As Harbinger tries to secure partners for its proposed 4G LTE network, it now looks like Craig McCaw and Clearwire could potentially end up playing spoiler.
A research note from Credit Suisse, published on Monday highlighted the potential value of MSS spectrum and suggested it could be worth as much as $0.50 per MHzPOP. It also noted that the spectrum value of Clearwire “is likely underappreciated”. No substantive justification is given for the $0.50 per MHzPOP valuation suggested in the research note, other than the fact that this is close to the median value in the 2006 AWS spectrum auctions, and there are a number of elementary mistakes in the analysis of TerreStar (most notably that no mention is made of the more than $400M of Convertible Preferred Stock which must be repaid before any value from the 1.4GHz spectrum is realized by the equity holders). The fact that the FCC “will expect a portion of the
step up in value for both [2GHz] MSS and broadcast spectrum” is also only mentioned in passing, despite the fact that this will have a huge impact on the realizable value if the spectrum is ultimately included in an incentive auction. Despite this, the research certainly seems to have stimulated some increased interest in MSS-ATC spectrum, as evidenced by the jump in the price of TerreStar’s stock on Monday, and could be seen as potentially helping Harbinger in its quest to attract investors for its 4G network.
However, perhaps coincidentally, last Friday, Business Week published an article highlighting Clearwire’s spectrum value (which it puts at “$20B or more” or $0.50/MHzPOP). In my view, this article clearly indicates that Clearwire is now open to offers for purchase of some of its spectrum. In that case, wireless operators who are looking for spectrum in the short or medium term, such as T-Mobile, would certainly have a viable alternative to MSS-ATC spectrum if they are looking to build out a 4G network. Given that there are only a few potential purchasers of wireless spectrum at the moment (and the it now looks like the FCC doesn’t want AT&T and Verizon to buy any more spectrum in forthcoming auctions), the fact that Clearwire and Harbinger may end up competing to attract one of this limited number of partners won’t do anything to push up the price that can be secured for their spectrum in the near term.
Back in March, I was lucky enough to try TerreStar’s Genus phone at the Satellite 2010 conference. At that time it was clear that the phone needed further work to get it ready for commercial service, and recent filings from TerreStar indicate that the company is now working with HNS on an “ATT-QoS Workaround” to address some “APN Issues” (APN or Access Point Node relates to data services, which I did not try to use back in March).
UPDATE: I’m told by a technical expert in this area that the GMR1-3G protocol used by TerreStar treats all information as packet data, including voice (which has the highest QoS). Given that AT&T’s terrestrial network carries voice over GSM and does not normally provide an equivalent voice over packet data service at this point in time, it appears quite plausible that the workaround relates to an attempt to optimize voice performance rather than being an issue for TerreStar’s data services.
However, my concerns about the viability of the Genus phone relate much more to whether the orientation-sensitivity of the phone will actually be acceptable in real world usage conditions. To use the phone you need to know where the satellite is located (roughly southwest when you are on the East Coast) and have clear visibility in that direction. Though that was simple at the Satellite conference, where this direction was out over the Potomac river, it certainly won’t always be the case in rural areas, unless its a desert or a prairie. I still remember only too well the joke I was told by a Globalstar engineer ten years ago – that their system was designed for a “man out standing/outstanding in his field”. More to the point you also need to stand still and not turn around – very different to the situation with Iridium and Globalstar handsets, where the extending antenna goes above your head and allows you to “walk and talk”.
If orientation-sensitivity does prove to be a big problem for potential users, as I think it will, then TerreStar is faced with an unpalatable choice: design a phone with an extending antenna, which will work fine, but would have no mass market appeal, or sell a phone like the Genus, which could conceivably have wider appeal, but won’t provide acceptable performance in satellite mode. Fundamentally, I therefore don’t see any reason to change the opinion I expressed last year during the DBSD bankruptcy, that “the part of TerreStar’s business plan directed to a mass market service is very unlikely to succeed”.
However, there has been one important change in the environment for TerreStar over the last six months, because the FCC has now held out the possibility that 2GHz MSS spectrum holders will be able to participate in an incentive auction, which would potentially allow them to return their spectrum to the FCC for re-auction as terrestrial spectrum without any ATC restrictions. Given the difficulty in realizing value from a satellite roaming business plan, then unless Harbinger negotiates a lease agreement for TerreStar’s satellite spectrum, as part of its planned L-band ATC deployment, it seems likely that this would be the best exit TerreStar could hope for. However, given that the FCC would only give TerreStar a proportion of the proceeds from the auction, and it would probably take a couple of years before that auction even happened (during which period TerreStar will have to raise more money to keep its satellite in operation), it is hard to imagine that the proceeds could exceed the secured debt load that TerreStar has accrued to date. Even if TerreStar did enter some sort of lease agreement with Harbinger (some details of the draft term sheet for the Spectrum Pooling Agreement, which appears to contemplate a “potential purchase of the S-band Spectrum” as one option, but not a takeover of TerreStar itself, have also been publicly filed), then it seems implausible that this payment would exceed the value of SkyTerra’s lease agreement with Inmarsat, which calls for payments of $115M per year, and it could very well be much less. Unfortunately even $115M per year would be insufficient to pay the interest on TerreStar’s secured debt, when it becomes cash pay next year. Remember also that if TerreStar stays out of bankruptcy, it will at some point have to pay Sprint’s spectrum clearing expenses, which Sprint has claimed exceed $100M for each of TerreStar and DBSD.
At this point in time, the future for TerreStar therefore looks pretty uncertain. More importantly for the rest of the MSS market, it is far from clear whether the Genus phone will provide meaningful competition to other handheld MSS providers, and even whether AT&T will actually go ahead with any large scale commercial launch of the handset. I’m sure everyone will be watching with interest to see what news emerges over the next few weeks.