REVISED 2008 SECTOR FORECAST
The U.S. Pay TV sector is being buffeted by several factors -- the slowing economy/possible
recession, housing bubble bursting, ramping RBOC wired competition, impact of the tripleplay
– which makes forecasting 2008 quite challenging. In aggregate, we believe these
factors will cause a continued slowdown in video sector growth, in particular for cable, but
not as severely as some had feared and as the stocks of the publicly-traded companies
appear to be discounting.
2008 Forecast
We now expect 1.820m growth in video subscriptions in 2008 (for the publicly-reporting
companies), down 14% from 2.115m in 2007, which was down about 7% from 2.278m in
2006. Our forecast is neither a recession scenario nor a rosy outlook that the pace of growth
from the last couple of years was somehow normal. We expect the sector to settle out at a
1.7-1.8m pace of annual growth, showing an about 50bp increase in pay TV penetration each
year on top of an about 1% growth in TV households. By 2010 we have pay TV penetration
at 85.7% of TV homes, up from 83.9% at year-end 2007. Video penetration has been slow to
grow ever since hitting 80% in 2000.
Wide Range of 2008 Possibilities
The potential scenarios for 2008 remain varied, though the question, in our mind, is whether
new subscriptions will be down a little or a lot. One could argue that the 2.1-2.3m video sub
growth posted the past two years (2006-07) has been overheated due to the dual impact of a
strong housing/refinancing market and drawing in new subscribers with low-priced triple-play
promotions by both cable and RBOC/DBS co-marketing. Not only does that imply that the
2+m run-rate was abnormally high, but now with the economic slowdown and housing
shifting to a net negative from a net positive (relative to historical norms), its possible the
sector shifts to a well-below-trend results in 2008. Conversely, the fed’s aggressive move,
reducing the fed funds rate by 125bp over the past two weeks, might have “rescued” the
sector from any further deterioration related to housing and the economy might begin to
rebound by 2Q08 (as our economist believes).
2001/02 Correlation Difficult To Ascertain
There are few historical lessons to help forecast 2008. Certainly 2001/02 was a disaster
scenario with 2001 U.S. Pay TV net adds down 28% to 2.980m and 2002 down a further
74% to 778k, before rebounding with 97% growth in 2003 to 1.528m. But the variance was
only shown in cable and it remains unclear whether that was related to just the economy or
also changes in subscriber accounting methodology as AT&T and Charter represented the
vast majority of the variance, while other cable companies were mixed in 2001/02 with some
even having increases.
1990/91 Correlation Difficult To Ascertain
Back in 1990-92, cable net adds slowed, but so was the pace of the industry’s build out of
homes passed. In 1990, cable net adds were down 33% to 2.334m, down a further 32% in
1991 to 1.595m and up 8% to 1.729m in 1992. But the expansion in homes passed slowed
from a new 5.667m homes passed in 1999 to 3.194m in 2000, 2.338m in 2001 and 1.300m in
2002. Penetration of homes passed increased every year during that cycle.
Why the 2H07 Slowdown?
The common perception seems to be that the housing slowdown and/or economy is what
drove 3Q07 pay TV net adds down 28% year-over-year to 430k, after a 5 quarter stretch of
growth. In fact, the variance from 2Q’s 16% growth was so dramatic that we are still seeing
the fallout in cable forecasts. We also expect 4Q07 to be down, -31% to 542k. With this
2H07 softness it is easy to gauge that 1H08 is also likely to be down, but 2H08 remains
intriguing since the root cause of the 2H07 softness is still unknown. If it was the economy,
then there should have been a bigger jump in bad debt expense (everyone from cable to
satellite to the RBOCs indicated consumer bad debt went up “slightly” or not at all, and
before anyone gets excited about DISH’s sharp churn increase remember a portion of its
churn was artificial as they changed the definition of churn during the quarter to a fewer

number of days outstanding), not to mention unemployment did not start to rise until the very
end of the year and is up only modestly. If it was housing, then why was the impact not
seen well before 3Q07 and why not an even more severe impact in 4Q07 rather than a
similar decline (31% y/y drop in pay TV net adds, vs. 28% in 3Q)? We also have had a hard
time finding any correlation between housing activity and pay TV subscriptions. Certainly
there must have been some positive impact from more people owning more homes, but at
the same time the primary driver of pay TV subscriptions is number of households, not
number of housing units. Even for those families suffering through foreclosures, they are
likely moving into a rental unit, and likely hooking up the cheapest form of entertainment
available, pay TV.
The Triple-Play Impact
Given the timing of the triple-play rollouts by both cable and the RBOCs, in particular by
Comcast, it appears to us that the rapid increase in pay TV net additions from 2Q06 through
2Q07 was driven by the triple-play promotions driving marginal customers into pay TV.
Similarly, the fall-off the sector is seeing in 2H07 is for the most part tough comparisons
against that stretch. One simple demonstration of this phenomenon is to compare 3Q07 and
4Q07 with two years ago -- 3Q07 pay TV net adds of 430k were up 8% from 3Q05, while
542k for 4Q07E are up 33% from 4Q05.
Conclusion
The subscriber growth rate for the U.S. Pay TV sector is slowing, partly due to the market
being overheated in 2006/1H07 from triple-play marketing, partly due to the impact of the
slowdown in the economy and housing, and partly due to its maturity at an 84% penetration
rate. We think the tough triple-play comps is the biggest factor impacting results, which
would be bullish relative to investor fears as it would imply that growth will normalize in
2H08. If we are wrong, or if the economy does deteriorate into a sustained recession, our
2008 forecast of 1.820m could arguably be high by as much as 800k-1.3m net adds. Our
new forecast would have been meaningfully more conservative in the absence of the
aggressive fed funds rate cuts.
WHO GETS WHAT?
Of course, figuring out the overall market growth is just the beginning. Dividing up the
market for 2008 remains controversial, but in our view the RBOCs will continue scaling,
DBS’s pace of net adds will continue to decline, but the brunt of the slowdown will be felt by
cable. Of course, everyone will continue to claim they are adding high quality subscribers
and losing low-end subs, which might be just as important as how many subs are being lost!
With the RBOCs building out in more affluent areas and satellite having the HD lead and
higher ARPUs, we wonder if analog cable is where the low-end resides, but at the same time
certainly would not dispute Time Warner Cable’s commentary that the majority of their sub
losses are life-line analog subscribers.
RBOCs
Verizon continues to scale its FiOS video service, adding more subscribers in each
subsequent quarter since launch and continuing to add new markets at a steady pace,
implying a ramp to 1m net adds in 2008. The same goes for AT&T’s U-Verse video service,
for which AT&T management has guided to 770k net adds in 2008 (1m subs by year-end).
Qwest continues to indicate a wired video service is not economical. This implies the
RBOCs will add about 1.8m subscribers in 2008, up from 964k in 2007. Our forecasts
already were, and remain, in line with these figures. It remains highly uncertain, in our view,
how much of the RBOC video sub growth is coming from new market launches vs. market
share gains in older markets. As new market launches begin to slow in terms of contribution
to growth (which is happening to Verizon now and will hit AT&T by 2H08), and as churn on a
larger subscriber base begins to kick in, it will be interesting to see if their pace of net
additions will peak earlier than the market expects. AT&T remains the most difficult to
ascertain, as its product appears capacity constrained and we hear the occasional anecdote
about technology challenges (number of servers required, etc.), and it is lagging comparable

FiOS penetration levels during its early launch phase by a wide margin. Yet, it is still ramping
nicely, exiting 2007 with net additions of 12k/week from a near standing start at the
beginning of the year, and management is promising a 40k/week net add pace by year-end
2008.
DBS
The satellite TV operators continue to perform well, especially given DirecTV has been
constantly increasing its new customer credit score requirements the past couple of years.
Several factors have been helping including (1) DBS still sources 30-40% of so of its
subscribers from more rural areas where cable competition is less, the RBOCs are not
building out, and the economy and housing market have not been having as meaningful an
impact; (2) another 8% or so of sector subs are locked in on exclusive sports programming;
(3) another 10-15% are locked in on often exclusive foreign language programming; (4) both
DISH and DirecTV have been locking up customers for 18-24 months with contract
requirements for HD/PVR upgrades; (5) triple-play bundling with the RBOCs has offset some
of cable’s triple-play benefit; and (6) satellite has taken the lead in terms of number of HD
channels available and HD marketing to consumers. We believe DBS slows in 2008, as it has
for the past 3 years, but as much due to churn on their larger subscriber bases as due to
RBOC and cable competition, and the impact of the slowing economy and housing. We are
forecasting 1.125m net DBS additions, down 31% from 2007’s 1.640m, which was down
13% from 1.885m for 2006, which had dropped 19% from 2.328m in 2005.
Cable
This does not leave much for cable. In fact, we show cable having the worst year in its
history, losing more than 1m subscribers (-1.085m), though 2007 was already the fourth
worst year on record losing 489k. It is not difficult to get to our estimate, and to the extent
there is any negative sector variance from our overall forecast, we would not be surprised if
cable bears the brunt of it. Assuming 1Q08 performance looks similar to 2H07 (271k worse
net adds in 3Q07 and 302k worse in 4Q), then the trailing four quarters pace by 1Q08 is
already a 795k decline. We assume cable softens a bit further in 2009 and 2010 as the
RBOCs continue to expand the footprint for their video offerings.
STOCK IMPLICATIONS
Investors have already discounted the slowdown in not just video growth, but broadband as
well (which has been slowing since 4Q06 due to maturation of broadband penetrations). In
fact, we think the uncertain environment, both macro and competitively, and uncertainty
regarding capex trends (HD, PVRs, broadband capacity, wireless) have driven the stocks to
levels way below fair value. Certainly, these companies are trading at the cheapest
valuations seen in at least the past 14 years, if not longer (see included chart), and Comcast
and Time Warner Cable are holding at 60% of asset value, typically the floor for media
stocks. We see 40-70% upside in each of the 5 stocks over the next 12 months, though we
continue to favor satellite including DirecTV still as our top pick in the sector. We expect the
sector to be revalued higher as (1) investors gain confidence that FCF is returning to rapid
growth -- +46% to $7.076b expected in 2008, after a disappointing 6% drop to $4.846b in
2007; and (2) as the RBOCs scaling slows, which could happen with FiOS during 2H08.
Several other factors can have a significant influence on the stocks in 2008.
􀂄 First, downside could come if the economy heads into a severe recession and we need
to reduce our forecasts significantly. Even absent such a downturn, analyst estimates
might still need to be trimmed for 2008 as managements are likely to offer conservative
guidance given the sector uncertainty.
􀂄 Second, upside could come if managements more aggressively pursue cost
realignments and share repurchases/dividends. Certainly, some promise for change in
behavior comes with DirecTV coming under new ownership soon (we hope), Time Warner
Cable potentially being fully spun-off to shareholders this spring, and Comcast having a
new CFO helping develop 2008 budgets. We still do not hold out any hope for either

Echostar or Cablevision managements deploying their excess capital more wisely, but both
stocks are cheap regardless
􀂄 Third, the market might have to absorb a meaningful amount of new cable shares to the
extent Time Warner Cable is fully spun-out to Time Warner shareholders. Time Warner
owns about $20b of Time Warner Cable stock, which would add over 20% to cable shares
outstanding. Of course, Time Warner might offset some of this pressure by offering to
exchange some of its Time Warner Cable shares for investors’ Time Warner shares, but
there is bound to be some meaningful level of displacement.
􀂄 Fourth, M&A activity might have a positive or negative effect, depending on the sector.
For cable, Comcast and Time Warner Cable will remain consolidators, though it appears
there is little for sale and debt financing has become less favorable. Cablevision’s
controlling shareholders, the Dolans, have made three attempts to privatize the company
and, debt markets willing, it would not surprise anyone to see a fourth bid given the decline
in Cablevision’s share price. For satellite, we continue to believe that outright
combinations with the RBOCs are still a possibility, though we doubt even preliminary
discussions would take place prior to the end of the current 700Mhz wireless auctions and
the close of the Newscorp/Liberty swap of DirecTV shares.
􀂄 Lastly, the regulatory environment can be a major factor. We would argue that despite a
constant stream of rhetoric from the FCC, the rule changes have been fairly modest, and
that is likely to continue.