused on smaller, tuck-in RLEC deals and increasing its business-revenue exposure
through CLEC/data center acquisitions. Overall, the companys revenue from
business and broadband has risen from 38% in 2007 to 58% most recently. The
increased exposure to stickier revenue streams which should see growth following a
macro turnaround position the company well for the medium term, in our view.

209

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

FCF offset by higher leverage and risk profile


We estimate Windstream will generate a free cash flow yield of 11.5% in 2011, and
pay out ~65% of free cash as its $1.00/share dividend. While the 11.5% yield is at
the high end of the telecom peer group, Windstream screens as one of the more
highly levered telcos (net debt of 3.4x trailing EBITDA, vs. the rest of RLECs below
3.0x). Its EV/EBITDA multiple of 6.9x 2011E EBITDA is at the higher end of its
RLEC peers, providing little valuation upside, in our view, especially given the lack
of opportunity for meaningful margin expansion.
Ongoing acquisitions an opportunity as well as risk
Despite its already high leverage, Windstrems positioning within the RLECs places
it as a likely consolidator of assets in 2011. With CTLs and FTRs hands likely full
with the integration of their recent large acquisitions, if a major transaction were to
occur in the space, we would expect Windstream could be involved. While, in
general, we view such consolidation opportunities positively from a long-term
perspective, the risk associated with a transaction likely would be negative for nearterm stock performance.

Investment Risks
Upside risks
Earlier-than-expected recovery in economic fundamentals could drive growth in
business revenues
Through acquisitions Windstream has increased its exposure to business markets,
positioning itself well for an eventual macroeconomic recovery. If macro trends were
to improve significantly in 2011, Windstream could see a meaningful growth on the
business revenue front, and our current revenue estimates could be too conservative.
Improvement in access line and HSI trends could drive revenue and cash flow
above expectations
If Windstreams access line losses were to improve more meaningfully, or data
revenue growth came in better than expected, consumer revenues could come in
better than we expect. Moreover, the improved trends could reflect well on margins,
rendering our cash flow estimates too conservative.
Downside risks
Further acquisitions could be risky for near-term free cash flow
The potential for additional acquisitions is another area of risk for the companys
cash flow prospects. While, in general, we like consolidators within the deteriorating
wireline industry from a synergy realization and margin maintenance perspective, a
potential acquisition in the near to medium term could take Windstreams leverage
above the companys historical levels of around 3.2x. This could raise companys
cost of capital and potentially limit its financial flexibility.
Regulatory revenue declines could be significant
Windstreams regulatory revenue exposure is toward the high end of its peer group.
While we expect that revenue stream to continue to diminish gradually rather than
see a major short-term disruption, a faster-than-expected decline in USF subsidies
and access rates could result in a significant hit to the top line and affect the
companys ability to maintain its dividend.

210

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Company Description
Windstream, previously Alltels local wireline business, was merged into Valor in
July 2006, and acquired CTC in 2007. The company currently operates in 16 states,
covering about 3 million access lines, 2 million long distance customers, around
1 million HSI customers, and 300,000-plus digital TV customers to whom it resells
DISH satellite service. Windstream spun off its Directory and Publishing business in
2007 and has closed six tuck-in acquisitions since 2009, most recently Iowa
Telecom, Lexcom, D&E, and NuVox. The companys largest market is Georgia,
followed by Kentucky, Texas, Ohio, and Nebraska, which cumulatively account for
60%+ of access lines.
Figure 89: Windstream Footprint

Source: Company reports (used with permission).

Valuation and Price Target Methodology


We initiate coverage of Windstream with a Neutral rating and establish a year-end
2011 price target of $14 per share. Our target price is based on our DCF analysis,
which assumes a -1% terminal growth rate and 7.2% discount rate. Our target price
implies a dividend yield of 6.9% and an EV/EBITDA multiple of 7x our 2011E
EBITDA estimate.

211

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Buying Assets on Business Side


Windstream followed an active acquisition strategy on the RLEC front, until
November 2009, when the company announced its acquisition of NuVox, a CLEC in
the Southeastern United States and the Midwest, with overlapping footprint.
Windstream leveraged NuVoxs existing billing systems; rather than integrating
NuVox into legacy operations, it transitioned its legacy CLEC customers onto
NuVoxs billing system. With its recent QComm acquisition Windstream continued
to grow broadband and business revenues, and more than doubled its existing fiber
footprint, enabling the company to better serve the business segment, as well as
expanding transport services for wireless carriers.
We expect Windstream to continue to be one of the major consolidators in the
wireline industry, and the company has indicated it will pursue more CLEC and
potentially data center acquisitions, such as Nuvox, KDL, and Hosting, as well as
tuck-in RLEC acquisitions, such as D&E, Lexcom, and Iowa Telecom, when the
pricing is reasonable.
Windstreams sweet spot when it comes to RLEC acquisitions is 100,000-plus
incremental access line from well-run companies in rural regions that would be
accretive to cash flow in the first year. While the company lately has turned to a
more aggressive acquisition strategy on the business lateral, we expect it to continue
to be involved in the RLEC consolidation wave, though at a more selective and
moderate pace.
Table 86: Windstreams 2009-2010 Acquisitions
Acquisition
Price, Close date Revenue OIBDA
Hosted Solutions
310m, 4Q10
51.7m 25.7m
Q-Comm
786m, 4Q10
231m 92.8m
Iowa
1.1bn, 2Q10
275m
130m
Nuvox
647m, 4Q09
561m
115m
Lexcom
141m, 4Q09
44m
24m
D&E
333m, 4Q09
145m
64m

Synergy
1.5m
25m
35m
30m
5m
25m

Access line overview


5 data centers in Raleigh, Charlotte (NC) and Boston, serving 600 customers
Fiber provider in 23 states, CLEC serving 5,500 SMB customers primarily in Midwest
Iowa, Minnesota, 256k access lines, 95k HSI, 26k digital TV
90k business customers in complementary markets in 16 states
Neighboring N. Carolina locations, 23k access lines, 9k HSI, 12k cable TV customers
Pennsylvania,114k ILEC, 46k HSI lines, 47k CLEC

Source: Company reports.

Given the increasing pressure in the consumer voice business, we view Windstreams
reorientation toward business markets as an attractive medium-term strategy. The
company has explicitly stated its interest in shifting its revenue mix toward
broadband and business and has increased the mix of these revenue streams from
38% in 2007 to 58% pro forma for the trailing twelve months.

212

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 90: WINs DSL Penetration Has Been Ramping

50.0%

40

40.0%

30

30.0%

20

20.0%

10

10.0%

0.0%
2Q08

4Q08

2Q09

4Q09

DSL adds

2Q10 4Q10E 2Q11E 4Q11E

DSL penetration of access lines

Source: Company reports and J.P. Morgan estimates.

Higher Leverage than Peers


We estimate that Windstream will generate a free cash flow yield of 11.5% in 2011,
and will pay out around 65% of that as its US$1.00/share dividend. The company has
taken on more leverage than peers, with net debt of 3.4x TTM EBITDA, the highest
among RLECS. This, coupled with the lack of meaningful synergies relative to
EBITDA, would create relatively higher risk to the companys dividend if an internal
or external factor were to hurt the companys FCF prospects. While we do not find
the companys ability to maintain its dividend in the near term or its level of leverage
worrisome, as the company does not have any significant maturities until 2013,
Windstreams dividend could be subject to higher risk compared to those of its
RLEC peers.

Wireline Trends Better


Windstreams more rural footprint has helped it maintain industry-low line loss
levels. The company has been losing access lines at a 4-5% rate since its formation in
2006 and lost 3.7% in 3Q10.
On the residential voice front, the introduction of the price for life promotion
helped improve trends starting in 3Q09. The price for life promotion provides a
wireline and broadband package with a lifetime price guarantee that, in an
environment with potential price hikes from cable MSOs, has resonated well with
Windstreams customer base. However, given the ongoing macro challenges and the
secular trends, there has not been a material improvement from a 4% loss rate in
consumer voice.
We model access line losses to continue to moderate, and expect the company to exit
2011 with an overall line loss rate of 3.4% (vs. 3.7% in 3Q10) and residential line
loss rate of 3.9% (vs. 4% in 3Q10).

213

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Upselling opportunity for higher speeds


With the majority of Windstreams existing customer base at the 1.5 Mbps
broadband speed, and 90%+ of gross adds selecting 3 Mbps speeds or higher,
Windstream is on a trajectory to increase data revenue as its customer base moves to
higher speeds. With broadband stimulus money of $180 million in grants plus its
$60 million internal contribution, the company plans to increase broadband reach
from 90%, and upgrade speeds in its footprint. Currently each speed increment sells
for $5 higher, and the companys top seller is the 6 Mbps plan.
With only 17 pops per square mile, Windstreams rural footprint sees minimal
competition from cable, and cable VOIP is available across only ~63% of the
companys footprint, with Time Warner Cable having the most overlap. There is a
broadband alternative for 75-80% of Windstreams lines, with cable available to
85%. We expect the voice competition from cable providers to increase from the
current 64% level and eventually top out in the low 70s. We estimate Windstreams
exposure to wireless substitution is ~20% and should remain below national averages
going forward.

Financial Outlook
Acquisitions and data growth offset voice revenue losses
Windstream sees a pro forma access line loss rate of ~3.7%, the lowest among its
peer group. Moreover, its increasing data penetration and video resale revenue are
helping keep organic revenue losses to a minimum. Data and special access revenue
accounts for ~40% of service revenue today, up from around 26% in 2008 and 23%
in 2007. We expect the line losses to continue to decline and model Windstream to
exit 2011 at a 3.4% y/y loss rate. We model 69k HSI sub additions for 2011, which
would bring penetration to ~44% of access lines.
Windstream has closed six acquisitions since the beginning of 2009, which add
around $1.3 billion in annual revenue and about $485 million in adjusted OIBDA.
Two of these acquisitions, Nuvox and QComm, were CLECs and one, Hosted
Solutions, was a data center. These acquisitions have helped increase business and
broadband revenue exposure to 58% pro forma, up from 38% in 2007, increasing the
stability of the top line for Windstream. We look for ~1.2% loss on the top line for
2011, excluding the acquisitions of QComm and Hosted Solutions. We look for a
~6.4% y/y decline in voice revenues, partly offset by 4.8% growth in data and
integrated solutions revenue.

214

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Figure 91: WIN 2007 Revenue Breakout

Figure 92: WIN LTM Pro Forma Revenue Breakout


Wholesale
17%

Wholesale
Broadband

28%

and business
38%
Broadband

Residential

and business

25%

58%

Residential
34%

Source: Company reports.

Earnings and Cash Flow Forecast


Managements 2010 FCF guidance implies a payout ratio in the 59-62% range, given
its $1.00/share dividend, and we model 61% payout ratio for 2010, supporting a
7.4% dividend yield at the high end of the range for its peers.
As more of the synergies from recent acquisitions come through, Windstreams
margins should expand in 2011, and we model 48.7% EBITDA margins (50.9% adj.
EBITDA). Net-net, we look for 2011 EBITDA to be essentially flat y/y.
Windstreams net leverage is 3.3x TTM EBITDA, vs. the rest of RLECs at or below
3.0x. The company has maintained a disciplined approach to leverage, capping
leverage at around 3.2x since its spinoff from Alltel in July 2006. We expect the
company to allocate excess FCF to decrease leverage to historical levels unless an
attractive acquisition opportunity arises.
Figure 93: Upcoming Maturities
$ (m)

Source: Company reports.

2,000

$1,746.0

1,500

$1,500.0

$1,421.0

$1,254.0

$1,100.0

1,000
500

$131.0

$47.0

$11.0

0
2011

2012

2013

2014

2015

2016

2017.00

Thereafter

Source: Company reports and J.P. Morgan.

215

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 87: Windstream Valuation Model


2009

2010E

2011E

2012E

2013E

2014E

2015E

2016E

2017E

2018E

2019E

2020E

CAGR
10E-20E

Residential voice lines


% change, y-t-y
Additional lines
Business voice lines
% change, y-t-y
Additional lines
Total Access lines
% change, y-t-y
Additional lines

2,134
NM
(117)
1,011
NM
(47)
3,392
NM
(161)

2,045
-4.2%
(89)
974
-3.7%
(37)
3,269
-3.6%
(123)

1,965
-3.9%
(80)
940
-3.4%
(33)
3,157
-3.4%
(112)

1,893
-3.7%
(72)
920
-2.1%
(20)
3,067
-2.9%
(90)

1,828
-3.4%
(65)
908
-1.3%
(12)
2,992
-2.5%
(75)

1,770
-3.2%
(58)
898
-1.2%
(11)
2,924
-2.3%
(68)

1,718
-3.0%
(52)
890
-0.9%
(8)
2,864
-2.0%
(60)

1,670
-2.8%
(47)
882
-0.8%
(7)
2,810
-1.9%
(54)

1,662
-0.5%
(8)
876
-0.7%
(6)
2,796
-0.5%
(14)

1,658
-0.2%
(4)
870
-0.7%
(6)
2,787
-0.3%
(10)

1,655
-0.2%
(4)
865
-0.6%
(5)
2,778
-0.3%
(9)

1,651
-0.2%
(4)
860
-0.5%
(5)
2,770
-0.3%
(8)

-2%

HSI connections
% change, y-t-y
Additional lines
% change, y-t-y
Consolidated Revenue
% change, y-t-y
EBITDA (operating cash flow)
% change, y-t-y
EBITDA margin

1,224
NM
194
NM
3,957
NM
1,830
NM
46.3%

1,323
8%
99
-49%
3,864
-2.3%
1,863
2%
48.2%

1,392
5%
69
-30%
3,827
-1.0%
1,865
0%
48.7%

1,447
4%
55
-20%
3,800
-0.7%
1,847
-1%
48.6%

1,491
3%
44
-20%
3,787
-0.3%
1,837
-1%
48.5%

1,527
2%
35
-20%
3,766
-0.5%
1,824
-1%
48.4%

1,555
2%
28
-20%
3,748
-0.5%
1,811
-1%
48.3%

1,578
1%
23
-20%
3,715
-0.9%
1,792
-1%
48.2%

1,596
1%
18
-20%
3,689
-0.7%
1,775
-1%
48.1%

1,611
1%
15
-20%
3,671
-0.5%
1,763
-1%
48.0%

1,622
1%
12
-20%
3,653
-0.5%
1,750
-1%
47.9%

1,631
1%
9
-20%
3,634
-0.5%
1,738
-1%
47.8%

2%

Adj. EBITDA (operating cash flow)


% change, y-t-y
EBITDA margin

1,958
NM
49.5%

1,945
-1%
50.3%

1,946
0%
50.9%

1,929
-1%
50.8%

1,919
-1%
50.7%

1,905
-1%
50.6%

1,893
-1%
50.5%

1,873
-1%
50.4%

1,857
-1%
50.3%

1,845
-1%
50.2%

1,832
-1%
50.2%

1,819
-1%
50.1%

-1%

Taxes Paid
Tax rate

(242)
37%

(228)
37%

(236)
37%

(239)
37%

(245)
37%

(250)
37%

(254)
37%

(255)
37%

(258)
37%

(261)
37%

(264)
37%

(267)
37%

2%

Capital Expenditures
Capital expenditures/revenue

412

404
10.5%

428
11.2%

425
11.2%

424
11.2%

421
11.2%

401
10.7%

397
10.7%

394
10.7%

392
10.7%

390
10.7%

388
10.7%

0%

0.95
1.00
(480)

1%
0%
0%

EPS
Dividend per share
Dividend paid

$
$

0.95 $
1.00 $
(437)

Private Market Valuation Summary


NPV of EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value @ 7.01x EBITDA
PMV of consolidated operations
Less: net long-term debt, end of period
PMV -- Equity
Current shares outstanding
Shares to be issued from warrants/options
Fully diluted shares outstanding
PMV per share
Private to public discount
12-month DCF value per share
Plus dividend
12-Month return

Source: Company reports and J.P. Morgan estimates.

216

0.83 $
1.00 $
(464)

0.84 $
1.00 $
(480)

0.85 $
1.00 $
(480)

0.87 $
1.00 $
(480)

0.89 $
1.00 $
(480)

0.90 $
1.00 $
(480)

Valuation Multiples @
2011E
11,626
(1,637)
(2,624)
7,365
6,516
13,881
(6,203)
7,678
480
3
483
$15.91
10%
$14.32
$1.00
13.4%

Current EV/EBITDA
Current PER
Current Dividend yield
Current Free Cashflow yield
Price Target EV/EBITDA
Target PER
Expected Dividend yield
Expected Free cash flow yield

0.91 $
1.00 $
(480)
$13.51
EV
12,723
7.1%

13,113

0.91 $
1.00 $
(480)

0.93 $
1.00 $
(480)

0.94 $
1.00 $
(480)

2011E
6.8x
16.2x
7.4%
11.5%

2012E
6.9x
15.9x
7.4%
11.2%

2013E
6.9x
15.5x
7.4%
11.0%

2014E
7.0x
15.2x
7.4%
10.7%

7.0x
17.1x
6.9%
10.8%

7.1x
16.8x
6.9%
11.2%

7.1x
16.4x
6.9%
11.0%

7.2x
16.2x
6.9%
10.8%

Valuation Grid - 2011E DCF Value per Share


Terminal Growth Rate (across) on 2020 EBITDA, Discount Rate (down)
14 -1.5%
-1.0%
-0.5%
0.0%
0.5%
6.0%
15
16
18
19
21
6.5%
14
15
16
17
18
7.0% 12.4
13.2
$14.03
15
16
7.5%
11
12
13
13
14
8.0%
10
11
11
12
13

-1%

-2%

-1%
-1%

North America Equity Research


13 January 2011

Wireless Tower Operators

Wireless Tower Operators

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

217

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

218

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

American Tower
Initiate with Overweight Rating and $60 YE11 Price Target
We are initiating coverage of American Tower with an Overweight rating and a yearend 2011 price target of $60. American Tower, with its scale, conservative financial
position, and focus on profitability, has been the most stable tower company and
offers an attractive risk/reward at this level, in our view. International opportunities
could drive higher returns for American Tower in comparison to those in domestic
markets and the company has diversified into eight countries already, with no
country but the US accounting for more than 10% of revenue. The firms relatively
low leverage at 4x EBITDA provides us with additional comfort and gives the
company the ability to do a larger tower deal or buyback stock. Finally, we see REIT
conversion as the most likely case for American Tower and include it in our target
valuation of $60 per share; however, if this does not occur, our estimated YE11
valuation for the company would be only $45, implying 11% downside risk from the
current level.

Overweight
American Tower (AMT;AMT US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ mn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

50.70
12 Jan 11
53.52 38.09
20,455.17
Dec
403
60.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.14
0.14
0.17
0.16
0.61
0.62

0.24A
0.25A
0.23A
0.24A
0.95A
0.93A

0.25
0.26
0.27
0.28
1.07
1.13

Source: Company dat a, Bloomberg, J.P. Morgan G AAP es timates . 'Bloomberg' above denotes Bloomberg
consensus estimates .

Key Investment Points


Solid revenue and adjusted EBITDA growth from positive operating trends
We model American Tower revenue growth of 15.2% and adjusted EBITDA growth
of 16.2% in 2011, after revenue growth of 14.6% in 2010, due to revenue growth
from incremental acquisitions as well as continued strength in lease-ups,
augmentations, and portfolio growth. There could be upside to our 2011 estimate
once Sprint Nextel starts to augment its CDMA sites for 800 MHz capability or if
Clearwire reaccelerates its network build. T-Mobile, Leap Wireless, and MetroPCS
could have an impact if any significant network investments are made. We expect
continued strength from Verizon LTE augmentations, but note that AT&Ts are now
built in to GAAP revenue.
Solid recurring FCF (RFCF) growth of 11.5% in 2011
We estimate solid RFCF growth of 11.0% to $1.1 billion in 2011, from $1.0 billion
in 2010. We model a 7% organic CAGR longer term. Our year-end 2011 price target
of $60 per share implies a yield of 4.8% on our 2011 RFCF estimate of $2.90 per
share.

219

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Low relative leverage gives American Tower flexibility, buyback potential


American Tower has a net leverage ratio of 3.9x versus 5.2x for Crown Castle and
7.3x for SBA Communications. The lower relative leverage allows the company to
potentially take on substantial debt for either a large tower purchase or stock
buyback. American Tower has already made some significant international moves
recently with a joint venture (TowerCo Ghana) and the purchase of Cell C towers in
South Africa. Large domestic tower portfolios are probably not as attractive at
current prices to American Tower, which leaves continued stock buyback as an
alternative. In our view, American Tower has the ability to lever up by 1-2 turns of
EBITDA to buy back $1.4-2.8 billion in stock, and will likely authorize more after it
completes the $450 million remaining on its current $1.5 billion authorization
(expiring July 2011).
International tower expansion could bring greater returns
American Tower sees higher potential returns from international tower builds or
purchases. While we expect international towers to have higher operating and
currency risks than domestic towers, American Towers international diversification
across eight countries is slowly reducing the risks. New builds are cheaper due to
lower soft and construction costs, and many markets are still developing and need
more new towers built than in the US. Collocation in some markets is starting to be
adopted by more carriers and some carriers are selling tower portfolios to focus on
core strategies. Our analysis currently indicates a substantially higher return potential
internationally than domestically (difference of 700-900bps).
Target of $60 assumes REIT conversion; fair value $45 without conversion
We expect American Tower to be the first tower company to opt to convert to a Real
Estate Investment Trust (REIT) structure to lower taxes as its NOL balance runs out,
which we expect could occur in early 2012. The company could hold a vote at its
annual shareholder meeting in May 2011 for actual conversion in 2012 or beyond. If
the company does not convert to a REIT and has to start paying a full cash tax rate in
2012 our valuation falls to $45 per share compared to $60 assuming REIT
conversion.

Investment Risks
High multiples could contract
We estimate AMT trades at 15.8x on an EV/2011E EBITDA basis and 10.8x on an
EV/2011E revenue basis, substantially higher than the S&P 500 on both metrics. On
a relative basis, the company trades in line with peers on an EV/2011E EBITDA
basis (15.8x versus average of 15.7x). Since 2002, the tower companies have gone
from trading at less than 10x forward EBITDA to ~15x-20x today.
iDEN shutdown could be a 3%-plus headwind for the industry
We see Sprint Nextel shutting down its 20,000 iDEN cell sites in the 2012-2016 time
frame with the bulk coming in 2013-2015. We model iDEN site shutdowns dragging
our base-case revenue scenario of 68,000 sites by 4% in 2012, and we expect an
incremental ~600 bps in Sprint revenue dilution annually until 2015. Peak dilution is
expected in 2016 at 26%, a substantial negative impact if CDMA augmentations do
not partially offset the revenue loss. We believe the industry could lose about 3% or
more in gross site revenue from Sprint Nextel once its network is fully optimized.

220

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Further carrier consolidation would impact growth


While we do not expect near-term consolidation for major national wireless carriers,
we would view another major merger as a long-term negative for tower companies.
The last substantial merger that led to cell site and tower consolidations was AT&T
Wireless and Cingular, which shaved off about a percentage point in growth for the
tower companies for about five years.
Debt refinancing risk
American Tower has leverage of 3.9x and most of its debt matures in the 2014 to
2020 time frame. The largest maturity of $1.75 billion comes due in 2014, and
represents about 30% of the companys total debt. If the credit markets were closed
or refinanced debt bore substantially higher interest rates, American Tower would be
negatively affected.
International expansion could create uncertainty
The company could become more aggressive in its international efforts in Mexico,
Brazil, Chile, Peru, Colombia, India, South Africa, or Ghana. While we do not model
substantial further investment in these or other countries, American Tower could
make a major move, which could negatively affect FCF through higher capex and
thus reduce potential stock buybacks. Building up a bigger presence in any emerging
market, in which the wireless industry is still developing and country-specific risks
are very different from those in the United States, could negatively impact the
companys risk profile and multiple.

Company Description
American Tower owns 20,333 US wireless towers, 12,347 wireless towers outside
the US in Mexico, Brazil, Chile, Peru, Colombia, and India, as well as 234 US
broadcast towers and 199 Mexican broadcast towers. The company has 204 inbuilding networks in the US and two in Mexico. Also, American Tower has
announced plans to acquire up to 1,400 sites in South Africa from Cell C and a joint
venture in Ghana with MTN Group called TowerCo Ghana for 1,876 towers.
American Tower, originally a subsidiary of American Radio Systems, was spun off
when American Radio Systems was acquired by CBS in 1998. In 2005, American
Tower purchased SpectraSite Holdings for $3.8 billion. SpectraSite began 1999 with
about 100 towers and, pro forma for all acquisitions, owned 8,191 towers in the US
at the end of 2000 after buying most of them from Nextel and SBC Communications;
at the time of acquisition, it had 7,400 towers.

Table 88: Revenues by Carrier, 2009


AT&T
Sprint Nextel
Verizon
T-Mobile
Total

Total Revenue
19%
18%
15%
9%
61%

Source: Company reports and J.P. Morgan estimates.

221

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Valuation and Price Target Methodology


Our main valuation methodology is based upon our discounted cash flow (DCF)
analysis. We also provide recurring free cash flow (RFCF) and EV/EBITDA multiple
analyses, so investors can evaluate relative valuation metrics. Our year-end 2011
price target of $60 per share is based upon our DCF analysis, using a discount rate of
8.1%, a terminal growth rate of 3.0% in 2020, and a 10% private-to-public discount.
At our price target, American Tower would trade at 23.0x 2011E RFCF and 18.3x
EV/EBITDA. Risks to our target price include lower-than-expected lease-up rates,
tower builds, and augmentations, higher interest rates, wireless carrier consolidation,
and disruptions in the companys international businesses.
Table 89: DCF Sensitivity Analysis

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
60
2.0%
2.5%
3.0%
3.5%
7.0%
66
72
80
90
7.5%
59
64
70
78
8.0%
53
57
62
68
8.1%
52
56
$60.41
66
8.6%
47
50
54
58

4.0%
104
87
75
73
64

Source: Company reports and J.P. Morgan estimates.

Table 90: Valuation Multiples Analysis


Current Price Multiples @

$50.50
EV ($b)
23.7x

Current EV/Adjusted EBITDA


Current PER (adjusted)
Current Price/RFCF per share (reported)
Current Price/RFCF per share (JPM)
Current Price/RFCF per share (JPM) (net of SL)

Price Target EV/Adj. EBITDA


27.6x
Price Target PER (adjusted)
Price Target/RFCF per share (reported)
Price Target/RFCF per share (JPM)
Price Target/RFCF per share (JPM) (net of SL)

2009
20.1x
70.6x
25.3x
24.1x
24.4x

2010E
17.6x
50.5x
22.3x
19.8x
21.6x

2011E
15.2x
47.4x
20.2x
17.4x
19.3x

2012E
14.2x
40.8x
17.2x
15.3x
16.5x

2013E
13.4x
35.6x
15.0x
13.7x
14.4x

23.3x
84.4x
30.3x
28.9x
29.2x

20.5x
60.4x
26.7x
23.7x
25.8x

17.6x
56.8x
24.2x
20.8x
23.1x

16.5x
48.9x
20.6x
18.4x
19.8x

15.5x
42.6x
18.0x
16.4x
17.3x

Source: Company reports and J.P. Morgan estimates.


Note: Using 2011E Net Debt and Share Count.

Operating and Financial Outlook


We model American Tower growing its tower base by 2.1% in 2011 and 1.7% in
2012 due mostly to new builds. Acquisitions could drive upside to our estimates, but
we do not model acquisitions until they are announced and expected to close. We
estimate revenue growth of 15.2% in 2011 and 6.4% in 2012 to $2.3 billion and
$2.4 billion, respectively. Revenue per tower could reach $64,643 and $67,551 in
2011 and 2012, up 0.5% and 4.5%, respectively. The lower average revenue per
tower in 2011 is being negatively impacted by lower-revenue international towers
acquired in 2010. Our model estimates adjusted EBITDA of $1.6 billion in 2011, up
16.2% year-over-year, and $1.7 billion in 2012, up 6.8% year-over-year. We estimate
an adjusted EBITDA margin of 68.8% in 2011 and 68.6% in 2012.

222

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Earnings and Cash Flow Forecast


We model EPS of $1.07 in 2011 and $1.90 in 2012, up 11.7% and 78.0%,
respectively. EPS for 2012E benefits from a 5% assumed tax rate in REIT status
versus 38% prior to REIT status. For 2011, we estimate capex of $377 million, or
about 17.0% of leasing revenue. Our 2012 capex estimate falls to $357 million due to
slightly lower tower builds, implying 15.1% capex to leasing revenue. Our recurring
free cash flow estimate (net of straight-line impacts) of $1.0 billion in 2011 is up
9.0% year-over-year, and our 2012 estimate is 13.4% higher at $1.2 billion. On a pershare basis, we estimate RFCF of $2.61 in 2011, up 11.5%, and $3.06 in 2012, up
17.1%.

223

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 91: Valuation Summary


CAGR
2020E 11E-20E
38,279
1%
0.9%
$87,745
3%
3.0%
3,344
5%
3.9%
64
2%
-1.3%
3,407
5%
3.8%
2,403
5%
3.9%
70.5%

2007
22,807
1.8%
$63,322
9.0%
1,426
10.2%
31
-90.2%
1,457
-9.3%
980
-14.5%
67.3%

2008
23,740
4.1%
$66,660
5.3%
1,547
8.5%
46
51.8%
1,594
9.4%
1,092
11.5%
68.5%

2009
27,256
14.8%
$65,682
-1.5%
1,668
7.8%
56
19.9%
1,724
8.2%
1,181
8.1%
68.5%

2010E
34,019
24.8%
$64,343
-2.0%
1,925
15.4%
52
-7.1%
1,976
14.6%
1,344
13.8%
68.0%

2011E
34,719
2.1%
$64,643
0.5%
2,222
15.4%
56
7.4%
2,277
15.2%
1,562
16.2%
68.6%

2012E
35,319
1.7%
$67,551
4.5%
2,366
6.5%
57
2.2%
2,422
6.4%
1,667
6.8%
68.8%

2013E
35,819
1.4%
$70,455
4.3%
2,506
5.9%
60
5.9%
2,566
5.9%
1,777
6.6%
69.2%

2014E
36,219
1.1%
$73,272
4.0%
2,639
5.3%
63
5.3%
2,703
5.3%
1,881
5.8%
69.6%

2015E
36,579
1.0%
$75,690
3.3%
2,755
4.4%
66
4.4%
2,821
4.4%
1,973
4.9%
69.9%

2016E
36,919
0.9%
$77,961
3.0%
2,865
4.0%
66
-0.3%
2,931
3.9%
2,060
4.4%
70.3%

2017E
37,259
0.9%
$80,299
3.0%
2,978
4.0%
66
-0.6%
3,044
3.9%
2,141
3.9%
70.3%

2018E
37,599
0.9%
$82,708
3.0%
3,096
3.9%
65
-0.8%
3,161
3.8%
2,225
3.9%
70.4%

2019E
37,939
0.9%
$85,190
3.0%
3,218
3.9%
64
-1.0%
3,282
3.8%
2,313
3.9%
70.5%

Less: Depreciation and Amortization


EBIT
Less: Taxes on EBIT @ 5.0%
Unlevered Net Income
Add: Depreciation and Amortization
Less: Capital Expenditures
Capex as % of leasing revenue

523
457
(60)
397
523
154
10.8%

405
687
(136)
551
405
243
15.7%

415
766
(183)
584
415
250
15.0%

463
881
(188)
693
463
315
16.4%

524
1,038
(257)
781
524
377
17.0%

545
1,122
(38)
1,084
545
357
15.1%

571
1,206
(42)
1,163
571
338
13.5%

595
1,286
(46)
1,240
595
318
12.0%

614
1,359
(50)
1,309
614
315
11.4%

630
1,430
(54)
1,376
630
317
11.1%

639
1,502
(58)
1,444
639
325
10.9%

648
1,577
(62)
1,516
648
334
10.8%

656
1,656
(66)
1,590
656
343
10.7%

664
1,739
(70)
1,668
664
353
10.6%

Unlevered Free Cash Flows


Recurring Free Cash Flows
y/y % change
Recurring Free Cash Flows (net of straight-line)
y/y % change

766
816
-28.8%
796
-30.6%

713
769
-5.8%
746
-6.2%

748
852
10.8%
842
12.9%

842
1,027
20.5%
946
12.3%

929
1,139
11.0%
1,026
8.4%

1,272
1,253
10.0%
1,164
13.5%

1,396
1,356
8.2%
1,289
10.7%

1,516
1,454
7.2%
1,410
9.4%

1,608
1,542
6.0%
1,521
7.8%

1,689
1,626
5.4%
1,612
6.0%

1,758
1,704
4.8%
1,692
5.0%

1,829
1,785
4.8%
1,776
4.9%

1,903
1,869
4.7%
1,863
4.9%

1,980
1,957
4.7%
1,954
4.9%

RFCF per share - reported


RFCF per share - JPM
RFCF per share - JPM (net of straight-line)

$0.69
$1.90
$1.85

$1.61
$1.84
$1.79

$1.99
$2.09
$2.07

$2.26
$2.54
$2.34

$2.50
$2.90
$2.61

$2.93
$3.29
$3.06

$3.36
$3.68
$3.50

$3.80
$4.08
$3.95

$4.25
$4.47
$4.41

$4.67
$4.89
$4.84

$5.09
$5.31
$5.27

$5.55
$5.78
$5.75

$6.06
$6.30
$6.28

$6.62
$6.87
$6.86

11%
10%
11%

Reported EPS - Diluted


Adjusted EPS - Diluted

$0.14
$0.33

$0.83
$0.62

$0.61
$0.72

$0.95
$1.00

$1.07
$1.06

$1.90
$1.24

$2.18
$1.42

$2.47
$1.61

$2.77
$1.80

$3.08
$2.01

$3.42
$2.23

$3.80
$2.48

$4.22
$2.75

$4.69
$3.06

18%
12%

$50.50
EV ($b)
23.7x
Current EV/Adjusted EBITDA
Current PER (adjusted)
Current Price/RFCF per share (reported)
Current Price/RFCF per share (JPM)
Current Price/RFCF per share (JPM) (net of SL)

2009
20.1x
70.6x
25.3x
24.1x
24.4x

2010E
17.6x
50.5x
22.3x
19.8x
21.6x

2011E
15.2x
47.4x
20.2x
17.4x
19.3x

2012E
14.2x
40.8x
17.2x
15.3x
16.5x

2013E
13.4x
35.6x
15.0x
13.7x
14.4x

Price Target EV/Adj. EBITDA


27.6x
Price Target PER (adjusted)
Price Target/RFCF per share (reported)
Price Target/RFCF per share (JPM)
Price Target/RFCF per share (JPM) (net of SL)

23.3x
84.4x
30.3x
28.9x
29.2x

20.5x
60.4x
26.7x
23.7x
25.8x

17.6x
56.8x
24.2x
20.8x
23.1x

16.5x
48.9x
20.6x
18.4x
19.8x

15.5x
42.6x
18.0x
16.4x
17.3x

Ending sites
y/y % change
Annual Revenue per site
y/y % change
Rental and management revenue
y/y % change
Network development services
y/y % change
Revenue
y/y % change
Adjusted EBITDA
y/y % change
Adjusted EBITDA margin

Private Market Valuation Summary


NPV of EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value
PMV of operations
Plus (less): NPV of NOLs
Less: net long-term debt, end of period
PMV of equity
Current shares outstanding
Shares to be issued from warrants/options
Fully diluted shares outstanding
DCF Value per Share
Private to public discount
Price Target
Upside to target

Current Price Multiples @

2011E
12,456
(324)
(2,073)
10,059
20,099
30,157
0
(4,184)
25,974
384.1
2.9
386.9
$67.13
10%
$60.41
19.6%

Source: Company reports and J.P. Morgan estimates.

224

REIT Conversion Impact:


yes

Yes = tax rate goes to 5.0% (adjust below). WACC, taxes affected.
No = tax rate = 38.0%. WACC lower, taxes higher.

5%

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
60
2.0%
2.5%
3.0%
3.5%
7.0%
66
72
80
90
7.5%
59
64
70
78
8.0%
53
57
62
68
8.1%
52
56
$60.41
66
8.6%
47
50
54
58

4.0%
104
87
75
73
64

3%
6%

3%
-1%
9%
6%
7%

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Crown Castle International


Initiate with Overweight Rating and $55 YE11 Price Target
We are initiating coverage of Crown Castle with an Overweight rating and a yearend 2011 price target of $55. Crown Castle has substantial scale and an attractive
tower portfolio in the top 100 markets. We are confident Crown Castle will be able to
turn the positive business trends into solid revenue, EBITDA, and cash flow growth
and view shares of the company as attractive at current levels. In addition, leverage at
Crown Castle is relatively modest at about 5x, which provides us with additional
comfort and gives the company the ability to do a larger tower deal or buy back
stock. Finally, we see REIT conversion as the most likely case for Crown Castle after
NOLs are forecast to run out in 2016 and include it in our target valuation of $55 per
share. We see little downside risk to our valuation if a conversion is delayed due to
minimal net impact of tax changes.

Overweight
Crown Castle International (CCI;CCI US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ bn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

42.09
12 Jan 11
44.46 34.12
12.04
Dec
286
55.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

0.02
(0.41)
(0.13)
0.04
(0.47)
(0.34)

(0.43)A
(0.36)A
(0.49)A
0.06A
(1.22)A
(0.80)A

0.06
0.06
0.10
0.11
0.33
0.52

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on GAAP results.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Positive operating trends drive strong 2011 revenue and adjusted EBITDA
We see Crown Castle revenue growth of 7.1% in 2011 and adjusted EBITDA growth
of 8.3% due to continued strength in lease-ups, augmentations, and domestic tower
portfolio growth. There could be upside to our 2011 estimate if Sprint Nextel
augments its CDMA sites for 800 MHz capability or if Clearwire reaccelerates its
network build. T-Mobile, Leap Wireless, and MetroPCS could have an impact if any
significant network investments are made. We expect continued strength from
Verizon LTE augmentations, but note that AT&Ts are now built in to GAAP
revenue.
Solid recurring FCF (RFCF) growth of 14.8% in 2011
We estimate solid RFCF growth of 14.8% to $727 million in 2011, from
$634 million in 2010. We model an 8% organic CAGR longer term. Our year-end
2011 price target of $55 per share implies a yield of 4.6% on our 2011 RFCF
estimate of $2.54 per share.

225

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Potential operating improvement


We expect Crown Castle to expand EBITDA margins from about 62.1% in 2010 to
63.9% in 2011. Longer term, we see Crown Castle improving its margins to about
66-67%, implying a 77% incremental EBITDA margin, which is in line with SBAs
long-term margins and lower than American Towers 72% margins. Any
outperformance would represent upside to our estimates.
Crown Castles 5.2x leverage provides some flexibility, buyback potential
Crown Castle has a net leverage ratio of 5.2x versus 7.3x for SBA Communications,
but higher than American Towers 3.9x. The lower relative leverage allows the
company to potentially take on some debt for either a modest private tower portfolio
purchase or stock buyback. Large domestic tower portfolios are probably not very
attractive at current prices to Crown Castle, which leaves a stock buyback as a
potential alternative. In our view, the company has the ability to lever up by one turn
of adjusted EBITDA to buy back $1.2 billion in stock.
CCI could convert to REIT in 2017; valuation of $55/share assuming conversion
We anticipate Crown Castle opting to convert to a Real Estate Investment Trust
(REIT) structure to lower taxes as its NOL balance runs out, which we expect could
occur after 2016. Our valuation for the company is $55 assuming REIT conversion
versus $53 without conversion due to the minimal expected net tax impact.

Investment Risks
High multiples could contract
We estimate CCI trades at 14.4x on an EV/2011E EBITDA basis and 9.1x on an
EV/2011E revenue basis, substantially higher than the S&P 500 on both metrics. On
a relative basis, the company trades below its peers on an EV/2011E EBITDA basis
(14.4x versus an average of 15.7x). Since 2002, the tower companies have gone from
trading at less than 10x forward EBITDA to ~15x-20x today.
iDEN shutdown could be a 3%-plus headwind for the industry
We see Sprint Nextel shutting down its 20,000 iDEN cell sites in the 2012-2016 time
frame with the bulk coming in 2013-2015. We model iDEN site shutdowns dragging
our base-case revenue scenario of 68,000 sites by 4% in 2012, and we expect an
incremental ~600 bps in dilution annually until 2015. Peak dilution is expected in
2016 at 26%, a substantial negative impact if CDMA augmentations do not partially
offset the revenue loss. We believe the industry could lose about 3% or more in site
revenue from Sprint Nextel once its network is fully optimized.
Further carrier consolidation would reduce growth
While we do not expect near-term consolidation for major national wireless carriers,
we would view another major merger as a long-term negative for tower companies.
The last substantial merger that led to cell site and tower consolidations was AT&T
Wireless and Cingular, which shaved off about a percentage point in growth for the
tower companies for about five years.
Debt refinancing risk
Crown Castle has leverage of 5.2x and most of its debt matures in the 2014 to 2020
time frame. The company has $2.25 billion coming due in 2020, representing 34% of
the companys total debt. If the credit markets were closed or refinanced debt bore
substantially higher interest rates, Crown Castle would be negatively affected.
226

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Company Description
Crown Castle has the largest portfolio of wireless towers in the US with 22,265
towers and 529 towers in Puerto Rico. It also has 1,595 wireless towers in Australia.
In addition, the company manages 2,000 rooftop and tower sites.
Crowns primary strategy has been to acquire the towers of wireless carriers, and a
great portion of its towers are former carrier towers or towers built for carriers.
Crown Castle has acquired towers from several wireless carriers, including Bell
Atlantic, BellSouth, GTE, Powertel, and Vodafone Australia. In October 2006, the
company purchased Global Signal, which had 10,800 towers. Of Global Signals
towers, approximately 60% are under a master lease agreement with Sprint Nextel.
Table 92: Revenues by Carrier, 2009
Total Revenue
20%
22%
18%
13%
73%

AT&T
Sprint Nextel
Verizon
T-Mobile
Total
Source: Company reports and J.P. Morgan estimates.

Valuation and Price Target Methodology


Our main valuation methodology is based upon our discounted cash flow (DCF)
analysis. We also provide recurring free cash flow (RFCF) and EV/EBITDA multiple
analyses, so investors can evaluate relative valuation metrics. Our year-end 2011
price target of $55 per share is based upon our DCF analysis, using a discount rate of
8.4%, a terminal growth rate of 3.0% in 2020, and a 10% private-to-public discount.
At our price target, Crown Castle would trade at 24.6x 2011E RFCF and 17.0x
EV/EBITDA. Risks to our target price include lower-than-expected lease-up rates,
tower builds, and augmentations, higher interest rates, wireless carrier consolidation,
and disruptions in the companys Australian businesses.
Table 93: DCF Sensitivity Analysis

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
55
7.4%
7.9%
8.4%
9.4%
10.4%

2.0%
59
52
47
38
31

2.5%
65
57
50
40
33

3.0%
72
63
$54.89
43
35

3.5%
81
69
60
47
37

4.0%
92
78
67
51
40

Source: Company reports and J.P. Morgan estimates.

227

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 94: Valuation Multiples Analysis


Current Price Multiples @

$42.60
EV ($b)
17.72

Current EV/Adjusted EBITDA


Current PER (adjusted)
Current Price/RFCF per share (reported)
Current Price/RFCF per share (JPM)
Current Price/RFCF per share (JPM) (net of SL**)

21.43
Price Target EV/Adj. EBITDA
Price Target PER (adjusted)
Price Target/RFCF per share (reported)
Price Target/RFCF per share (JPM)
Price Target/RFCF per share (JPM) (net of SL**)

2009
17.5x
NM
22.7x
23.7x
24.0x

2010E
15.2x
NM
18.7x
19.3x
20.3x

2011E
14.0x
226.7x
16.6x
16.8x
19.1x

2012E
13.0x
113.6x
14.7x
14.7x
16.0x

2013E
12.3x
82.0x
13.3x
13.2x
14.0x

21.2x
NM
29.3x
30.5x
30.9x

18.4x
NM
24.1x
24.8x
26.1x

17.0x
292.1x
21.3x
21.6x
24.6x

15.8x
146.3x
18.9x
18.9x
20.6x

14.8x
105.7x
17.2x
17.0x
18.0x

Source: Company reports and J.P. Morgan estimates.


Note: Using 2011E Net Debt and Share Count.

Operating and Financial Outlook


We model Crown Castle growing its tower base by 0.7% in 2011 and 0.8% in 2012
due mostly to new builds. Acquisitions could drive upside to our estimates, but we
do not model acquisitions until they are announced and expected to close. We
estimate revenue growth of 7.1% in 2011 and 5.8% in 2012 to $2.0 billion and
$2.1 billion, respectively. Revenue per tower could reach $76,569 and $81,097 in
2011 and 2012, up 7.8% and 5.9%, respectively. Our model estimates adjusted
EBITDA of $1.3 billion in 2011, up 8.3% year-over-year, and $1.4 billion in 2012,
up 7.7% year-over-year. We estimate an adjusted EBITDA margin of 63.0% in 2011
and 64.1% in 2012.

Earnings and Cash Flow Forecast


We model EPS of $0.33 in 2011 and $0.61 in 2012. For 2011, we estimate capex of
$271 million, or about 14.8% of leasing revenue. Our 2012 capex estimate is flat at
$272 million, implying 13.9% capex to leasing revenue. Our recurring free cash flow
(net of straight-line impacts) estimate of $638 million in 2011 is up 6.1% year-overyear, and our 2012 estimate is 19.3% higher at $761 million. On a per-share basis,
we estimate RFCF of $2.23 in 2011, up 6.3%, and $2.66 in 2012, up 19.2%.

228

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 95: Valuation Summary


CAGR
2020E 11E-20E
25,835
1%
0.8%
$107,551
4%
3.0%
2,768
5%
3.8%
221
3%
3.8%
$2,989
5%
3.8%
1,988
5%
3.8%
66.5%

2007
23,846
0.7%
$54,191
6.7%
1,286
7.8%
99
8.2%
$1,385
7.8%
762
14.4%
55.0%

2008
24,079
1.0%
$58,562
8.1%
1,403
9.0%
124
25.2%
$1,527
10.2%
839
10.2%
55.0%

2009
23,957
-0.5%
$64,242
9.7%
1,543
10.0%
142
14.7%
$1,685
10.4%
1,013
20.7%
60.1%

2010E
23,875
-0.3%
$71,027
10.6%
1,698
10.0%
173
21.8%
$1,871
11.0%
1,165
15.0%
62.3%

2011E
24,035
0.7%
$76,569
7.8%
1,834
8.0%
169
-2.2%
$2,003
7.1%
1,262
8.3%
63.0%

2012E
24,235
0.8%
$81,097
5.9%
1,957
6.7%
162
-4.3%
$2,119
5.8%
1,359
7.7%
64.1%

2013E
24,435
0.8%
$85,123
5.0%
2,071
5.8%
166
2.2%
$2,237
5.6%
1,445
6.3%
64.6%

2014E
24,635
0.8%
$88,925
4.5%
2,182
5.3%
175
5.3%
$2,356
5.3%
1,531
6.0%
65.0%

2015E
24,835
0.8%
$92,368
3.9%
2,285
4.7%
183
4.7%
$2,467
4.7%
1,613
5.4%
65.4%

2016E
25,035
0.8%
$95,505
3.4%
2,381
4.2%
190
4.2%
$2,572
4.2%
1,692
4.9%
65.8%

2017E
25,235
0.8%
$98,384
3.0%
2,473
3.8%
198
3.8%
$2,670
3.8%
1,768
4.5%
66.2%

2018E
25,435
0.8%
$101,350
3.0%
2,567
3.8%
205
3.8%
$2,773
3.8%
1,847
4.5%
66.6%

2019E
25,635
0.8%
$104,405
3.0%
2,666
3.8%
213
3.8%
$2,879
3.8%
1,916
3.8%
66.6%

Less: Depreciation and Amortization


EBIT
Less: Taxes on EBIT @ 5.0%
Unlevered Net Income
Add: Depreciation and Amortization
Less: Capital Expenditures
Capex as % of leasing revenue

540
222
94
316
540
300
23.3%

526
313
104
417
526
451
32.1%

530
484
76
560
530
174
11.2%

550
615
21
636
550
220
12.9%

611
651
(6)
645
611
271
14.8%

625
734
(10)
723
625
272
13.9%

649
796
(14)
782
649
273
13.2%

683
848
(16)
831
683
283
13.0%

716
897
(19)
878
716
292
12.8%

746
946
(22)
924
746
301
12.7%

774
993
(25)
968
774
310
12.5%

804
1,043
(27)
1,015
804
319
12.4%

835
1,081
(30)
1,051
835
328
12.3%

867
1,121
(32)
1,089
867
337
12.2%

Unlevered Free Cash Flows


Recurring Free Cash Flows
y/y % change
Recurring Free Cash Flows (net of straight-line)
y/y % change

556
373
-20.3%
373
-20.3%

493
435
16.8%
435
16.8%

916
518
19.1%
510
17.3%

967
634
22.3%
602
17.9%

984
727
14.8%
638
6.1%

1,077
830
14.2%
761
19.3%

1,159
923
11.1%
872
14.5%

1,231
1,015
10.0%
983
12.8%

1,301
1,104
8.8%
1,091
11.0%

1,368
1,190
7.8%
1,184
8.5%

1,433
1,274
7.1%
1,270
7.3%

1,501
1,361
6.9%
1,360
7.1%

1,558
1,439
5.7%
1,441
5.9%

1,619
1,520
5.6%
1,524
5.8%

RFCF per share - reported


RFCF per share - JPM
RFCF per share - JPM (net of straight-line)

$1.39
$1.33
$1.33

$1.61
$1.53
$1.53

$1.87
$1.80
$1.77

$2.28
$2.21
$2.10

$2.57
$2.54
$2.23

$2.91
$2.90
$2.66

$3.20
$3.22
$3.04

$3.49
$3.54
$3.43

$3.77
$3.85
$3.80

$4.04
$4.15
$4.12

$4.30
$4.43
$4.42

$4.57
$4.74
$4.73

$4.80
$5.01
$5.01

$5.04
$5.28
$5.30

8%
8%
10%

Reported EPS - Diluted


Adjusted EPS - Diluted

($0.87)
($0.24)

($0.35)
($0.04)

($0.47)
$0.02

($1.22)
($0.27)

$0.33
$0.19

$0.61
$0.38

$0.84
$0.52

$1.02
$0.64

$1.20
$0.76

$1.38
$0.88

$1.56
$0.99

$1.75
$1.11

$1.90
$1.21

$2.05
$1.31

23%
24%

$42.60
EV ($b)
Current EV/Adjusted EBITDA
17.72
Current PER (adjusted)
Current Price/RFCF per share (reported)
Current Price/RFCF per share (JPM)
Current Price/RFCF per share (JPM) (net of SL**)

2009
17.5x
NM
22.7x
23.7x
24.0x

2010E
15.2x
NM
18.7x
19.3x
20.3x

2011E
14.0x
226.7x
16.6x
16.8x
19.1x

2012E
13.0x
113.6x
14.7x
14.7x
16.0x

2013E
12.3x
82.0x
13.3x
13.2x
14.0x

Price Target EV/Adj. EBITDA


21.43
Price Target PER (adjusted)
Price Target/RFCF per share (reported)
Price Target/RFCF per share (JPM)
Price Target/RFCF per share (JPM) (net of SL**)

21.2x
NM
29.3x
30.5x
30.9x

18.4x
NM
24.1x
24.8x
26.1x

17.0x
292.1x
21.3x
21.6x
24.6x

15.8x
146.3x
18.9x
18.9x
20.6x

14.8x
105.7x
17.2x
17.0x
18.0x

Ending sites
y/y % change
Annual Revenue per site
y/y % change
Rental and management revenue
y/y % change
Network development services
y/y % change
Revenue
y/y % change
Adjusted EBITDA
y/y % change
Adjusted EBITDA margin

Plus (less): NPV of NOLs


Less: preferred stock
Less: net long-term debt, end of period
PMV of equity
Current shares outstanding
Fully diluted shares outstanding

2011E
10,103
(125)
(1,828)
8,150
15,027
23,177
125
(316)
(5,531)
17,454
286.2
286.2

REIT Conversion Impact:


yes

DCF Value per Share


Private to public discount
Price Target
Upside to target

6%
4%
2%
6%
9%
10%

Current Price Multiples @

Private Market Valuation Summary


NPV of EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value
PMV of operations

4%
6%

$60.99
10%
$54.89
28.8%

** Estimated straight line effects.

Yes = tax rate remains 5.0% (adjust below). WACC, taxes affected.
No = tax rate goes to 38% when NOLs end. WACC lower, taxes higher.

5%

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
55
7.4%
7.9%
8.4%
9.4%
10.4%

2.0%
59
52
47
38
31

2.5%
65
57
50
40
33

3.0%
72
63
$54.89
43
35

3.5%
81
69
60
47
37

4.0%
92
78
67
51
40

Source: Company reports and J.P. Morgan estimates.

229

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

230

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

SBA Communications
Initiate with Overweight Rating and $53 YE11 Price Target
We are initiating coverage of SBA Communications with an Overweight rating and a
year-end 2011 price target of $53. The company has been a superior operating and
financial performer and, as a result, we believe warrants a premium valuation to
American Tower and Crown Castle. SBA has not signed a master lease agreement
with a major carrier, and so should report faster growth in 2H11 and 2012 than its
peers as the carrier signs LTE amendments. SBA has started to expand
internationally with a focus on markets that are fairly mature but still offer higher
growth and better return prospects than the US. While we see REIT conversion as a
possibility when the companys NOLs run out, we dont think this will happen in our
forecasting time frame (before 2020).

Overweight
SBA Communications (SBAC;SBAC US)
Company Data
Price ($)
Date Of Price
52-week Range ($)
Mkt Cap ($ mn)
Fiscal Year End
Shares O/S (mn)
Price Target ($)
Price Target End
Date

40.19
12 Jan 11
41.29 30.47
4,611.72
Dec
115
53.00
31 Dec 11

EPS ($)
Q1 (Mar)
Q2 (Jun)
Q3 (Sep)
Q4 (Dec)
FY
Bloomberg EPS FY ($)

2009A

2010E

2011E

(0.15)
(0.25)
(0.43)
(0.37)
(1.20)
(1.03)

(0.32)A
(0.72)A
(0.30)A
(0.25)A
(1.60)A
(1.52)A

(0.21)
(0.20)
(0.20)
(0.19)
(0.80)
(0.63)

Source: Company dat a, Bloomberg, J.P. Morgan estimates. Not e: EPS f igures bas ed on GAAP results.
'Bloom berg' abov e denot es Bloomberg cons ensus est imat es .

Key Investment Points


Fastest grower and most aggressive operator
We see SBA Communications as the most aggressive tower company with regard to
acquisitions, use of leverage and equity, and guidance for 2011. Managements
ability to outgrow its peers and continuously deliver strong results supports its
premium valuation, in our view. We believe managements lease-up ability and
tower acquisitions give it confidence in a guidance range, which it could very well
beat.
Strong revenue and adjusted EBITDA growth for 2011
We see SBA Communications revenue growth of 8.1% in 2011 and adjusted
EBITDA growth of 11.3% due to continued strength in lease-ups, augmentations,
and domestic tower portfolio growth. There could be upside to our 2011 estimate if
Sprint Nextel augments its CDMA sites for 800 MHz capability or if Clearwire
reaccelerates its network build. T-Mobile, Leap Wireless, and MetroPCS could have
an impact if any significant network investments are made. We expect continued
strength from Verizon and AT&T LTE augmentations, and note that unlike its peers
SBA will report AT&T augmentation growth as leases are signed.

231

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Solid recurring FCF (RFCF) growth of 18.3% in 2011


We estimate solid RFCF growth of 18.3% to $261 million in 2011, from
$220 million in 2010. We model a 9% organic CAGR longer term. Our year-end
2011 price target of $53 per share implies a yield of 4.3% on our 2011 RFCF
estimate of $2.27 per share.
International tower expansion could bring greater returns
SBA Communications sees higher potential returns from international tower builds
or purchases, despite higher risk. SBA has started to expand its international
operations in more mature markets, but still with higher growth and better return
prospects than in the US. New builds are cheaper due to lower soft and construction
costs. In addition, many markets are still developing and need new tower builds.
Collocation in some markets is starting to be adopted by more carriers and some
carriers are selling tower portfolios to focus on core strategies, like most US carriers.
Our analysis currently indicates a substantially higher return potential internationally
than domestically (difference of 700-900bps).
Potential to benefit from scarcity value as smallest tower company
As the smallest tower company in a more consolidated industry, SBA could benefit
from scarcity value, supporting a premium valuation. Although the company is still
in its growth phase, over the long term other players may find SBAs assets
strategically attractive. Historically strategic assets have changed hands for as much
as 20-25x RFCF.

Investment Risks
High multiples could contract
We estimate SBAC trades at 16.9x on an EV/2011E EBITDA basis and 10.8x on an
EV/2011E revenue basis, substantially higher than the S&P 500 on both metrics. On
a relative basis, the company trades at a premium to its peers on an EV/2011E
EBITDA basis (16.9x versus an average of 15.7x). Since 2002, the tower companies
have gone from trading at less than 10x forward EBITDA to ~15x-20x today.
iDEN shutdown could be a 3%-plus headwind for the industry
We see Sprint Nextel shutting down its 20,000 iDEN cell sites in the 2012-2016 time
frame with the bulk coming in 2013-2015. We model iDEN site shutdowns dragging
our base-case revenue scenario of 68,000 sites by 4% in 2012, and we expect an
incremental ~600 bps in dilution annually until 2015. Peak dilution is expected in
2016 at 26%, a substantial negative impact if CDMA augmentations do not partially
offset the revenue loss. We believe the industry could lose about 3% or more of site
revenue from Sprint Nextel once its network is fully optimized.
Further carrier consolidation would reduce growth
While we do not expect near-term consolidation for major national wireless carriers,
we would view another major merger as a long-term negative for tower companies.
The last substantial merger that led to cell site and tower consolidations was AT&T
Wireless and Cingular, which shaved off about a percentage point in growth for the
tower companies for about five years.

232

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Debt refinancing risk


SBA Communications has leverage of 7.3x and most of its debt matures in the 2013
to 2019 time frame. The company has $550-680 million of debt maturing in each
year starting in 2013 and ending in 2015. If the credit markets were closed or
refinanced debt bore substantially higher interest rates, SBA Communications would
be negatively affected.
International expansion could create uncertainty
The company could become more aggressive in its international efforts in Panama
and Costa Rica. We view its Canada operations as relatively stable and comparable
to those in the US. While we do not model substantial further investment in these or
other countries, SBA Communications could make a major move, which could
negatively affect FCF through higher capex and thus reduce potential stock
buybacks. Building up a bigger presence in any emerging market, in which the
wireless industry is still developing and country-specific risks are very different from
those in the United States, could negatively impact the companys risk profile and
multiple.

Company Description
SBA Communications has about 8,700 US wireless towers and is the smallest of the
public independent tower companies. SBA has stayed away from the large carrier
tower deals. The company plans to build 200-240 towers in 2010 and it has stated a
target of growing its tower builds by about 10% per annum.
The company has traded some of the benefits of economies of scale for what we
consider to be the most capital-efficient tower addition strategy of the public
independent tower companies. SBAs tower addition strategy has combined building
towers, acquiring mom-and-pop towers, and executing smaller carrier transactions.
The company consistently has led the sector in terms of organic lease-up rates and
same-tower revenue growth. SBA was founded as a site acquisition firm and has
developed a strong reputation in network development services, having participated
in the development of more than 14,000 sites. In 1997, the company decided to
expand the business to include tower ownership.
Table 96: Revenues by Carrier, 2009
AT&T
Sprint Nextel
Verizon
T-Mobile
Total

Consolidated Revenue
24%
22%
15%
14%
75%

Source: Company reports and J.P. Morgan estimates.

Valuation and Price Target Methodology


Our main valuation methodology is based upon our discounted cash flow (DCF)
analysis. We also provide recurring free cash flow (RFCF) and EV/EBITDA multiple
analyses, so investors can evaluate relative valuation metrics. Our year-end 2011
price target of $53 per share is based upon our DCF analysis, using a discount rate of
8.5%, a terminal growth rate of 3.5% in 2020, and a 10% private-to-public discount.
At our price target, SBA Communications would trade at 22.6x 2011E RFCF (net of
233

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

straight-line impacts) and 20.4x EV/EBITDA. Risks to our target price include
lower-than-expected lease-up rates, tower builds, and augmentations, higher interest
rates, wireless carrier consolidation, and disruptions in the companys international
businesses.
Table 97: DCF Sensitivity Analysis

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
53
2.5%
3.0%
3.5%
4.0%
4.5%
7.5%
58
64
73
84
98
8.0%
50
55
62
70
80
8.5%
44
48
$53.09
59
67
9.5%
34
37
40
44
49
10.5%
26
28
31
33
37
Source: Company reports and J.P. Morgan estimates.

Table 98: Valuation Multiples Comparison


Current Price Multiples @

$39.66
EV ($b)
7.20

Current EV/Adjusted EBITDA


Current Price/EFCF per share (reported)
Current Price/RFCF per share (JPM)
Current Price/RFCF per share (JPM) (net of SL)

8.73
Price Target EV/Adj. EBITDA
Price Target/EFCF per share (reported)
Price Target/RFCF per share (JPM)
Price Target/RFCF per share (JPM) (net of SL)

2009
21.3x
23.4x
23.6x
22.9x

2010E
18.7x
20.7x
20.8x
20.1x

2011E
16.8x
17.4x
17.5x
16.9x

2012E
15.3x
14.9x
14.9x
14.5x

2013E
14.1x
13.0x
13.0x
12.7x

25.8x
31.3x
31.7x
30.7x

22.7x
27.7x
27.8x
27.0x

20.4x
23.3x
23.4x
22.7x

18.6x
19.9x
20.0x
19.5x

17.1x
17.4x
17.4x
17.0x

Source: Company reports and J.P. Morgan estimates.


Note: Using 2011E Net Debt and Share Count.

Operating and Financial Outlook


We model SBA Communications growing its tower base by 2.5% in 2011 and 2.7%
in 2012 due mostly to new builds. Acquisitions could drive upside to our estimates,
but we do not model acquisitions until they are announced and expected to close. We
estimate revenue growth of 8.1% in 2011 and 7.7% in 2012 to $675 million and
$727 million, respectively. Revenue per tower could reach $65,641 and $68,923 in
2011 and 2012, up 5.0% in both years. Our model estimates adjusted EBITDA of
$429 million in 2011, up 11.3% year-over-year, and $471 million in 2012, up 9.9%
year-over-year. We estimate an adjusted EBITDA margin of 63.9% in 2011 and
65.1% in 2012.

Earnings and Cash Flow Forecast


We model negative EPS of $0.80 in 2011 and $0.61 in 2012. For 2011, we estimate
capex of $134 million, or about 22.8% of leasing revenue, and 2012 capex of
$144 million, or 22.8% of leasing revenue. Our recurring free cash flow estimate (net
of straight-line impacts) of $269 million in 2011 is up 18.0% year-over-year, and our
2012 estimate is 15.0% higher at $309 million. On a per-share basis, we estimate
RFCF of $2.34 in 2011, up 18.9%, and $2.73 in 2012, up 16.5%.

234

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 99: Valuation Summary


CAGR
11E-20E
2%

2007
6,220
12.1%
$54,680
-5.5%
322
25.6%
86
-9.0%
408
16.3%
211
31.6%
53.9%

2008
7,854
26.3%
$57,002
4.2%
396
22.9%
79
-8.1%
475
16.4%
270
27.7%
57.8%

2009
8,324
6.0%
$59,501
4.4%
477
20.6%
79
-1.1%
556
17.0%
339
25.4%
61.6%

2010E
8,830
6.1%
$62,515
5.1%
535
12.2%
89
13.2%
624
12.4%
385
13.7%
62.1%

2011E
9,050
2.5%
$65,641
5.0%
587
9.6%
88
-0.9%
675
8.1%
429
11.3%
63.9%

2012E
9,290
2.7%
$68,923
5.0%
632
7.7%
95
7.7%
727
7.7%
471
9.9%
65.1%

2013E
9,550
2.8%
$72,025
4.5%
679
7.3%
102
7.3%
780
7.3%
510
8.2%
65.7%

2014E
9,830
2.9%
$74,906
4.0%
726
7.0%
109
7.0%
835
7.0%
552
8.3%
66.4%

2015E
10,070
2.4%
$77,528
3.5%
771
6.3%
116
6.3%
887
6.3%
593
7.4%
67.1%

2016E
10,270
2.0%
$79,853
3.0%
812
5.3%
122
5.3%
934
5.3%
623
5.1%
67.0%

2017E
10,430
1.6%
$82,249
3.0%
851
4.8%
128
4.8%
979
4.8%
652
4.7%
66.9%

2018E
10,570
1.3%
$84,716
3.0%
890
4.5%
133
4.5%
1,023
4.5%
681
4.4%
66.8%

2019E
10,690
1.1%
$87,258
3.0%
928
4.3%
139
4.3%
1,067
4.3%
709
4.2%
66.8%

2020E
10,790
0.9%
$89,876
3.0%
965
4.1%
145
4.1%
1,110
4.1%
738
4.0%
66.7%

Less: Depreciation and Amortization


EBIT
Less: Taxes on EBIT @ -1.0%
Unlevered Net Income
Add: Depreciation and Amortization
Less: Capital Expenditures
Capex as % of leasing revenue

169
42
(1)
41
169
48
14.8%

211
59
(1)
58
211
56
14.2%

259
80
(0)
80
259
67
14.0%

276
109
(1)
108
276
92
17.2%

290
138
(1)
137
290
134
22.8%

309
162
(1)
161
309
144
22.8%

328
182
(0)
181
328
154
22.7%

346
205
(0)
205
346
164
22.6%

364
229
(0)
229
364
150
19.4%

378
245
0
245
378
136
16.7%

392
261
0
261
392
121
14.3%

404
277
0
277
404
115
12.9%

416
293
1
294
416
109
11.7%

427
310
1
311
427
102
10.6%

Unlevered Free Cash Flows


Recurring Free Cash Flows
y/y % change
Recurring Free Cash Flows (net of straight-line)
y/y % change

163
99
49.2%
104
50.9%

213
151
51.5%
153
47.5%

271
197
30.4%
203
32.4%

292
220
12.2%
228
12.2%

294
261
18.3%
269
18.0%

326
301
15.4%
309
15.0%

355
338
12.2%
346
11.9%

388
378
11.8%
386
11.6%

443
416
10.3%
424
10.0%

487
445
6.8%
453
6.7%

531
472
6.1%
480
6.0%

566
498
5.6%
506
5.5%

601
524
5.3%
533
5.2%

636
550
4.9%
558
4.9%

EFCF per share - reported


RFCF per share - JPM
RFCF per share - JPM (net of straight-line)

$1.14
$0.95
$0.99

$1.48
$1.37
$1.39

$1.70
$1.68
$1.73

$1.92
$1.91
$1.97

$2.28
$2.27
$2.34

$2.66
$2.66
$2.73

$3.06
$3.05
$3.12

$3.50
$3.49
$3.56

$3.95
$3.94
$4.02

$4.32
$4.31
$4.39

$4.71
$4.69
$4.77

$5.10
$5.07
$5.16

$5.52
$5.48
$5.56

$5.96
$5.91
$5.99

11%
11%
11%

Reported EPS
Adjusted EPS

($0.74)
($0.74)

($0.43)
($0.59)

($1.20)
($1.13)

($1.60)
($1.20)

($0.80)
($0.49)

($0.61)
($0.37)

($0.44)
($0.27)

($0.23)
($0.14)

($0.01)
($0.00)

$0.15
$0.09

$0.32
$0.20

$0.50
$0.30

$0.69
$0.42

$0.90
$0.55

NM
NM

$39.66
EV ($b)
Current EV/Adjusted EBITDA
7.20
Current Price/EFCF per share (reported)
Current Price/RFCF per share (JPM)
Current Price/RFCF per share (JPM) (net of SL)

2009
21.3x
23.4x
23.6x
22.9x

2010E
18.7x
20.7x
20.8x
20.1x

2011E
16.8x
17.4x
17.5x
16.9x

2012E
15.3x
14.9x
14.9x
14.5x

2013E
14.1x
13.0x
13.0x
12.7x

Price Target EV/Adj. EBITDA


8.73
Price Target/EFCF per share (reported)
Price Target/RFCF per share (JPM)
Price Target/RFCF per share (JPM) (net of SL)

25.8x
31.3x
31.7x
30.7x

22.7x
27.7x
27.8x
27.0x

20.4x
23.3x
23.4x
22.7x

18.6x
19.9x
20.0x
19.5x

17.1x
17.4x
17.4x
17.0x

Ending sites
y/y % change
Annual Revenue per site
y/y % change
Rental and management revenue
y/y % change
Network development services
y/y % change
Revenue
y/y % change
Adjusted EBITDA
y/y % change
Adjusted EBITDA margin

2011E
3,649
0
(835)
2,813
6,597
9,410

Plus (less): NPV of NOLs


Less: net long-term debt, end of period
PMV of equity

0
(2,657)
6,752

Current shares outstanding


Fully diluted shares outstanding

114.5
114.5

REIT Conversion Impact:

DCF Value per Share


Private to public discount
Price Target
Upside to target

6%
6%
6%
6%

4%
9%
10%
-3%
9%
9%
8%

Current Price Multiples @

Private Market Valuation Summary


NPV of EBITDA 12E -- 20E
NPV of taxes paid 12E -- 20E
NPV of capex 12E -- 20E
NPV of free cash flow 12E -- 20E
NPV of 2020 terminal value
PMV of operations

4%

$58.99
10%
$53.09
33.9%

no

Yes = tax rate goes to 5.0% (adjust below), WACC, taxes affected.
No = tax rate = -1.0%, WACC lower, taxes higher.

5%

Valuation Grid - DCF Value of Equity


Terminal Growth Rate (across) on 2020E FCF, Discount Rate (down)
53
2.5%
3.0%
3.5%
4.0%
4.5%
7.5%
58
64
73
84
98
8.0%
50
55
62
70
80
8.5%
44
48
$53.09
59
67
9.5%
34
37
40
44
49
10.5%
26
28
31
33
37

Source: Company reports and J.P. Morgan estimates.

235

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

236

North America Equity Research


13 January 2011

North America Equity Research


13 January 2011

Company Models

Company Models

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

237

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

238

North America Equity Research


13 January 2011

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

AT&T Model
Table 100: AT&T Wireless Operating Statistics
$ millions, except per subscriber and per share figures
Beginning subscribers
+ Gross additional subscribers
- Disconnects
+ Acquisition subscribers
Ending subscribers
% change, y-t-y

1Q09
77,009
4,856
(3,633)

2Q09
78,232
4,872
(3,504)

3Q09
79,600
5,429
(3,433)

78,232
10%

79,600
9%

81,596
9%

4Q09
81,596
6,212
(3,551)
863
85,120
11%

2009
77,009
21,369
(14,121)
863
85,120
11%

1Q10
85,120
5,213
(3,356)
86,987
11%

2Q10
86,987
4,989
(3,427)
1,581
90,130
13%

3Q10
90,130
6,252
(3,621)

4Q10E
92,761
6,243
(3,668)

92,761
14%

95,336
12%

2010E
85,120
22,697
(14,072)
1,581
95,336
12%

1Q11E
95,336
5,048
(3,828)

2Q11E
96,556
4,428
(3,778)

3Q11E
97,206
5,192
(3,962)

4Q11E
98,436
5,925
(4,025)

96,556
11%

97,206
8%

98,436
6%

100,336
5%

2011E
95,336
20,593
(15,593)
0
100,336
5%

Churn rate, monthly

1.6%

1.5%

1.4%

1.4%

1.5%

1.3%

1.3%

1.3%

1.3%

1.3%

1.3%

1.3%

1.4%

1.4%

1.3%

Net additional subscribers


% change, y-t-y

1,223
-6%

1,368
3%

1,996
1%

2,661
27%

7,248
8%

1,857
52%

1,562
14%

2,631
32%

2,575
-3%

8,635
19%

1,220
-34%

650
-58%

1,230
-53%

1,900
-26%

5,000
-42%

77,621
306,000

78,916
306,000

80,598
306,000

83,358
306,000

80,123
306,000

86,054
307,000

88,559
307,000

91,446
307,000

94,049
307,614

90,027
307,154

95,946
308,537

96,881
309,462

97,821
310,236

99,386
310,857

97,509
309,773

25.6%
1.60%

26.0%
1.79%

26.7%
2.61%

27.8%
3.48%

27.8%
2.37%

28.3%
2.42%

29.4%
2.04%

30.2%
3.43%

31.0%
3.35%

31.0%
2.81%

31.3%
1.58%

31.4%
0.84%

31.7%
1.59%

32.3%
2.44%

32.3%
1.61%

Calculated ARPU
% change, y-t-y
Reported ARPU
% change, y-t-y

$50.01
0.0%
$50.02
0.1%

$50.52
0.3%
$50.60
0.5%

$51.17
1.0%
$51.10
1.0%
58.86

$50.33
-0.5%
$50.60
-0.3%

$50.51
0.2%
$50.51
0.2%

$49.78
-0.5%
$49.81
-0.4%

$49.63
-1.8%
$50.02
-1.1%

$48.71
-3.2%
$48.71
-3.7%

$49.48
-2.0%
$49.48
-2.0%

$48.28
-3.0%
$48.28
-3.1%

$48.39
-2.5%
$48.39
-3.3%

$48.60
-2.5%
$48.60
-2.6%

$47.50
-2.5%
$47.50
-2.5%

$48.19
-2.6%
$48.19
-2.6%

Data
Annual data revenue
Average revenue per user
Data as % of ARPU

3,175
$13.64
27.3%

3,444
$14.57
28.8%

3,644
$15.05
29.4%

3,871
$15.56
30.9%

14,134
$14.70
29.1%

4,122
$15.98
32.1%

4,380
$16.61
33.5%

$49.85
-2.6%
$49.91
-2.3%
59.93
1.8%
4,754
$17.35
34.8%

4,989
$17.68
36.3%

18,245
$16.89
34.1%

5,253
$18.25
37.8%

5,457
$18.78
38.8%

5,677
$19.34
39.8%

5,778
$19.38
40.8%

22,165
$18.94
39.3%

Postpaid
Beginning subscribers
+ Gross additional subscribers
- Disconnects
+ Acquisition subscribers
Ending subscribers
% change, y-t-y
% of Total subs

59,653
2,970
(2,073)
0
60,535
9%
77.4%

60,535
3,089
(1,961)
0
61,647
9%
77.4%

61,647
3,464
(2,131)
0
62,961
8%
77.2%

62,961
3,042
(2,201)
825
64,627
8%
75.9%

59,653
12,565
(8,366)
825
64,627
8%
75.9%

64,627
2,594
(2,082)
0
65,108
8%
74.8%

65,108
2,497
(2,001)
1,366
66,970
9%
74.3%

66,970
3,048
(2,303)
0
67,688
8%
73.0%

67,688
2,869
(2,344)
0
68,213
6%
71.6%

64,627
11,008
(8,730)
1,366
68,213
6%
71.6%

68,213
2,558
(2,658)
0
68,113
5%
70.5%

68,113
2,554
(2,854)
0
67,813
1%
69.8%

67,813
2,747
(2,547)
0
68,013
0%
69.1%

68,013
2,754
(2,454)
0
68,313
0%
68.1%

68,213
10,613
(10,513)
0
68,313
0%
68.1%

Average subs

Average subscribers
Pops
Ending penetration
Annual penetration gain

60,094

61,091

62,304

63,794

62,140

64,868

66,039

67,329

67,951

66,420

68,163

67,963

67,913

68,163

68,263

Churn rate, monthly

1.15%

1.07%

1.14%

1.15%

1.12%

1.07%

1.01%

1.14%

1.15%

1.10%

1.30%

1.40%

1.25%

1.20%

1.28%

Net additional subscribers


% change, y-t-y
% of total net adds

897
21%
73.3%

1,128
30%
82.5%

1,333
-19%
66.8%

841
-34%
31.6%

4,199
-7%
57.9%

512
-43%
27.6%

496
-56%
31.8%

745
-44%
28.3%

525
-38%
20.4%

2,278
-46%
26.4%

(100)
NM

(300)
NM

200
-73%
16.3%

300
-43%
15.8%

100
-96%
2.0%

ARPU
% change, y-t-y
Data ARPU
% change, y-t-y
Voice ARPU
% change, y-t-y

$59.55
2.2%
$16.48
26.8%
$43.98
-4.1%

$60.56
2.4%
$17.72
26.0%
$43.77
-4.3%

$61.61
3.9%
$18.43
25.4%
$44.20
-2.8%

$61.53
2.6%
$19.16
17.5%
$43.31
-2.4%

$60.60
2.5%

$62.63
3.4%
$21.07
18.9%
$42.51
-2.9%

$62.84
2.0%
$22.02
19.5%
$41.74
-5.6%

$62.61
1.8%
$23.00
20.0%
$40.49
-6.5%

$62.52
3.2%

$63.00
1.8%

$63.85
1.9%

$64.77
3.1%

$63.92
2.1%

$64.04
2.4%

NM

-100.0%

-100.0%

-100.0%

-100.0%

NM

NM

$61.89
3.9%
$20.13
22.1%
$42.73
-2.8%

NM

-100.0%

-100.0%

-100.0%

-100.0%

NM

Prepaid
Beginning subscribers
+ Gross additional subscribers
- Disconnects
+ Acquisition subscribers
Ending subscribers
% change, y-t-y
% of Total subs

6,116
1,232
(1,387)
0
5,961
-2%
7.6%

5,970
950
(1,362)
0
5,558
-9%
7.0%

5,562
904
(1,080)
0
5,386
-12%
6.6%

5,386
1,079
(1,137)
22
5,350
-12%
6.3%

6,106
4,165
(4,966)
22
5,350
-12%
6.3%

5,353
1,076
(1,052)
0
5,377
-10%
6.2%

5,377
1,489
(1,189)
204
5,881
6%
6.5%

5,888
1,417
(1,096)
0
6,209
15%
6.7%

6,209
1,546
(1,096)
0
6,659
24%
7.0%

5,350
5,528
(4,433)
204
6,659
24%
7.0%

6,659
1,344
(1,044)
0
6,959
29%
7.2%

6,959
943
(793)
0
7,109
21%
7.3%

7,109
1,429
(1,279)
0
7,259
17%
7.4%

7,259
1,728
(1,428)
0
7,559
14%
7.5%

6,659
5,444
(4,544)
0
7,559
14%
7.5%

Average subs

6,039

5,764

5,474

5,368

5,728

5,365

5,629

6,049

6,434

6,005

6,809

7,034

7,184

7,409

7,109

7.7%

7.9%

6.6%

7.1%

7.2%

6.5%

7.0%

6.0%

5.7%

6.2%

5.1%

3.8%

5.9%

6.4%

5.3%

Net additional subscribers


% change, y-t-y
% of total net adds

(155)
NM
-12.7%

(412)
NM
-30.1%

(176)
-389%
-8.8%

(58)
-152%
-2.2%

(801)
NM
-11.1%

24
NM
1.3%

300
NM
19.2%

321
NM
12.2%

450
NM
17.5%

1,095
NM
12.7%

300
1150%
24.6%

150
-50%
23.1%

150
-53%
12.2%

300
-33%
15.8%

900
-18%
18.0%

ARPU
% change, y-t-y

$27.50
-2.3%

$27.50
3.3%

$27.50
-5.2%

$27.50
-5.2%

$27.50
-2.5%

$27.50
0.0%

$27.50
0.0%

$27.50
0.0%

$27.50
0.0%

$27.50
0.0%

$27.00
-1.8%

$27.00
-1.8%

$27.00
-1.8%

$27.00
-1.8%

$27.00
-1.8%

Reseller
Beginning subscribers
+ Gross additional subscribers
- Disconnects
+ Acquisition subscribers
Ending subscribers
% change, y-t-y
% of Total subs

8,589
468
(131)
0
8,931
15%
11.4%

8,931
485
(137)
0
9,286
18%
11.7%

9,286
778
(144)
0
9,934
24%
12.2%

9,934
637
(153)
21
10,439
22%
12.3%

8,589
2,368
(565)
21
10,439
22%
12.3%

10,439
412
(143)
0
10,717
20%
12.3%

10,717
14
(144)
10
10,597
14%
11.8%

10,597
519
(113)
0
11,021
11%
11.9%

11,021
401
(101)
0
11,321
8%
11.9%

10,439
1,346
(501)
10
11,321
8%
11.9%

11,321
369
(69)
0
11,621
8%
12.0%

11,621
170
(70)
0
11,721
11%
12.1%

11,721
271
(71)
0
11,921
8%
12.1%

11,921
372
(72)
0
12,221
8%
12.2%

11,321
1,182
(282)
0
12,221
8%
12.2%

Average subs

8,760

9,109

9,610

10,187

9,514

10,578

10,657

10,809

11,171

10,880

11,471

11,671

11,821

12,071

11,771

0.5%

0.5%

0.5%

0.5%

0.5%

0.5%

0.5%

0.4%

0.3%

0.4%

0.2%

0.2%

0.2%

0.2%

0.2%

Net additional subscribers


% change, y-t-y
% of total net adds

337
20%
27.6%

348
219%
25.4%

634
373%
31.8%

484
-16%
18.2%

1,803
64%
24.9%

269
-20%
14.5% NM

(130)
NM

406
-36%
15.4%

300
-38%
11.7%

845
-53%
9.8%

300
12%
24.6%

100
NM
15.4%

200
-51%
16.3%

300
0%
15.8%

900
7%
18.0%

ARPU
% change, y-t-y
Connected Devices
Beginning subscribers
+ Gross additional subscribers
- Disconnects
+ Acquisition subscribers
Ending subscribers
% change, y-t-y
% of Total subs

$4.00
-14.0%

$3.90
-13.3%

$3.80
-32.2%

$3.70
-32.8%

$3.85
-24.0%

$4.00
0.0%

$3.90
0.0%

$3.80
0.0%

$3.80
2.7%

$3.88
0.6%

$3.80
-5.0%

$3.70
-5.1%

$3.60
-5.3%

$3.60
-5.3%

$3.68
-5.2%

2,661
185
(41)
0
2,805
55%
3.6%

2,805
348
(44)
0
3,109
44%
3.9%

3,109
283
(48)
0
3,315
38%
4.1%

3,315
1,454
(60)
(6)
4,704
77%
5.5%

2,661
2,271
(194)
(6)
4,704
77%
5.5%

4,704
1,131
(79)
0
5,785
106%
6.7%

5,785
990
(94)
1
6,682
115%
7.4%

6,682
1,268
(109)
0
7,843
137%
8.5%

7,843
1,427
(127)
0
9,143
94%
9.6%

4,704
4,816
(409)
1
9,143
94%
9.6%

9,143
777
(57)
0
9,863
70%
10.2%

9,863
761
(61)
0
10,563
58%
10.9%

10,563
745
(65)
0
11,243
43%
11.4%

11,243
1,070
(70)
0
12,243
34%
12.2%

9,143
3,354
(254)
0
12,243
34%
12.2%

2,733

2,957

3,212

4,010

3,683

5,245

6,234

7,263

8,493

6,924

9,503

10,213

10,903

11,743

10,693

0.5%

0.5%

0.5%

0.5%

0.4%

0.5%

0.5%

0.5%

0.5%

0.5%

0.2%

0.2%

0.2%

0.2%

0.2%

Net additional subscribers


% change, y-t-y
% of total net adds

144
-8%
11.8%

304
-12%
22.2%

235
-2%
11.8%

1,394
430%
52.4%

2,077
107%
28.7%

1,052
631%
56.7%

896
195%
57.4%

1,159
393%
44.1%

1,300
-7%
50.5%

4,407
112%
51.0%

720
-32%
59.0%

700
-22%
107.7%

680
-41%
55.3%

1,000
-23%
52.6%

3,100
-30%
62.0%

ARPU
% change, y-t-y

$3.00
0.0%

$3.00
0.0%

$3.00
0.0%

$3.00
0.0%

$3.00
0.0%

$3.00
0.0%

$3.10
3.3%

$3.10
3.3%

$3.10
3.3%

$3.08
2.5%

$3.20
6.7%

$3.30
6.5%

$3.30
6.5%

$3.30
6.5%

$3.28
6.5%

Churn rate, monthly

Churn rate, monthly

Average subs
Churn rate, monthly

NM

NM

NM

Source: Company reports and J.P. Morgan estimates.

239

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 101: AT&T Wireless Cost Metrics and Income Statement


$ millions, except per subscriber and per share figures
1Q09

2Q09

3Q09

4Q09

2009

1Q10

2Q10

3Q10

4Q10E

2010E

1Q11E

2Q11E

3Q11E

4Q11E

2011E

691

703

687

670

660

661

662

662

662

662

662

4,202
4,567
8,769
0%
-14%

4,039
5,193
9,232
5%
5%

4,368
5,607
9,976
-7%
8%

4,121
5,614
9,735
-4%
-2%

3,671
4,735
8,406
-4%
-14%

3,986
5,151
9,137
-1%
9%

4,465
6,800
11,265
13%
23%

4,415
5,776
10,191
5%
-10%

661
108.0%
16,536
22,462
38,999
3%
3%

662

Gross adds - direct


Postpaid upgrades
Total retail unit sales
% change, y-t-y
% change, sequential

688
103.3%
16,730
20,981
37,711
-2%
-2%

3,902
5,112
9,015
7%
-12%

3,497
5,437
8,934
-2%
-1%

4,175
6,452
10,627
-6%
19%

4,482
5,794
10,276
1%
-3%

662
108.4%
16,057
22,795
38,852
0%
0%

Upgrade formula
% of base upgraded
% base w new handsets

1.09x
7.6%
11.3%

1.29x
8.5%
11.7%

1.28x
9.0%
12.4%

1.36x
8.8%
11.7%

1.25x
33.8%
47.1%

1.29x
7.3%
9.8%

1.29x
7.8%
10.3%

1.52x
10.1%
12.3%

1.31x
8.5%
10.8%

1.36x
33.8%
43.3%

1.31x
7.5%
9.4%

1.55x
8.0%
9.2%

1.55x
9.5%
10.9%

1.29x
8.5%
10.3%

1.42x
33.4%
39.8%

Eq. cost / postpaid unit


Eq. rev. / postpaid unit
Handset subsidy in postpaid

$349
$158
$191

$334
$152
$182

$330
$138
$192

$315
$142
$173

$331
$147
$184

$321
$143
$178

$335
$138
$196

$412
$153
$259

$365
$130
$236

$362
$141
$221

$364
$129
$235

$347
$123
$224

$358
$127
$231

$361
$132
$228

$357
$128
$230

20.5%
28.3%
$10.27

20.9%
33.9%
$10.58

17.9%
38.3%
$9.16

16.6%
29.4%
$8.34

18.9%
32.6%
$9.56

18.5%
-0.8%
$9.22

19.7%
7.3%
$9.76

16.7%
5.2%
$8.32

20.0%
4.0%
$9.74

18.7%
16.6%
$9.26

19.5%
31.5%
$9.41

19.0%
8.9%
$9.19

19.0%
72.5%
$9.23

19.0%
-14.0%
$9.02

19.1%
26.5%
$9.22

Eq. rev./gross add (handset price)


Eq. cost/gross add (handset cost)
Handset subsidy
Cost of eq. as % of eq. revenue

$136
$300
$164
221%

$137
$300
$163
219%

$126
$300
$174
238%

$127
$280
$153
221%

$131
$295
$164
225%

$125
$280
$155
225%

$116
$280
$164
242%

$134
$360
$226
269%

$110
$310
$200
282%

$121
$311
$190
256%

$110
$310
$200
282%

$110
$310
$200
282%

$110
$310
$200
282%

$110
$300
$190
273%

$110
$307
$197
279%

Customer acquisition cost (per gross add)


Marketing/gross add
Marketing as % of service revenue

$420
$256
9%

$440
$277
9%

$420
$246
8%

$440
$287
12%

$431
$267
9%

$388
$233
8%

$380
$216
8%

$450
$224
8%

$400
$200
8%

$407
$217
8%

$410
$210
7%

$410
$210
6%

$400
$200
6%

$400
$210
8%

$405
$207
7%

G&A cost/subscriber/month
G&A as % of service revenue

$8.43
15%

$9.26
16%

$10.77
19%

$10.44
18%

$9.76
17%

$10.30
18%

$10.00
18%

$10.00
18%

$9.65
17%

$9.99
18%

$9.60
18%

$9.60
17%

$9.60
17%

$9.60
18%

$9.60
18%

CCPU (Cash cost per user)


% change, y-t-y

$18.70
0%

$19.84
1%

$19.93
4%

$18.78
1%

$19.32
8%

$19.52
4%

$19.76
0%

$18.32
-8%

$19.39
3%

$19.25
0%

$19.01
-3%

$18.79
-5%

$18.83
3%

$18.62
-4%

$18.82
-2%

879
9%
7.5%

1,340
-13%
11.2%

1,495
-9%
12.1%

2,193
17%
17.4%

5,907
1%
12.2%

1,218
39%
9.5%

2,174
62%
16.5%

2,518
68%
18.4%

2,611
19%
19.0%

8,521
44%
16%

2,154
77%
15.5%

2,180
0%
15.5%

2,211
-12%
15.5%

2,195
-16%
15.5%

8,740
3%
16%

1Q09
11,646
1,192
12,838
9%

2Q09
11,960
1,262
13,222
10%

3Q09
12,372
1,255
13,627
8%

4Q09
12,585
1,232
13,817
8%

2009
48,563
4,941
53,504
9%

1Q10
12,850
1,047
13,897
8%

2Q10
13,186
1,056
14,242
8%

3Q10
13,675
1,505
15,180
11%

4Q10E
13,745
1,121
14,866
8%

2010E
53,456
4,729
58,185
9%

1Q11E
13,897
992
14,889
7%

2Q11E
14,065
983
15,047
6%

3Q11E
14,263
1,169
15,432
2%

4Q11E
14,162
1,130
15,292
3%

2011E
56,386
4,274
60,660
4%

Cost of service
General & administrative
Pre-marketing cash flow
Pre-marketing cash flow margin
% change, y-t-y

2,390
1,742
7,514
64.5%
10%

2,505
1,940
7,515
62.8%
9%

2,215
2,294
7,863
63.6%
9%

2,085
2,293
8,207
65.2%
9%

9,195
8,268
31,099
64.0%
9%

2,381
2,332
8,137
63.3%
8%

2,594
2,337
8,255
62.6%
10%

2,284
2,419
8,972
65.6%
14%

2,749
2,399
8,596
62.5%
5%

10,007
9,488
33,961
63.5%
9%

2,710
2,433
8,755
63.0%
8%

2,672
2,454
8,938
63.6%
8%

2,710
2,477
9,076
63.6%
1%

2,691
2,515
8,956
63.2%
4%

10,783
9,878
35,725
63.4%
5%

Cost of equipment
Marketing costs
Total operating expenses
EBITDA (operating cash flow)
EBITDA margin
% change, y-t-y

2,631
1,123
7,886
4,952
42.5%
7%

2,769
1,214
8,428
4,794
40.1%
2%

2,993
1,143
8,645
4,982
40.3%
26%

2,726
1,598
8,701
5,116
40.7%
18%

11,119
5,078
33,660
19,844
40.9%
13%

2,354
1,117
8,183
5,714
44.5%
15%

2,558
1,073
8,562
5,680
43.1%
18%

4,055
1,282
10,040
5,140
37.6%
3%

3,159
1,168
9,476
5,390
39.2%
5%

12,127
4,639
36,261
21,924
41.0%
10%

2,795
983
8,920
5,969
42.9%
4%

2,770
894
8,790
6,257
44.5%
10%

3,294
984
9,465
5,966
41.8%
16%

3,083
1,166
9,454
5,838
41.2%
8%

11,941
4,027
36,630
24,030
42.6%
10%

Depreciation and amortization


Operating income
Operating income margin
% change, y-t-y

1,499
3,453
26.9%
-6%

1,504
3,290
24.9%
-11%

1,490
3,492
25.6%
17%

1,550
3,566
25.8%
8%

6,043
13,801
25.8%
1%

1,558
4,156
29.9%
20%

1,578
4,102
28.8%
25%

1,640
3,500
23.1%
0%

1,640
3,750
25.2%
5%

6,416
15,508
26.7%
12%

1,640
4,329
29.1%
4%

1,640
4,617
30.7%
13%

1,640
4,326
28.0%
24%

1,640
4,198
27.5%
12%

6,560
17,470
28.8%
13%

Equity in net income(loss) of affiliates


Segment income

3,453

3,290

3,492

9
3,575

9
13,810

13
4,169

7
4,109

(6)
3,494

(6)
3,744

8
15,516

(6)
4,323

(6)
4,611

(6)
4,320

(6)
4,192

(24)
17,446

Revenue and minutes of use


Total minutes of use

Operating expenses
Cost of service as % of service revenue
Incremental cost of service %
Cost of service/subscriber/month
Cost of service/minute of use

Capital Expenditures (wireless)


% change, y-t-y
% of service revenue

Service revenue
Equipment revenue
Total revenue
% change, y-t-y

Source: Company reports and J.P. Morgan estimates.

240

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 102: AT&T Wireline Operating Statistics


$ millions, except per subscriber and per share figures
1Q09
29,969
-12.3%
-2.8%
(869)
26,780
-11.7%
-2.5%
(699)
394
9750%
76%
170

2Q09
29,048
-12.1%
-3.1%
(921)
26,008
-11.4%
-2.9%
(772)
570
1800%
45%
176

3Q09
28,098
-11.8%
-3.3%
(950)
25,205
-11.0%
-3.1%
(803)
735
607%
29%
165

4Q09
27,332
-11.4%
-2.7%
(766)
24,562
-10.6%
-2.6%
(643)
954
326%
30%
219

2009
27,332
-11.4%

2Q10
25,780
-11.3%
-3.2%
(853)
23,264
-10.6%
-3.0%
(730)
1,328
133%
16%
183

3Q10
24,908
-11.4%
-3.4%
(872)
22,509
-10.7%
-3.2%
(755)
1,494
103%
13%
166

4Q10E
24,258
-11.2%
-2.6%
(650)
21,970
-10.6%
-2.4%
(539)
1,638
72%
10%
144

2010E
24,258
-11.2%

(3,506)
24,562
-10.6%
-10.6%
(2,917)
954
326%
326%
730

1Q10
26,633
-11.1%
-2.6%
(699)
23,993
-10.4%
-2.3%
(569)
1,145
191%
20%
191

(3,074)
21,970
-10.6%
-10.6%
(2,592)
1,638
72%
72%
684

1Q11E
23,758
-10.8%
-2.1%
(500)
21,545
-10.2%
-1.9%
(425)
1,785
56%
9%
147

2Q11E
23,258
-9.8%
-2.1%
(500)
21,120
-9.2%
-2.0%
(425)
1,932
45%
8%
147

3Q11E
22,758
-8.6%
-2.1%
(500)
20,695
-8.1%
-2.0%
(425)
2,079
39%
8%
147

4Q11E
22,271
-8.2%
-2.1%
(488)
20,283
-7.7%
-2.0%
(413)
2,226
36%
7%
147

(1,988)
20,283
-7.7%
-7.7%
(1,688)
2,226
36%
36%
588

3,189
-18%
-5%
(170)

3,040
-18%
-5%
(149)

2,893
-18%
-5%
(147)

2,770
-18%
-4%
(123)

2,770
-18%
-18%
(589)

2,640
-17%
-5%
(130)

2,517
-17%
-5%
(124)

2,399
-17%
-5%
(117)

2,288
-17%
-5%
(111)

2,288
-17%
-17%
(482)

2,213
-16%
-3%
(75)

2,138
-15%
-3%
(75)

2,063
-14%
-4%
(75)

1,988
-13%
-4%
(75)

1,988
-13%
-13%
(300)

13,344
6.4%
2.9%
3,303

13,454
6.9%
0.8%
110

13,550
6.4%
0.7%
96

13,717
5.7%
1.2%
151

13,717
5.7%
5.7%
3,660

13,989
4.8%
2.0%
255

13,925
3.5%
-0.5%
(92)

14,092
4.0%
1.2%
148

14,217
3.6%
0.9%
125

14,217
3.6%
3.6%
436

14,367
2.7%
1.1%
150

14,467
3.9%
0.7%
100

14,542
3.2%
0.5%
75

14,667
3.2%
0.9%
125

14,667
3.2%
3.2%
450

Satellite connections
% change, y-t-y
% change, q-t-q
Net change

2,205
-1%
1%
15

2,210
-1%
0%
5

2,195
1%
-1%
(15)

2,174
-1%
-1%
5

2,174
-1%
-1%
10

2,127
-4%
-2%
(47)

2,053
-7%
-3%
(74)

1,994
-9%
-3%
(59)

1,935
-11%
-3%
(59)

1,935
-11%
-11%
(239)

1,888
-11%
-2%
(47)

1,841
-10%
-3%
(47)

1,793
-10%
-3%
(47)

1,746
-10%
-3%
(47)

1,746
-10%
-10%
(189)

U-verse video connections


% change, y-t-y
% change, q-t-q
Net change
Average U-verse subs

1,329
251%
27%
284
1,187

1,577
187%
19%
248
248

1,816
133%
15%
239
239

2,064
98%
14%
250
1,940

2,064
98%
98%
1,019
1,555

2,295
73%
11%
231
2,180

2,504
59%
9%
209
2,400

2,739
51%
9%
235
2,622

2,944
43%
7%
205
2,842

2,944
43%
43%
880
2,504

3,154
37%
7%
210
3,049

3,364
34%
7%
210
3,259

3,574
30%
6%
210
3,469

3,784
29%
6%
210
3,679

3,784
29%
29%
840
3,364

U-verse Homes Passed (TV)


Net New Homes Passed
U-verse TV Penetration

18,350
1,350
7.2%

19,700
1,350
8.0%

21,050
1,350
8.6%

22,400
1,350
9.2%

22,400
5,400
9.2%

23,750
1,350
9.7%

25,100
1,350
10.0%

26,000
900
10.5%

27,350
1,350
10.8%

27,350
4,950
10.8%

28,150
800
11.2%

28,950
800
11.6%

29,750
800
12.0%

30,550
800
12.4%

30,550
3,200
12.4%

Total consumer revenue connections


% change, y-t-y
% change, q-t-q
Net change

46,847
-5%
0%
(198)

46,289
-4%
-1%
(558)

45,659
-4%
-1%
(630)

45,287
-4%
-1%
(372)

45,287
-4%
-4%
(1,758)

45,044
-4%
-1%
(243)

44,262
-4%
-2%
(782)

43,733
-4%
-1%
(529)

43,354
-4%
-1%
(379)

43,354
-4%
-4%
(1,933)

43,167
-4%
0%
(187)

42,930
-3%
-1%
(237)

42,667
-2%
-1%
(262)

42,468
-2%
0%
(200)

42,468
-2%
-2%
(886)

Switched access lines


Total homes passed
% change, y-t-y

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,000
0.0%

52,520
1.0%

52,520
1.0%

52,520
1.0%

52,520
1.0%

52,520
1.0%

Retail consumer primary


% change, y-t-y
% change, q-t-q
Net change

26,386
-13%
NM
(869)

25,438
-13%
-4%
(948)

24,470
-13%
-4%
(968)

23,608
-13%
-4%
(862)

23,608
-13%
-13%
(3,647)

22,848
-13%
-3%
(760)

21,936
-14%
-4%
(913)

21,015
-14%
-4%
(921)

20,333
-14%
-3%
(682)

20,333
-14%
-14%
(3,275)

19,761
-14%
-3%
(572)

19,189
-13%
-3%
(572)

18,617
-11%
-3%
(572)

18,057
-11%
-3%
(560)

18,057
-11%
-11%
(2,276)

3,194
-17%
-5%
(169)

3,025
-18%
-5%
(169)

2,856
-19%
-6%
(169)

2,687
-20%
-6%
(169)

2,687
-20%
-20%
(676)

2,518
-21%
-6%
(169)

2,349
-22%
-7%
(169)

2,180
-24%
-7%
(169)

2,011
-25%
-8%
(169)

2,011
-25%
-25%
(676)

1,867
-26%
-7%
(144)

1,724
-27%
-8%
(144)

1,580
-28%
-8%
(144)

1,436
-29%
-9%
(144)

1,436
-29%
-29%
(575)

Retail business
% change, y-t-y
% change, q-t-q
Net change

21,364
-6%
-2%
(462)

20,864
-7%
-2%
(500)

20,464
-8%
-2%
(400)

20,106
-8%
-2%
(358)

20,106
-8%
-8%
(1,720)

19,854
-7%
-1%
(252)

19,465
-7%
-2%
(389)

19,089
-7%
-2%
(376)

18,763
-7%
-2%
(326)

18,763
-7%
-7%
(1,343)

18,462
-7%
-2%
(301)

18,186
-7%
-1%
(276)

17,935
-6%
-1%
(251)

17,835
-5%
-1%
(100)

17,835
-5%
-5%
(928)

Total retail access lines


% change, y-t-y
% change, q-t-q
Net change

50,944
-10%
-3%
(1,500)

49,327
-11%
-3%
(1,617)

47,790
-11%
-3%
(1,537)

46,401
-12%
-3%
(1,389)

46,401
-12%
-12%
(6,043)

45,220
-11.2%
-3%
(1,181)

43,750
-11%
-3%
(1,471)

42,284
-12%
-3%
(1,466)

41,107
-11%
-3%
(1,177)

41,107
-11%
-11%
(5,294)

40,090
-11%
-2%
(1,017)

39,098
-11%
-2%
(992)

38,132
-10%
-2%
(967)

37,329
-9%
-2%
(803)

37,329
-9%
-9%
(3,778)

Wholesale access lines


% change, y-t-y
% change, q-t-q
Net change
Total switched access lines
% change, y-t-y
Net change
% change, y-t-y

3,053
-11%
-4%
(117)
53,992
-11%
(1,622)

2,961
-9%
-3%
(92)
52,379
-11%
(1,613)

2,936
-8%
-1%
(25)
50,833
-11%
(1,546)

2,908
-8%
-1%
(28)
49,392
-11%
(1,441)

2,908
-8%
-8%
(262)
49,309
-11%
(6,305)

2,741
-10%
-6%
(167)
48,083
-11%
(1,309)
48,083

2,641
-11%
-4%
(100)
46,558
-11%
(1,525)

2,605
-11%
-1%
(36)
45,108
-11%
(1,450)

2,515
-14%
-3%
(90)
43,622
-12%
(1,486)

2,515
-14%
-14%
(393)
43,622
-12%
(5,687)

2,448
-11%
-3%
(68)
42,538
-12%
(1,084)

2,380
-10%
-3%
(68)
41,478
-11%
(1,059)

2,313
-11%
-3%
(68)
40,444
-10%
(1,034)

2,245
-11%
-3%
(68)
39,574
-9%
(871)

2,245
-11%
-11%
(270)
39,574
-9%
(4,048)

Total wired broadband connections


Penetration of homes passed
% change, y-t-y
% change, q-t-q
Net change

15,436
30%
5%
2%
359

15,548
30%
6%
1%
112

15,638
30%
5%
1%
90

15,789
30%
5%
1%
151

15,789
30%
5%
5%
712

16,044
31%
4%
2%
255

15,952
31%
3%
-1%
(92)

16,100
31%
3%
1%
148

16,225
31%
3%
1%
125

16,225
31%
3%
3%
436

16,375
31%
2%
1%
150

16,475
31%
3%
1%
100

16,550
32%
3%
0%
75

16,675
32%
3%
1%
125

16,675
32%
3%
3%
450

14,220
-1%
1%
101
14,169
12,697
1,472
54%

14,107
-1%
-1%
(113)
14,163
14,163
0
55%

13,980
-1%
-1%
(127)
14,043
14,043
0
57%

13,905
-2%
-1%
(74)
13,943
13,943
0
59%

13,905
-2%
-2%
(213)
14,012
12,540
1,472
59%

13,951
-2%
0%
45
13,928
13,928
0
61%

13,668
-3%
-2%
(282)
13,809
13,809
0
62%

13,603
-3%
0%
(66)
13,636
13,636
0
65%

13,541
-3%
0%
(62)
13,572
13,572
0
67%

13,541
-3%
-3%
(364)
13,723
13,723
0
67%

13,500
-3%
0%
(41)
13,521
13,521
0
68%

13,409
-2%
-1%
(91)
13,455
13,455
0
70%

13,293
-2%
-1%
(116)
13,351
13,351
0
71%

13,227
-2%
0%
(66)
13,260
13,260
0
73%

13,227
-2%
-2%
(314)
13,384
13,384
0
73%

14,220
0

14,107
0

13,980
0

13,905
0

13,905
0

13,951
0

13,668
0

13,603
0

13,541
0

13,541
0

13,500
0

13,409
0

13,293
0

13,227
0

13,227
0

1,196
251%
27%
256
1,068

1,419
187%
19%
223
1,308

1,634
133%
15%
215
1,527

1,858
98%
14%
223
1,746

1,858 0
98%
98%
917 0
1,399

2,066
73%
11%
208
1,962

2,254
59%
9%
188
2,160

2,465
51%
9%
212
2,359

2,650
43%
7%
185
2,557

2,650 0
43%
43%
792 0
2,254

2,839
37%
7%
189
2,744

3,028
34%
7%
189
2,933

3,217
30%
6%
189
3,122

3,406
29%
6%
189
3,311

3,406
29%
29%
756
3,028

20
67%
11%
2

22
57%
10%
2

24
50%
9%
2

26
44%
8%
2

26
44%
44%
8

28
40%
8%
2

30
36%
7%
2

32
33%
7%
2

34
31%
6%
2

34
31%
31%
8

36
29%
6%
2

38
27%
6%
2

40
25%
5%
2

42
24%
5%
2

42
24%
24%
8

Retail consumer access lines


% change, y-t-y
% change, q-t-q
Net change
Retail consumer primary lines (switched/VOIP)
% change, y-t-y
% change, q-t-q
Net change
VOIP lines (U-Verse voice)
% change, y-t-y
% change, q-t-q
Net change
Retail consumer secondary lines (switched/VOIP)
% change, y-t-y
% change, q-t-q
Net change
Consumer broadband connections
% change, y-t-y
% change, q-t-q
Net change

Retail consumer additional


% change, y-t-y
% change, q-t-q
Net change

DSL subs
% change, y-t-y
% change, q-t-q
Net change
Average DSL subs
Average consumer DSL subs
Average business DSL subs
Consumer DSL penetration of primary lines
Consumer DSL subs
Business DSL subs
U-verse broadband connections
% change, y-t-y
% change, q-t-q
Net change
Average U-verse broadband subs
Satellite broadband subs
% change, y-t-y
% change, q-t-q
Net change

2011E
22,271
-8.2%

Source: Company reports and J.P. Morgan estimates.

241

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 103: AT&T Wireline Revenue by Segment


$ millions, except per subscriber and per share figures
REVENUE BY CUSTOMER TYPE
Households in service
% change, y-t-y
% change, q-t-q
Net change
Average lines

1Q09
28,057
-8.5%
-1.5%
(413)
28,264

2Q09
27,505
-8.0%
-2.0%
(552)
27,781

3Q09
26,948
-7.4%
-2.0%
(557)
27,227

4Q09
26,527
-6.8%
-1.6%
(421)
26,738

2009
26,527
-6.8%
-6.8%
(1,943)
27,499

1Q10
26,155
-6.8%
-1.4%
(411)
26,341

2Q10
25,521
-7.2%
-2.4%
(401)
25,838

3Q10
25,144
-6.7%
-1.5%
(391)
25,333

4Q10E
24,814
-6.5%
-1.3%
(330)
24,979

2010E
24,814
-6.5%
-6.5%
(1,713)
25,671

1Q11E
24,504
-6.3%
-1.2%
(310)
24,659

2Q11E
24,214
-5.1%
-1.2%
(290)
24,359

3Q11E
23,924
-4.9%
-1.2%
(290)
24,069

4Q11E
23,634
-4.8%
-1.2%
(290)
23,779

2011E
23,634
-4.8%
-4.8%
(1,180)
24,224

ARPU
% change, y-t-y
% change, q-t-q

$63.29
1.9%
0.2%

$64.24
2.0%
1.5%

$65.01
2.3%
1.2%

$65.50
3.8%
0.8%

$64.50
2.3%
2.3%

$66.99
5.9%
2.3%

$69.05
7.5%
3.1%

$70.03
7.7%
1.4%

$70.28
7.3%
0.4%

$68.93
6.9%
6.9%

$70.34
5.0%
0.1%

$71.81
4.0%
2.1%

$72.48
3.5%
0.9%

$72.74
3.5%
0.4%

$71.81
4.2%
4.2%

5,366
-6.9%

5,354
-6.4%

5,310
-5.5%

5,254
-3.6%

21,284
-5.6%

5,294
-1.3%

5,352
0.0%

5,322
0.2%

5,267
0.2%

21,235
-0.2%

5,204
-1.7%

5,247
-2.0%

5,234
-1.7%

5,189
-1.5%

20,874
-1.7%

Global enterprise solutions


Voice & other revenue
% change, y-t-y
% change, q-t-q

1,739
-8.6%
-2.7%

1,692
-10.0%
-2.7%

1,651
-12.0%
-2.4%

1,559
-12.8%
-5.6%

6,641
-10.8%
NM

1,528
-12.1%
-1.9%

1,467
-13.3%
-4.0%

1,419
-14.1%
-3.3%

1,409
-9.6%
-0.7%

5,823
-12.3%
NM

1,390
-9.0%
-1.3%

1,357
-7.5%
-2.4%

1,341
-5.5%
-1.1%

1,321
-6.3%
-1.5%

5,409
-7.1%
NM

Data revenue
% of enterprise revenue
% change, y-t-y
% change, q-t-q

2,104
55%
-5%
-2%

2,069
55%
-6%
-2%

2,039
55%
-8%
-1%

2,130
58%
-1%
4%

8,342
56%
-5%
-5%

2,111
58%
0%
-1%

2,111
59%
2%
0%

2,129
60%
4%
1%

2,114
60%
-1%
-1%

8,465
59%
1%
1%

2,176
61%
3%
3%

2,168
62%
3%
0%

2,189
62%
3%
1%

2,202
63%
4%
1%

8,735
62%
3%
3%

Global enterprise solutions revenue


% change, y-t-y

3,843
-6.6%

3,761
-8.0%

3,690
-10.0%

3,689
-6.1%

14,983
-7.7%

3,639
-5.3%

3,578
-4.9%

3,548
-3.8%

3,523
-4.5%

14,288
-4.6%

3,566
-2.0%

3,524
-1.5%

3,530
-0.5%

3,523
0.0%

14,144
-1.0%

Small business & alternate channels


Retail business access lines
% change, y-t-y
% change, q-t-q
Net change
Average lines

21,364
-5.6%
-2.1%
(462)
21,595

20,864
-7.0%
-2.3%
(500)
21,114

20,464
-7.6%
-1.9%
(400)
20,664

20,106
-7.9%
-1.7%
(358)
20,285

20,106
-7.9%
-7.9%
(1,720)
20,966

19,854
-7.1%
-1.3%
(252)
19,980

19,465
-6.7%
-2.0%
(389)
19,660

19,089
-6.7%
-1.9%
(376)
19,277

18,763
-6.7%
-1.7%
(326)
18,926

18,763
-6.7%
-6.7%
(1,343)
19,435

18,462
-7.0%
-1.6%
(301)
18,613

18,186
-6.6%
-1.5%
(276)
18,324

17,935
-6.0%
-1.4%
(251)
18,061

17,835
-4.9%
-0.6%
(100)
17,885

17,835
-4.9%
-4.9%
(928)
18,299

ARPU
% change, y-t-y
% change, q-t-q

$34.65
1.2%
2.2%

$34.18
1.3%
-1.4%

$34.02
-1.5%
-0.5%

$34.66
2.2%
1.9%

$34.29
0.2%
0.2%

$35.09
1.2%
1.2%

$35.00
2.4%
-0.3%

$35.29
3.7%
0.8%

$35.52
2.5%
0.7%

$35.27
2.8%
2.8%

$35.96
2.5%
1.2%

$35.70
2.0%
-0.7%

$36.00
2.0%
0.8%

$36.23
2.0%
0.7%

$35.82
1.6%
1.6%

1,545
-6%
0%

1,481
-7%
-4%

1,400
-13%
-5%

1,358
-12%
-3%

5,784
-10%
NM

1,333
-14%
-2%

1,278
-14%
-4%

1,212
-13%
-5%

1,198
-12%
-1%

5,021
-13%
NM

1,190
-11%
-1%

1,160
-9%
-3%

1,150
-5%
-1%

1,141
-5%
-1%

4,640
-8%
NM

700
31%
1%
2%

684
32%
0%
-2%

709
34%
1%
4%

751
36%
9%
6%

2,844
33%
3%
3%

770
37%
10%
3%

786
38%
15%
2%

829
41%
17%
5%

819
41%
9%
-1%

3,204
39%
13%
13%

818
41%
6%
0%

803
41%
2%
-2%

801
41%
-3%
0%

803
41%
-2%
0%

3,224
41%
1%
1%

2,245
-3.9%

2,165
-5.0%

2,109
-8.7%

2,109
-5.7%

8,628
-5.8%

2,103
-6.3%

2,064
-4.7%

2,041
-3.2%

2,017
-4.4%

8,225
-4.7%

2,008
-4.5%

1,962
-4.9%

1,950
-4.4%

1,944
-3.6%

7,865
-4.4%

Wholesale & GEM solutions


Voice & other revenue
% change, y-t-y
% change, q-t-q

2,123
-4%
-6%

2,135
-6%
1%

2,108
-8%
-1%

1,981
-12%
-6%

8,347
-8%
NM

1,908
-10%
-4%

1,918
-10%
1%

1,854
-12%
-3%

1,835
-7%
-1%

7,515
-10%
NM

1,769
-7%
-4%

1,799
-6%
2%

1,804
-3%
0%

1,780
-3%
-1%

7,153
-5%
NM

Data revenue
% of wholesale revenue
% change, y-t-y
% change, q-t-q

1,996
48%
-1%
-5%

2,018
49%
-4%
1%

2,009
49%
-5%
0%

2,086
51%
-1%
4%

8,109
49%
-3%
-3%

2,022
51%
1%
-3%

2,045
52%
1%
1%

2,090
53%
4%
2%

2,069
53%
-1%
-1%

8,227
52%
1%
1%

2,003
53%
-1%
-3%

2,045
53%
0%
2%

2,061
53%
-1%
1%

2,046
53%
-1%
-1%

8,155
53%
-1%
-1%

4,119
-2.8%

4,153
-4.8%

4,117
-6.7%

4,067
-6.8%

16,456
-5.3%

3,930
-4.6%

3,963
-5.5%

3,944
-5.5%

3,904
-4.0%

15,741
-4.3%

3,773
-4.0%

3,844
-3.0%

3,865
-2.0%

3,826
-2.0%

15,308
-2.8%

10,207
-4.5%

10,079
-6.1%

9,916
-8.4%

9,865
-6.3%

40,067
-6.3%

9,672
-5.2%

9,605
-4.7%

9,533
-3.9%

9,444
-4.3%

38,254
-4.5%

9,347
-3.4%

9,331
-2.9%

9,346
-2.0%

9,293
-1.6%

37,317
-2.5%

961
19%

997
14%

1,046
15%

1,087
16%

4,091
16%

1,104
15%

1,155
15%

1,207
15%

1,249
15%

4,715
15%

1,214
10%

1,271
10%

1,328
10%

1,374
10%

5,186
10%

521
-28%
-9%

480
-29%
-8%

446
-29%
-7%

409
-29%
-8%

1,856
-29%
NM

377
-28%
-8%

344
-28%
-9%

333
-25%
-3%

296
-28%
-11%

1,350
-27%
NM

290
-23%
-2%

265
-23%
-9%

256
-23%
-3%

228
-23%
-11%

1,039
-23%
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

0
0%
NM
NM

521
-28%

480
-29%

446
-29%

409
-29%

1,856
-29%

377
-28%

344
-28%

333
-28%

296
-28%

1,350
-27%

290
-23%

265
-23%

256
-23%

228
-23%

1,039
-23%

68
NM
16,162
-6%

76
NM
15,989
-7%

77
NM
15,749
-8%

86
682%
15,614
-6%

307
NM
63,514
-6%

78
15%
15,421
-5%

95
25%
15,396
-4%

87
13%
15,275
-3%

87
1%
15,094
-3%

347
13%
61,186
-4%

87
12%
14,928
-3%

87
-8%
14,930
-3%

87
0%
14,923
-2%

87
0%
14,797
-2%

348
0%
59,578
-3%

Consumer markets revenue


% change, y-t-y

Voice & other revenue


% change, y-t-y
% change, q-t-q
Data revenue
% of business in-region revenue
% change, y-t-y
% change, q-t-q
Small business & alternate channels revenue

Wholesale & GEM solutions revenue


% change, y-t-y
AT&T business solutions
% change, y-t-y
Strategic business services
% change, y-t-y
National mass markets
Voice & other revenue
% change, y-t-y
% change, q-t-q
Data revenue
% of national mass markets revenue
% change, y-t-y
% change, q-t-q
National mass markets revenue
% change, y-t-y
Other
% change, y-t-y
Wireline operating revenues
% change, y-t-y

Source: Company reports and J.P. Morgan estimates.

242

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 104: AT&T Wireline Income Statement


$ millions, except per subscriber and per share figures
Customer
Consumer
Business Solutions
Mass Markets
Other
Operating
Voice
Data
Other
Total revenue
% change, y-t-y
Cost of sales
SG&A
Total operating expenses
EBITDA (operating cash flow)
EBITDA margin
% change, y-t-y
Depreciation and amortization
Operating income

1Q09

2Q09

3Q09

4Q09

2009

1Q10

2Q10

3Q10

4Q10E

2010E

1Q11E

2Q11E

3Q11E

4Q11E

2011E

5,366
10,207
521
68

5,354
10,079
480
76

5,310
9,916
446
77

5,254
9,865
409
86

21,284
40,067
1,856
307

5,294
9,672
377
78

5,352
9,605
344
95

5,322
9,533
333
87

5,267
9,444
296
87

21,235
38,254
1,350
347

5,204
9,347
290
87

5,247
9,331
265
87

5,234
9,346
256
87

5,189
9,293
228
87

20,874
37,317
1,039
348

8,503
6,282
1,377
16,162
-5.7%

8,255
6,323
1,411
15,989
-6.6%

7,943
6,448
1,358
15,749
-7.7%

7,623
6,508
1,483
15,614
-5.7%

32,324
25,561
5,629
63,514
-6.4%

7,479
6,631
1,311
15,421
-4.6%

7,219
6,848
1,329
15,396
-3.7%

6,973
6,928
1,374
15,275
-3.0%

6,637
7,034
1,424
15,094
-3.3%

28,308
27,441
5,438
61,186
-3.7%

6,573
7,083
1,272
14,928
-3.2%

6,486
7,155
1,289
14,930
-3.0%

6,349
7,241
1,333
14,923
-2.3%

6,123
7,293
1,381
14,797
-2.0%

25,531
28,772
5,275
59,578
-2.6%

7,696
3,232
10,928
5,234
32.4%
-12.5%

7,570
3,350
10,920
5,069
31.7%
-19.5%

7,403
3,355
10,758
4,991
31.7%
-16.4%

7,304
3,326
10,630
4,984
31.9%
-11.1%

29,974
13,262
43,236
20,278
31.9%
-15.0%

7,266
3,346
10,612
4,809
31.2%
-8.1%

7,048
3,341
10,389
5,007
32.5%
-1.2%

6,912
3,406
10,318
4,957
32.5%
-0.7%

6,898
3,441
10,339
4,755
31.5%
-4.6%

28,123
13,535
41,658
19,528
31.9%
-3.7%

6,792
3,374
10,166
4,762
31.9%
-1.0%

6,763
3,374
10,138
4,793
32.1%
-4.3%

6,730
3,373
10,103
4,820
32.3%
-2.8%

6,659
3,344
10,003
4,794
32.4%
0.8%

26,945
13,465
40,409
19,169
32.2%
-1.8%

3,176
2,058

3,196
1,873

3,228
1,763

3,243
1,741

12,843
7,435

3,099
1,710

3,123
1,884

3,118
1,839

3,155
1,600

12,495
7,033

3,120
1,642

3,120
1,672

3,119
1,701

3,093
1,702

12,452
6,717

Source: Company reports and J.P. Morgan estimates.

Table 105: AT&T Consolidated Income Statement


$ millions, except per subscriber and per share figures
1Q09

2Q09

3Q09

4Q09

2009

1Q10

2Q10

3Q10

4Q10E

2010E

1Q11E

2Q11E

3Q11E

4Q11E

2011E

11,646
1,192
0

11,960
1,262
0

12,372
1,255
0

12,585
1,232
0

48,563
4,941
0

12,850
1,047
0

13,186
1,056
0

13,675
1,505
0

13,745
1,121
0

53,456
4,729
0

13,897
992
0

14,065
983
0

14,263
1,169
0

14,162
1,130
0

56,386
4,274
0

8,503
6,282
1,377
(1)

8,255
6,323
1,411
0

7,943
6,448
1,358
0

7,623
6,508
1,483
(2)

32,324
25,561
5,629
(3)

7,479
6,631
1,311
0

7,219
6,848
1,329
0

6,973
6,928
1,374
0

6,637
7,034
1,424
0

28,308
27,441
5,438
0

6,573
7,083
1,272
0

6,486
7,155
1,289
0

6,349
7,241
1,333
0

6,123
7,293
1,381
0

25,531
28,772
5,275
0

1,249
0

1,211
0

1,162
0

1,102
0

4,724
0

1,041
0

1,007
0

961
0

959
0

3,968
0

958
0

947
0

932
0

930
0

3,766
0

323
0
30,571
-0.6%

313
(1)
30,734
-0.4%

317
0
30,855
-1.6%

325
2
30,858
-0.7%

1,278
1
123,018
0.4%

290
0
30,649
0.3%

163
0
30,808
0.2%

165
0
31,581
2.4%

169
0
31,087
0.7%

787
0
124,125
0.9%

151
0
30,926
0.9%

155
0
31,079
0.9%

157
0
31,443
-0.4%

161
0
31,180
0.3%

623
0
124,628
0.4%

Total Revenues - Reported


Wireless service
Voice
Data
Directory
Other
Total revenue
% change, y-t-y

11,646
8,503
6,281
1,249
2,892
30,571
-1%

11,960
8,255
6,323
1,211
2,985
30,734
0%

12,372
7,943
6,448
1,162
2,930
30,855
-2%

12,585
7,622
6,507
1,102
3,042
30,858
-1%

48,563
32,323
25,559
4,724
11,849
123,018
-1%

12,850
7,479
6,631
1,041
2,648
30,649
0%

13,186
7,219
6,848
1,007
2,548
30,808
0%

13,675
6,973
6,928
961
3,044
31,581
2%

13,745
6,637
7,034
959
2,714
31,087
1%

53,456
28,308
27,441
3,968
10,954
124,125
1%

13,897
6,573
7,083
958
2,414
30,926
1%

14,065
6,486
7,155
947
2,427
31,079
1%

14,263
6,349
7,241
932
2,659
31,443
0%

14,162
6,123
7,293
930
2,672
31,180
0%

56,386
25,531
28,772
3,766
10,172
124,628
0%

Cost of services and sales


SG&A
Total operating expenses
EBITDA (operating cash flow)
EBITDA margin
% change, y-t-y

12,195
7,753
19,948
10,623
34.7%
-2%

12,550
7,775
20,325
10,409
33.9%
-10%

12,897
7,660
20,557
10,298
33.4%
-3%

12,844
8,138
20,982
9,876
32.0%
-1%

50,486
31,326
81,812
41,206
33.5%
-4%

12,329
7,484
19,813
10,836
35.4%
2%

12,381
7,475
19,856
10,952
35.5%
5%

13,519
7,707
21,226
10,355
32.8%
1%

12,962
7,725
20,687
10,400
33.5%
5%

51,191
30,391
81,582
42,543
34.3%
3%

12,434
7,494
19,929
10,997
35.6%
1%

12,341
7,427
19,768
11,311
36.4%
3%

12,866
7,536
20,402
11,042
35.1%
7%

12,566
7,729
20,295
10,885
34.9%
5%

50,208
30,185
80,394
44,234
35.5%
4%

4,886
5,737
18.8%
-4%

4,903
5,506
17.9%
-16%

4,910
5,388
17.5%
-4%

5,015
4,861
15.8%
-1%

19,714
21,492
17.5%
-7%

4,826
6,010
19.6%
5%

4,838
6,114
19.8%
11%

4,891
5,464
17.3%
1%

4,928
5,472
17.6%
13%

19,483
23,060
18.6%
7%

4,892
6,105
19.7%
2%

4,891
6,420
20.7%
5%

4,888
6,154
19.6%
13%

4,861
6,024
19.3%
10%

19,532
24,703
19.8%
7%

Int. (expense) income, net

(849)

(879)

(853)

(798)

(3,379)

(767)

(754)

(729)

(713)

(2,963)

(710)

(703)

(693)

(704)

(2,811)

Equity in net income(loss) of affiliates


Other, net
Income before taxes

137
(90)
4,935

231
(47)
4,811

181
(56)
4,660

185
37
4,285

734
(156)
18,691

217
(109)
5,351

195
(78)
5,477

217
47
4,999

217
47
5,023

846
(93)
20,850

217
0
5,612

180
0
5,897

180
0
5,641

180
0
5,499

757
0
22,649

(1,809)
3,126

(1,613)
3,198

(1,468)
3,192

(1,266)
3,019

0
(6,156)
12,535

(1,880)
3,471

(1,884)
3,593

(1,753)
3,246

(1,765)
3,258

0
(7,282)
13,568

(1,972)
3,640

(2,072)
3,825

(1,982)
3,659

(1,932)
3,567

0
(7,957)
14,692

Weighted average shares, basic


Stock options
Other stock-based compensation
Weighted average shares, fully diluted

5,896
15
11
5,922

5,900
15
8
5,923

5,901
15
6
5,922

5,902
15
6
5,923

5,900
15
8
5,923

5,905
15
15
5,935

5,909
15
13
5,937

5,909
15
14
5,938

5,906
15
14
5,935

5,907
15
14
5,936

5,881
15
14
5,910

5,856
15
14
5,885

5,839
15
14
5,868

5,822
15
14
5,851

5,849
15
14
5,878

EPS (loss), basic


EPS (loss), fully diluted
% change, y-t-y

$0.53
$0.53
-8%

$0.54
$0.54
-15%

$0.54
$0.54
-1%

$0.51
$0.51
26%

$2.12
$2.12
-2%

$0.59
$0.59
12%

$0.61
$0.61
12%

$0.55
$0.55
1%

$0.55
$0.55
8%

$2.30
$2.29
8%

$0.62
$0.62
4%

$0.65
$0.65
7%

$0.63
$0.62
14%

$0.61
$0.61
11%

$2.51
$2.50
9%

Dividend Paid Per Share


Payout ratio (dividend + buyback)

$0.41

$0.41

$0.41

$0.42

$1.65

$0.42
61%

$0.42
64%

$0.42
59%

$0.43
115%

$1.69
70%

$0.43
82%

$0.43
79%

$0.43
77%

$0.45
96%

$1.74
83%

Operating Revenues
Wireless
Service revenue
Equipment revenue
Eliminations
Wireline
Voice
Data
Other
Eliminations
Advertising & Publishing
Total revenue
Eliminations
Other
Total revenue
Eliminations
Total Revenues - From Segments
% change, y-t-y

Depreciation and amortization


Operating income
Operating income margin
% change, y-t-y

One-time gains (loss)


Tax benefit (expense)
Net income (Loss)

Source: Company reports and J.P. Morgan estimates.

243

Philip Cusick, CFA


(1-212) 622-1444
philip.cusick@jpmorgan.com

North America Equity Research


13 January 2011

Table 106: AT&T Balance Sheet


$ millions, except per subscriber and per share figures
Assets
Cash and cash equivalents
Accounts receivable
Prepaid expenses
Deferred income taxes
Other current assets
Total current assets
Net PPE

1Q09

2Q09

3Q09

4Q09

2009

1Q10

2Q10

3Q10

4Q10E

2010E

1Q11E

2Q11E

3Q11E

4Q11E

2011E

3,812
14,965
1,704
957
2,083
23,521

7,348
14,846
1,786
964
1,996
26,940

6,167
14,796
1,791
991
2,176
25,921

3,741
14,845
1,562
1,247
3,792
25,187

3,741
14,845
1,562
1,247
3,792
25,187

2,617
14,167
1,820
1,293
2,465
22,362

1,377
13,780
1,666
1,225
3,344
21,392

3,246
13,606
1,686
1,059
2,380
21,977

2,893
13,606
1,686
1,059
2,380
21,624

2,893
13,606
1,686
1,059
2,380
21,624

3,592
13,606
1,686
1,059
2,380
22,323

4,458
13,606
1,686
1,059
2,380
23,189

3,458
13,606
1,686
1,059
2,380
22,189

3,573
13,606
1,686
1,059
2,380
22,304

3,573
13,606
1,686
1,059
2,380
22,304

98,339

98,229

98,321

100,053

100,053

99,360

100,302

101,475

100,649

100,649

100,300

99,978

99,689

99,982

99,982

71,694
47,461
9,605
5,797
2,413
5,528
264,358

71,691
47,674
8,682
5,773
2,749
6,180
267,918

71,727
47,946
7,814
5,656
2,813
6,370
266,568

72,782
48,741
7,393
5,494
2,921
6,275
268,846

72,782
48,741
7,393
5,494
2,921
6,275
268,846

73,052
49,028
6,624
5,680
3,278
6,317
265,701

73,484
49,957
6,047
5,539
4,346
6,489
267,556

73,447
50,113
5,369
5,525
4,544
6,802
269,252

73,447
50,113
5,369
5,525
4,544
6,802
268,073

73,447
50,113
5,369
5,525
4,544
6,802
268,073

73,447
50,113
5,369
5,525
4,544
6,802
268,423

73,447
50,113
5,369
5,525
4,544
6,802
268,967

73,447
50,113
5,369
5,525
4,544
6,802
267,678

73,447
50,113
5,369
5,525
4,544
6,802
268,086

73,447
50,113
5,369
5,525
4,544
6,802
268,086

10,790
17,359
4,065
2,399
2,419
37,032

10,155
18,046
3,932
1,667
2,419
36,219

6,755
18,093
4,036
1,965
2,419
33,268

7,361
21,260
4,170
1,681
2,479
36,951

7,361
21,260
4,170
1,681
2,479
36,951

9,437
18,087
4,061
2,639
2,482
36,706

9,721
18,157
3,943
1,879
2,482
36,182

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

6,426
18,417
3,933
1,416
2,482
32,674

Long-term debt
Deferred income taxes
Postemployment benefit obligation
Other noncurrent liabilities
Total liabilities

63,560
19,376
32,032
14,663
166,663

66,565
20,354
31,985
13,783
168,906

65,909
22,279
31,750
13,361
166,567

64,720
23,781
27,847
13,226
166,525

64,720
23,781
27,847
13,226
166,525

60,024
25,520
27,709
13,276
163,235

60,277
25,615
27,421
14,578
164,073

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

62,540
20,651
27,071
13,023
155,959

Common shares issued ($1 par value)


Capital in excess of par value
Retained earnings
Treasury shares (at cost)
Accumulated other comprehensive loss
Shareholder's equity
Total liabilities and shareholder's equity

6,495
91,638
37,296
(21,283)
(16,451)
97,695
264,358

6,495
91,637
38,069
(21,284)
(15,905)
99,012
267,918

6,495
91,678
38,841
(21,280)
(15,733)
100,001
266,568

6,495
91,707
39,366
(21,260)
(13,987)
102,321
268,846

6,495
91,707
39,366
(21,260)
(13,987)
102,321
268,846

6,495
91,557
39,373
(21,137)
(13,822)
102,466
265,701

6,495
91,628
40,909
(21,134)
(14,415)
103,483
267,556

6,495
91,748
50,751
(21,112)
(14,589)
113,293
269,252

6,495
91,748
49,572
(21,112)
(14,589)
112,114
268,073

6,495
91,748
49,572
(21,112)
(14,589)
112,114
268,073

6,495
91,748
49,922
(21,112)
(14,589)
112,464
268,423

6,495
91,748
50,466
(21,112)
(14,589)
113,008
268,967

6,495
91,748
49,177
(21,112)
(14,589)
111,719
267,678

6,495
91,748
49,585
(21,112)
(14,589)
112,127
268,086

6,495
91,748
49,585
(21,112)
(14,589)
112,127
268,086

Goodwill
Licenses
Customer lists and relationships - net
Other intangible assets - net
Investments in equity affiliates
Postemployment benefit
Other assets
Total assets
Liabilities and Shareowners' Investment
Current liabilities
Debt maturing within one year
Accounts payable and accrued liabilities
Advanced billing and customer deposits
Accrued taxes
Dividends payable
Total current liabilities

Source: Company reports and J.P. Morgan estimates.

Table 107: AT&T Cash Flow Statement


$ millions, except per subscriber and per share figures
QUARTERLY
Cash Flows From Operating Activities
Net Income
Adjustments to reconcile net income to net cash provided by
Depreciation and amortization
Undistributed earnings from investments in equity affiliates
Provision for uncollectible accounts
Deferred income tax expense (benefit)
Net gain on sales of investments
Gain on license exchange
Changes in operating assets and liabilities:
Accounts receivable
Other current assets
Accounts payable and accrued liabilities
S