It goes without saying that the forced privatization of Intelsat and Inmarsat, both of which began life as intergovernmental cooperatives, hastened the transformation of satellite telecommunications into the competitive global enterprise it is today.
Another watershed came in 2006, when Intelsat swallowed up its longtime antagonist-turned-competitor PanAmSat, which led the push for privatization in the 1990s. The move re-established Intelsat as the biggest player in a rapidly changing industrial landscape.
Washington-based attorney Phillip Spector has been a key player in the industry’s transformation, first as a hired gun for PanAmSat’s crusade. In 2005, he took up with his former adversary, joining Intelsat as general counsel and executive vice president for business development.
By that time, Intelsat was four years into the commercial era but still had a reputation for being a highly bureaucratic, if technically capable organization, Spector says. But Intelsat was already changing, and acquiring the entrepreneurial PanAmSat accelerated the process, he adds.
Spector, who left Intelsat in 2013 to join the space practice of the law firm Milbank, Tweed, Hadley & McCloy, spoke recently with SpaceNews Editor Warren Ferster.
Can the satellite industry maintain the growth it has enjoyed over the last decade?
The game is going to have to be played out on some new playing fields. A lot of what drove the success the satellite industry’s had over the last 10 years were large orders by the U.S. government spurred by two wars. A second factor was telecom services. Many of those services are now migrating to fiber, and fiber is finding its way not just to the developed world but to the less developed world, to places like Africa which did not have fiber 10 years ago. Direct-to-home television is still a big factor for this business but over the next 10-20 years that could change too as the models move toward the Internet and more toward fiber delivery of video. The satellite operators need to find other areas of growth. I think they’ve done a good job of that by starting to look at mobility applications and high-throughput satellites, but it remains to be seen whether they will discover the appropriate new business model that will fuel future growth at the same level they had the last 10 years.
You’ve said satellite technology hasn’t changed dramatically in the last 30 years. Can you elaborate?
Satellites have certainly gotten better but you have nothing comparable to what’s happened in telephones where we’ve gone, in 30 years, from having a desk phone to today having a mobile phone, an iPad and a multiplicity of mobility and other options. Satellite technology has been relatively slow in advancing partly because this industry builds for the long term and is fundamentally conservative. If Steve Jobs made a mistake with a version of the iPhone he could correct it very quickly in the next version. It’s not so easy to do with satellites.
What about onboard processing?
Onboard processing has been new but many of the applications for it haven’t really been used in the way that perhaps they could have. The changes are going to come with more of these high-throughput satellites. ViaSat has them, as do Hughes and Eutelsat. Intelsat is building the Epic satellite system, you have Inmarsat with Global Xpress and you have O3b. These are all potentially game changers in terms of technology but none is really proven yet. These are all examples of satellites that may use onboard processing and other techniques to meaningfully change how you push bits through a satellite pipe.
What’s different about Epic?
Intelsat is building a system that takes advantage of advances in how you can move bits through a satellite pipe and get more throughput, which is what the customers are demanding. They also looked at spectrum and realized that you do not need to be in Ka-band to do many of these applications. In fact, C-band and Ku-band may be better in some situations because they’re lower in the frequency range and less susceptible to rain, which is a big issue particularly in tropical and equatorial regions. On the other hand, there is clearly a place for Ka-band and the market will decide over time.
Why are companies like Inmarsat placing such big bets on Ka-band?
When Inmarsat decided that it needed a response to what Intelsat and others were doing, Inmarsat really could not make that response in Ku-band because there are no good Ku slots left in the world. So they had to move to a higher-frequency band where there were slots available. That’s also true for O3b, Viasat and Hughes. They really had no choice but to go to Ka-band. Now they’ve done a good job of making a virtue out of necessity but it really was necessity.
Has the hosted payloads concept been more talk than action?
So far yes, it has been. But that’s mostly because the government has not been able to get its act together. The case for hosted payloads is very compelling, particularly in this time of challenged military budgets. But the government has to change some of its procurement practices and move more quickly. The essence of hosted payloads comes down to a metaphor: There’s a train that’s going to be leaving the station and if you want to ride on that train, you had better be there on time and ready to go. The government is the passenger, but it never seems to get to the station at the right time to meet the train that’s going in the direction it wants to go.
Might the U.S. Air Force’s new contracting vehicle for hosted payloads make a difference?
It’s great that they are open to the idea. I think that in terms of getting from that broad contract approach to actual contracts that yield payloads on commercial satellites, we still have a long way to go.
Why so skeptical?
It’s very hard to change the military’s procurement practices. The military first of all has a vested interest in maintaining current programs and when budgets are cut the first reaction is to hunker down and protect those programs rather than to try to deal with a new reality and save the taxpayers’ money. I do think that at the end of the day there will be more hosted payloads, but it’s pathetic that the last major hosted payload was an Australian order with Intelsat in 2009. The U.S. government in that case couldn’t get its act together to buy the payload directly from Intelsat, and ultimately it bought the same capacity through the Australians because it was easier. That’s a problem.
Intelsat funded a UHF payload on its IS-27 satellite but couldn’t get the Pentagon to commit to buying capacity. What happened?
The government had given very strong signals to Intelsat in terms of conversations with Navy and other officials that if Intelsat put the UHF payload on the IS-27 the government would buy it. So Intelsat took a chance and spent a lot of money to place that payload on the satellite. The government then began backing away from the idea, not necessarily because they didn’t need it — I think they did need it — but because they were protecting other programs and felt threatened at the idea that a commercial operator might be operating in their spectrum.
Might this experience deter other operators from taking a similar risk?
It certainly would keep me, if I were at one of the satellite operators, from going forward on the “if you build it they will come” theory. What I would insist upon instead would be a contract in advance that says “if I build this payload the government will sign up for x amount of capacity over y number of years.”
Doesn’t the government face legal or policy hurdles to doing that?
Yes, but it’s got to find a way to get over those. Somehow they’re able to make a commitment that gives, for example, Lockheed Martin the confidence to build a factory to construct fighter jets. The same people who figure that one out ought to be able to figure out how to make multiyear commitments to buy satellite capacity.
Is the use of government spectrum going to be a challenge for Inmarsat’s Global Xpress system?
I think Inmarsat will face some of the same challenges. They’re assisted in the fact that they laid off some of the risk to [Global Xpress satellite manufacturer] Boeing and that was very smart. Boeing is now trying to market to the government and I don’t know what challenges they’ve faced but I think the use of government spectrum will be part of the challenge. But if the government really needs the capacity the government will find a way to buy it. That’s what we thought would happen with IS-27 eventually. [Editor’s note: IS-27 was lost in a launch failure.]
Do you envision a day in which the U.S. military procures most or even all of its wideband satellite capacity from commercial operators?
In the long term I think all but the most hardened communications will go over commercial satellites. It does not make sense for the military to build satellites on its own at the very high cost that they pay in order to meet relatively routine communications needs.
China and India are viewed as having tremendous potential yet both markets have been all but impossible to crack by outside operators. Do you have any sense that this will change?
The arc of history is long and at the end of the day the current systems in India and China will not survive. And I don’t just mean in communications and communications regulation but more generally. It may be a very long process but as these countries are forced to open up, as is happening already with the Internet, with mobile phones and the like, there will be changes and some of those changes will lead to more need for satellite communications that the governments cannot meet on their own. They have some terrific space technology. What they don’t have is the ability in a private marketplace to utilize that appropriately and grow it. They have to take their orders from centralized planners, and that system never works.
Do you expect to see much in the way of industry mergers and acquisitions in the next few years?
I think we will see some of the smaller regional operators sold, but many of them may also remain independent or form their own sorts of regional consortia. A good example of regional operators getting together would be the Arabsat-Hellas Sat deal that happened last year, and we may see more of those deals as well as the major operators buying smaller operators.
How many mergers might we expect in 2014?
I would say the range is zero to two. There are probably three or four companies that may be put into a sales process during 2014, but not all these sales processes have been successful.
