by Armand Musey

We have always been a strong supporter of the business model for satellite radio. While initial upfront investment is high, satellite radio’s low variable costs and a nationwide distribution footprint should result in breakeven economics with modest market penetration and strong bottom line growth thereafter. When both Sirius and XM Radio were in their development stages it was really too early to differentiate between the two companies. They were both “concept stocks” and we were a strong supporter of the “concept” behind both companies. However, now that both companies have launched their services and execution has become an issue, we find it easier to differentiate between the two. In this regard, we believe XM Radio has clearly pulled ahead of Sirius.

When Sirius sought financing in 1999, the company’s prospectus stated it expected to begin service in late 2000. A series of chipset development problems have now delayed this launch by more than a year and pushed-out OEM factory installation until the 2004 model year. Moreover, our sources working with Sirius’ receiver manufacturers indicate that while they went forward with their manufacturing plans, the chipset available in Sirius receivers at the time of its initial launch was only a temporary fix. We suspect that chipset problems and network optimization issues are behind Sirius’ six-month rollout schedule, while XM Radio took just 45 days to complete its national rollout. We also think the delay of Sirius’ OEM rollout until the 2004 model year is a competitive setback.

We have sought to investigate potential system architecture problems behind Sirius’ execution problems and have gotten few concrete answers. In addition to problems with the development of its chipset, which management has acknowledged, Sirius has also been plagued by speculation that its repeater network does not function effectively. Specifically, that it suffers from “dead spots” where Sirius’ receivers do not receive the satellites’ signals. This speculation surfaced about a year ago when Sirius announced that it would be conducting extensive “rolling” tests before commercial introduction, as opposed to XM, which went straight to market once its satellites came on line. These concerns have persisted, however, because of the prolonged nature of Sirius’ rollout.

On a fundamental level, we suspect the constant motion and unique orbital ellipses of Sirius’ three satellite constellation could be making it difficult for Agere to program the chips to correlate the broadcast signals and/or for Sirius to locate shadows behind buildings and other obstacles where the signal fades out. We believe XM is not facing these same issues because of the more traditional stationary orbit of its two geostationary satellites and its higher concentration of smaller signal repeaters. While we have not been able to confirm this theory, if it is true, it is entirely possible that Sirius will be able to “engineer” a solution. Consultations with satellite engineers give us a high level of comfort that this is a likely scenario.

Following the announcement by GM that XM Radio receivers would be available as factory installed options on 23 models in its 2003 line-up (more than one-half of the total line- up), Sirius was compelled to announce that it would be offered as a dealer installed option in Daimler Chryslers and Fords. The fact that Sirius will not be available as a factory installed option in the 2003 model line-up is significant for two reasons. First, we believe it confirms that the Sirius chipset is still not “production enough” for the automobile OEM to risk integrating it into their complex production schedules. Second, because we think that the “push distribution” of the OEM channel will be key to mass market subscriber acquisitions, a one-year delay in entering this channel represents a competitive setback, although one that is hard to quantify.

As we write this column, shares of Sirius are trading at a significant discount to those of XM Radio. We believe this discount is appropriate given Sirius’ disappointing execution and increased financial risk. Its slow rollout may even cause it to break some of its debt covenants. Since Sirius has a long history of execution problems and management’s past guidance has proved unreliable, until we see that Sirius has a production chipset that actually works in multiple markets we would remain on the sidelines. Instead we encourage investors looking for a pure play on the demand for satellite radio to look instead to XM Radio, with its track record of strong execution

Armand Musey is the satellite communications analyst at Salomon Smith Barney (SSB). The foregoing article should not be considered as a recommendation with respect to any security. SSB and its affiliates may maintain a long or short position in, act as a market maker for, or purchase or sell a position in, securities of referenced entities and may also perform investment banking, advisory, or other services for any such entity.