Goldman Sachs issued a report today on XM's Q4 and full year operations.

REPORT EXCERPTS

XM Satellite Radio (XMSR)

Sell

Approaching a fork in the road; 4Q07 review What's changed

XM’s 4Q results largely met consensus and our financial estimates, though fell modestly short on metrics with lighter net adds and higher costs per sub. Specifically, XM’s 4Q net adds of 460k was lower than our 533k estimate. SAC (ex 1x items) was slightly worse than expected at $81 (GSe $80), as was CPGA of $133 vs. our $127 estimate. Conversion improved to 53.9% vs. 52.5% in 3Q and 52.4% a year ago, churn held flat at 1.72% vs. 1.69% in 3Q2007 and 1.79% , in 4Q06 and ARPU was flat at $11.71.

Implications

Trends and management focus has shifted to an OEM-centric model, and the cost structure, as expected, is following the transition, with more costs as penetration, revenue share and royalty agreements ramp in lockstep.

However, what is unlikely to be linear is the path post the merger decision. Upon deal approval, we would expect an aggressive integration plan with an improved combined profitability profile, save for the cost of complying with deal conditions. Under the no merger scenario, we would expect fairly dramatic cuts in the overall expense base and likely business model changes, owing to the continued losses ($565mn at Sirius + $682mn at XMin 2007) and cash flow burn ($500mn total in 2007) of the companies. Thus the least likely path is one that follows linearly from the current trends. XM could be fully funded, but having drawn on its credit facility, further deviations from internal targets (as the company chose not to provide 2008 targets) may result in little if any cash cushion.

Valuation

We are maintaining our Sell rating and 12-month price target of $11.50 based on our DCF and the expected value upon a merger. Key risks Upside risks include merger approval, stable costs, accelerating net adds, steady conversion rates and churn; downside risks include the opposite.

Stock Recommendation

We continue to rate XMSR shares Sell. Even with the market incorporating a fair probability of merger approval, the stock is down about 20% since the merger announcement a year ago as lower expected cash flows and deteriorating channel trends have weighed on the stock. Given the fall off in the retail satellite radio market and little anticipation of new content deals or compelling technology on the horizon that might drive an upside surprise, we still find it challenging to be comfortable with XM’s current valuation. On February 20, 2007 XM and Sirius announced a definitive agreement to combine in a merger of equals. Based on our analysis, we continue to estimate a 30% chance of a clean approval (meaning non-substantive conditions). Excluding the merger proposal, we do not believe the fundamentals, subscriber growth curves, or risk-adjusted estimates support a premium or equal valuation for Sirius versus XM. Therefore, we continue to think the risk to the downside for SIRI shares remains greater relative to XMSR, even though we also rate XMSR shares SELL, owing to the risk/reward profile of the stock assuming the assorted merger outcomes.

Subscriber analysis

We are adjusting our 2008 year-end subscriber estimates of 10.4mn. Our 2008 estimates imply net adds of 1.4mn, with only 70k in retail and the remaining 1.33mn in OEM. This compares to 2006, a year in which XM added 1.7mn subscribers (789k retail and 907k OEM) to end the year with at 7.6 million subscribers.

The increasing proportion of OEM net adds has been expected, though the scale has increased faster, likely owing to displacement of what might have otherwise been a retail net add. We understand that on average, 9 out of every 10 retail radios are sold with car kits, implying that as the factory installation rates grow, the retail market requires new growth drivers. Already we forecast fewer net adds in the retail channel yoy for the next five-year forecast period.

To illustrate our thoughts on why Sirius has more relative risk to weakness in the retail channel versus XM, look at how Sirius’ net retail additions as a percentage of gross retail additions has tracked XM’s on a one year lag (See arrows in Exhibit 1). As Sirius has significant retail net adds, there is more to lose if the channel deteriorates further.

Our probability-weighted approach yields $11.50 for XMSR shares

Our $11.50 12-month price target for shares of XM Satellite Radio is based on our probability weighted approach of our DCF-based valuation versus the expected value upon merger. In arriving at our price targets, we handicap the announcement of a merger proposal, receipt of and conditions attached to various approvals, and the estimated success in executing potential synergies.

Fundamental valuation yields $10.50 per XMSR share For our fundamental valuation framework, we employ a discounted cash flow analysis. Based on this methodology, our fundamental analysis yields a value of $10.50 per share for XM. In our view, EV/subscriber metrics are viable in mature subscriber businesses with known ARPU, per sub profitability, churn, and associated metrics, but not in early stage, unprofitable businesses.

4Q07 review and 1Q08 outlook

XM’s 4Q results largely met consensus and our financial estimates, though fell modestly short on metrics with lighter net adds and higher costs per sub. In 3Q2007, revenue of $307mn (GSe $303mn) rose 20% yoy, while SAC and CPGA (ex-1x items) were slightly higher than expected at $81 and $133 versus GSe $80 and $127. That said, owing to lower subscriber additions, adjusted OI loss of ($134)mn (ex-ESO and 1x items) was slightly better than our ($144)mn estimate, leading to adjusted LPS (ex-1x items) of ($0.56), a penny better than our ($0.57) estimate. Encouragingly, conversion improved to 53.9% vs. 52.5% in 3Q and 52.4% a year ago, churn held flat at 1.72% vs. 1.69% in 3Q2007 and 1.79% ,a year ago and ARPU was essentially flat at $11.71. XM ended the year with $157mn in cash, versus $218mn last year. Note that a sale leaseback in 1Q2007 netted $288mn in proceeds, and 4Q2007 use of cash was $75mn.

For the first quarter, we expect net additions 310k vs. 285k a year ago. With the retail market continuing to weaken, we expect retail & OEM gross additions of 253k & 784k, respectively. On a net basis, we estimate only 4,000 retail net adds vs. 306,000 OEM. Said differently, our estimates imply net adds to be 99% OEM in 1Q08.

We expect 2008 revenue & EBITDA (pre SBC) of $308mn & ($15mn), respectively, resulting in LPS of ($0.36). Additionally, our SAC & CPGA estimates are $68 & $100. For the full year we revenue & EBITDA (pre SBC) of $1,316mn & ($145mn), respectively, resulting in LPS of ($1.70). Additionally, our SAC & CPGA estimates are $75 & $112.

What to watch for: Issues that will drive the stock

(1) Larger-than-expected slowdown in auto sales: Recent comments from the automotive industry (both OEM and dealer chain) indicate that many of the issues impacting the subprime, CDO and CLO market are projected to trickle down and carry negative implications on new car sales, implying around 15-16 million units sold in 2008. As satellite radio operators progress towards an OEM centric model, watch for primary and derivative effects of a housing and auto slowdown to temper the subscriber ramp.

2)Marketing shifting toward OEM channel: XM continues to shift marketing dollars from the retail channel toward strengthening the OEM channel given that the OEM channel is expected to contribute a larger portion of new subscribers. On the OEM side, we expect XM to weight more of its ad budget towards co-promotions with OEM partners rather than traditional retail marketing. XM believes that additional communication with trial OEM subscribers has helped bolster conversion and retention rates. Within the retail channel we expect XM to allocate more promotional dollars to event-based marketing.

3) OEM penetration to ramp:Watch for acceleration in installs across OEMs as XM pushes to ramp penetration across all models. Specifically, the total number of vehicles produced with factory installed XM is over 9mn, with GM expected to manufacture more than 2.5mn vehicles in the model year 2008. That said, XM does not expect the revenue share agreements to increase as penetration ramps, stating that it would most likely be flat to down over time. We believe this is due to GM having installed its 8 millionth vehicle, after which point the revenue share (percentage) is static.

4) OEM conversion ratios relatively static:Watch for the conversion of OEM promotional subscribers to full paying subscribers to decline over the long term. That said, currently the conversion rate is improving, as XM reported 53.9% vs. 52.5% in 3Q and 52.4% a year ago. We estimate a gradually more conservative outlook in our model, predicting rates approaching 40%-45% by 2010 for both XM and Sirius. XM has historically stated that they expect the conversion rate to fluctuate around the mid 50% range. OEM conversion rates will gradually become less important as only GM and Honda subscribers are measured for conversion and all other OEM partners are counted as subscribers after electing to become a paying subscriber. We think the arrangements between Toyota, Hyundai and Nissan will be large drivers of the OEM ramp in 2008 and will alleviate focus on declining conversion rates. Since the OEM does not prepay for the initial subscription term, XM is not initially recording the customer as a subscriber. Rather, the “trial sub” will not be reflected in the sub roll until they voluntarily convert to a self-paid subscription.

5) Sufficient Liquidity:XM could be fully funded given the current operating plan. Notes from XM 10-K regarding liquidity:“Provided that we meet the revenue, expense and cash flow projections of our current business plan, we expect to be fully funded and not need additional liquidity to continue operations beyond our existing assets, credit facilities and cash generated by operations; our current business plan is based on estimates regarding expected future costs, expected future revenue and assumes the refinancing or renegotiating of certain of our obligations as they become due, including the maturity of our existing credit facilities and $400 million of convertible notes in 2009.” Consumption of $75mn of cash in 4Q contradicts the $75mn of cash flow generated by Sirius. Mapping out the quarterly cash flow estimates absent a merger seems critical in assigning value in a ‘no-merger’ situation

Upside:

•Ultimate satellite radio subscriber universe is larger than our expectations and it takes less time than expected to reach.

•A merger is approved by all regulatory bodies and shareholders and expected synergies are fully realized.

•Programming and distribution costs remain at current levels or decrease, bringing cash flow break-even sooner than our estimates indicate.

•Widespread consumer acceptance of both the service and new products.

Downside:

•A merger is rejected and the stocks trade down to levels consistent with fundamentals.

•Satellite problems, chipset delays, or OEM partner production issues.

•Disruption to key exclusive content, or rising royalty rates for music.

•Failure to attract and retain new subscribers with exclusive content.

•Further complications with regulators including the FCC or DOJ

Tyler Savery Position - Long Sirius, Long XM