TerreStar and AT&T have now announced that AT&T will distribute TerreStar’s Genus (Elektrobit) phone to “government, first responders, public safety, energy, utility, transportation and maritime users” and given additional details of the expected pricing for this satellite-cellular roaming service. The phone will cost $800 to $900 (subsidies were not mentioned, which is perhaps not too surprising given that the service is not being targeted at consumers initially), and then in addition to a standard voice plan, subscribers will need to pay $24.99/month for satellite access and an additional usage fee for satellite roaming (in the United States) of $0.65/minute for voice and $5/Mbyte for data.
TerreStar and AT&T are basically targeting the existing handheld MSS market, but expect that the phone will be deployed more widely, because it will become an “everyday” phone for end users, rather than being a shared phone which is kept in a cupboard for emergencies. In our view, the size of the potential addressable market is probably not too far off that for Land Mobile Radio, where we estimate there are around 2.5M to 3M handsets in use in the US today, with about half in public safety and the rest in other government and private sector market segments (parks, utilities, railroads, energy, etc.). Of course only a fraction of those users also carry a employer-provided cellphone, and at this point in time, TerreStar’s solution is more likely to be a replacement for that device rather than for an LMR radio itself. So let’s assume AT&T and TerreStar are targeting a potential market of say 1M people. For comparison, today there are about 170K MSS phones in use in the US and Canada (including many Globalstar phones with limited two-way service at present), of which probably just under 100K are in the lower 48 states (and remember that the TerreStar phone is unlikely to work as reliably as Iridium or Globalstar in Alaska or Northern Canada).
In that context, TerreStar would be doing amazingly well if it could gain 50K subscribers by the end of 2010 and 100K-150K subscribers by the end of 2011 (when SkyTerra also expects to be offering next generation services, and Globalstar will be back in full two-way service). Remember also that in 1999-2000, despite massive advertising (at least in Iridium’s case), Iridium and Globalstar only managed to gain a few tens of thousands of users on a global basis in their first year of commercial service. TerreStar will probably share the end user revenue about 50/50 with AT&T (Iridium and Inmarsat expect to get about 70% of retail revenue but AT&T would likely want a bigger incentive to promote the service) and monthly retail ARPUs (including the satellite access fee but excluding any terrestrial voice plan) might be expected to be between $40 and $50. So let’s assume TerreStar receives $25 per user per month. If there are (in the most optimistic scenario) an average of 20K subs in 2010 (assuming deployment ramps up towards the end of the year) and 100K subs in 2011, then that would generate service revenues for TerreStar of $6M in 2010 and $30M in 2011. There would presumably be additional equipment revenues (although we doubt TerreStar is looking to make a profit on equipment sales) and perhaps revenue from some other services like M2M data (once chipsets are available, which means only a limited amount of revenue could be produced even in 2011).
However, its pretty clear that TerreStar is going to need to raise additional funds to cover its operating costs in the near future, since its cash burn rate even before going into full commercial service is about $25M per quarter, and excluding the money purchase agreement for the second satellite (not part of this $25M cash burn), TerreStar had $110M of cash at the end of June 2009. Thus regardless of what happens to the $430M of preferred shares (which must be redeemed or converted next April), TerreStar will need to raise more money by mid 2010. From mid 2011, TerreStar will also need to begin paying cash interest on its debt, at an annual rate of more than $150M per year in 2012.
Given the relatively limited opportunity in professional markets (at best a few hundred thousand subscribers for TerreStar, once the overall market is shared with SkyTerra and other MSS operators), TerreStar will eventually need to either achieve considerable success in consumer markets (requiring a dramatically different price point for both equipment and airtime) or find a strategic partner interested in buying or leasing the company’s 20MHz of spectrum for a terrestrial ATC deployment (which doesn’t look likely in the near term, not least because TerreStar doesn’t even have an ATC license as yet). In this context, its probably most appropriate to look at TerreStar’s initial service offering as a proof-of-concept (and a means to justify further funding) for one of these two longer term possibilities, rather than as the means to generate an immediate financial return on its own. We’ll see over the next 12 months whether TerreStar is able to provide the proof that both the financial markets and potential strategic partners will be looking for.
It seems that people are now coming round to the view, which we’ve expressed since 2006, that there won’t be enough paying users of in-flight broadband for both network providers and airlines to make a profit on the costs of deploying equipment and running a network (as Boeing found out after spending somewhere between $1B and $2B on Connexion). Our view was that only airlines who are interested in offering a differentiated service would be able to justify the costs involved. However, to date the leading service providers (Aircell and Row44) have apparently not only been installing the equipment for free, but have also been offering a cut of revenues to the airline. Its no wonder that this “no lose” proposition has led to fleetwide installation commitments from most of the major US airlines. In comparison, installations of Inmarsat equipment for in-flight cellular services on aircraft in other parts of the world have slowed dramatically over the last 18 months, as most airlines no longer have the money to pay for fleetwide upgrades (with the possible exception of Ryanair, which we suspect may have a similarly attractive deal from OnAir).
Lost in the noise of Southwest’s commitment to install Row44 service across its entire fleet of 540 aircraft was the footnote that there isn’t “a solid timeframe for [installation]” because “certain specific details concerning the cost and financing of equipage are still being worked out”. From what we’ve heard, Row44 needs to raise a lot more money very soon in order to move forward with full-scale deployment (pretty obviously, since fitting equipment on 500 planes at $250K+ each would cost $125M), and presumably Southwest’s announcement was timed to help them secure that funding. However, with Southwest also demanding “control [over] the price point that our customer sees”, it seems a pretty unpalatable deal for potential investors if Row44 must front the installation costs and pay for the network and then let Southwest set the pricing to maximize its own return (probably more dictated by customer loyalty) rather than Row44′s revenues. Similarly unreasonable expectations appear to have been the reason why the oft-mentioned return of Connexion service on Lufthansa (who refused to provide any revenue guarantee to the network provider but wanted to make the provider liable for any future equipment deinstallations) has not happened to date.
What is the solution to achieving a sustainable business model for in-flight broadband? Whether it lies in airlines providing connectivity for free as a differentiator for their customers, or airlines using the link to the aircraft as a means to reduce their own operating costs, what we’re ultimately going to have to see is a change in the direction that the money flows. Instead of airlines getting the equipment for free and receiving a share of the service revenues, the airline is going to have to pay for the equipment and maybe in some cases even offer a revenue guarantee to the network provider (particularly on long-haul international routes where the cost of providing Ku-band coverage is much greater).
How palatable will in-flight connectivity be then to airlines that are currently losing hundreds of millions of dollars a year? At the very least we’d expect them to be a lot more discriminating in deciding whether to provide connectivity or not (who needs it on a one hour shuttle flight?). Perhaps its only if one of the providers goes bust that we’ll see a return to rationality in pricing (of course, it would be very unlikely for the service itself to disappear completely as Connexion did, because the costs of operating either Aircell or Row44′s networks domestically aren’t that high). Until that point is reached, expect airlines to continue to scramble to get something for nothing with their in-flight connectivity installations. In the meantime we’ll be watching carefully to see if the discussions over “cost and financing of equipage” between Southwest and Row44 get resolved and if investors are willing to put more money into in-flight connectivity providers.
Unfortunately its not new services, but the prices of current and future satellite phones and airtime that seem to be headed upwards. The last year has seen Iridium introduce its new, improved 9555 handset at a higher price than the 9505A that it replaced, with phones now selling for about $1500, while Thuraya has “simplified” (i.e. increased) its airtime pricing and introduced the more expensive ruggedized XT phone. Inmarsat admitted in June that its new GSPS handset may sell for up to $750 at launch in 2010, compared to the $500 retail price point it suggested previously. Even TerreStar has now indicated that its new handset may cost up to $800, with airtime pricing at “less than $1 per minute”.
We’ve commented before on how satellite phone revenues have been falling since 2005, and competition has certainly diminished as Globalstar has experienced problems with its two-way services over the last couple of years. However, it seems the consensus amongst current participants in the handheld MSS market is that there is little if any growth potential still left in satellite phones, and the actions of Iridium and Thuraya appear to indicate that their remaining customers are relatively price insensitive.
Even more surprising is that so far, at least, the new entrants do not seem to be particularly keen on shaking up the existing “premium price” paradigm for satellite phones. In the case of TerreStar this is rather worrying, given that their objective is to greatly expand the satellite phone market, and bring satellite-cellular roaming to a mass market, which seems very unlikely to happen with an $800 phone. Is TerreStar simply trying not to give too much away about its future pricing plans, while it focuses on developing all the other elements needed for a commercial service, such as distribution channels, billing systems, etc.? Will TerreStar actually be able to convince a cellular operator to subsidize its phone (which would require a significantly greater commitment from a partner than its current roaming agreement with AT&T)? We should find out soon, as TerreStar intends to launch commercial services at the end of 2009.