Interview: Karim Michel Sabbagh, chief executive, SES
March 6, 2017
SES Chief Executive Karim Michel Sabbagh discusses the Intelsat-OneWeb merger proposal, the near-term prospects for UHD TV, the cost of SpaceX launch delays, and investment in O3b MEO versus expanding the company’s GEO fleet, and O3b’s proposed new constellation.
How do you view the proposed Intelsat-OneWeb merger?
There are a number of questions that need to be answered. I have seen what has been published by the companies. There are a number of open questions. We went through a very steep learning curve at SES when we were thinking through what the model will be to create a scalable, flexible, distributed network. How do you create capabilities around these networks that mirror the market segments that you want to serve. Aligning our business to the markets that we serve takes a lot of effort. There is a lot of work to be done and I think the management and the board of this new entity need to clarify things. I am in no position to add to that.
It’s certainly good news for de-levering at Intelsat. But having Intelsat and OneWeb dance together will be difficult.
Having a mega constellation is not in itself what success looks like; those who look at sheer size miss the point. The long-term winners are those who have a global scale AND look at distinct markets, focusing on one, two, three or four segments that they want to serve, and where they can make a difference and create value for customers. For satellite to remain relevant in the future, you have to invest in these verticals and you have to align everything around these verticals.
I do hope that the discourse in the industry will shift from mega fleets and mega satellites, to the specific markets we can enable.
Do you see any changes for you following the management shakeups of two of your big in-flight-connectivity customers, Panasonic Avionics and GEE?
The short answer is no. First because these management changes are shareholder matters which happen in any organization. And second, the service we are providing to these companies is essential to the agreements they have with their clients — the current connectivity agreements that they are serving, or the ramp-up they committed to. Some of these contracts go through the end of this decade and some of them go through the middle of the next decade.
One of these companies might be weakened enough to be in play now as a takeover target, as part of the consolidation of the sector.
We will find out. The industrial logic around consolidation, either in the satellite operating industry or any of the vertical markets it services will continue to prevail. You could argue that what was announced yesterday [Intelsat and OneWeb] is part of this. Consolidations succeed if they are compatible and accretive from a capabilities standpoint.
Is it too early to plot five years ahead and see a growth in takeup in UHD TV? How confident can you be now on the success of this?
Are we going to see a fast ramp-up? The answer is yes, provided that we figure out the business model where we bring in the smaller players, whether in the DTH space or as part of the second group of cable operators in the United States, which we are doing right now.
You have to be very proactive in creating the ecosystem. In the US we created a platform for the cable operators to be able to ease their transition into it.
Look: It is not going to be as straightforward as SD to HD transition, but what we have achieved so far moving up from 8 commercial channels in 2015 to 21 by end 2016 proves that we are heading in the right direction.
We said the inflection point was going to come from sports. It turns out it came from other areas but sports is going to be there. To illustrate, MX1 has a long term agreement for NFL content distribution. They were part of the distribution operation for the Super Bowl. I would invite you to watch the Hyundai Super Bowl ad. That was made possible by MX1. The 4K plus virtual reality output was MX1’s work.
How sensitive is the 2017-18 to a slip in the scheduled launch dates of SES-10, SES-11, SES-14 and SES-16/GovSat — all with SpaceX?
They certainly matter to our business. Having said that, our assumptions of when these launches are going to happen have built-in contingencies that are based on our experience.
I translate: You have factored in delays in the launches when forecasting the start of revenue generation.
That’s right. You could imagine that our Chief Technology Officer Martin Halliwell is in constant communication with Gwynne Shotwell, SpaceX’s President. Some of these launches are using first-time launch models like SES-10 on the maiden flight of a reusable rocket. We had our team embedded there for a long period of time.
We get updates for our executive committee in a fortnightly manner. This is not a distant focus for us as a management team.
How did EchoStar-23 jump ahead of you in the queue at SpaceX?
They did not jump ahead of us, it is just that the idea that we could eventually jump ahead of them finally did not work out.
Do you have a launch date from SpaceX for SES-10?
The end of March. We need to get this one because if we succeed with a reusable rocket — and we will succeed, as an industry — then we open up an entirely new phase of economics and cadence for launches in the future.
It looked like SES might invest, with partners, in a global Ka-band GEO system for government and aero/ maritime mobility verticals. Now it seems that if you expand somewhere, it’s more likely to be in MEO with O3b. You said you could do things more quickly in MEO with O3b than GEO.
So if we are looking for fleet expansions beyond what you’ve announced so far, it it is more likely to come in MEO, through O3b. Is that a fair assessment?
Certainly we have the O3b constellation, and then we have customized programs like SES-17 with Thales. What I said was that we have a playbook for how to expand this globally and how we want to go about it. It took us a year and a half to get a playbook that made sense to us and to a client like Thales.
At the same time, in October, once we completed the acquisition of O3b, we became much more vocal about payload technologies, digital processing or antennas that make the distinction between MEO and GEO almost irrelevant.
Because of that and because of the differentiated performance of MEO, we don't need to think any more in terms of two different worlds. If we can serve markets that are data-centric in MEO than that is an advantage. That’s where our thought process is.
We continue also to see the opportunity of hybrid technologies on our satellite replacement programs. We have a number of these coming by the end of this decade and we have been transparent about this with the investment community.
But half of your planned capex is uncommitted.
That’s right, it’s uncommitted but in our guidance. And that half enables us to adjudicate whether we want to pursue data-centric satellite programs in GEO or MEO. The latter already has a 5-year advantage vs. competing technologies, and if the next generation of O3b satellite can markedly augment this advantage why not pursuing it.
We are quite advanced on this, and we expect to bring clarity on this front during the course of 2017.
You’ve already started being clear with the U.S. Federal Communications Commission (FCC), where you have filed for added Ka-band frequencies for O3b in Ka-band and a new orbit with higher inclination.
You’re right and that is an option. But I would de-emphasize the higher-inclination now. It is an option but it is not the nominal plan. The nominal plan is how do you augment the fleet that you have today. We are dealing today with a situation where a client with data-centric and sensitive requirements cannot be served anymore on the O3b network. We are working hard to re-allocate other clients out of this beam to accommodate the required surge. So you could appreciate the potential in the medium term to augment the O3b network with the next generation of spacecraft.
The FCC procedure requires that you respect deadlines for construction once you get approval.
Yes, and if the question is will we make this window, the answer is yes and all this will fit well within our capex plan.
But instead of the 16 higher-inclination satellites in the FCC filing we should pay attention to the expanded Ka-band in the current orbit?
No decision has been made. But the one that takes priority is the equatorial one. Augmenting the equatorial fleet is where I think we have the biggest market to serve. Having demand for capacity we don't have is a nice problem, but we need to deal with it as soon as possible.
You have 11 satellites in geostationary orbit that are in inclined orbit to save fuel as they near retirement. You recently sold one to GEE for IFC. Was this one-off or will this be a trend?
It’s not just one-off. It’s going to be one of the important capabilities in our toolkit. We are going to learn along the way how to best commercialize it and combine it to other solutions in our portfolio. Will incline orbit assets replace others? The answer is no because it lacks other benefits you can get with a customized high-throughput payload like the ones we are deploying with SES-14 and SES-15 for mobility over the Americas. It is a great augmentation capability that is very efficient. Net, this is not a one-off thing.
You see growth in the future in your government vertical. You think the EU will be able to do something to federate their governments’ demand?
A: We see the European Commission paying more attention to this topic at a high level. This is a positive sign. Is there a common roadmap? This is a work in progress. We need a common blueprint and it has to be a public-private initiative.
But the commission has no money. Wouldn't it be better to see support in London, Paris, Rome, Berlin, rather than in Brussels?
I was at the annual Berlin Security Conference in November and was pleasantly surprised about the level of knowledge about milsatcom and govsatcom. The Germans and the French were very involved. Germany published a White Paper on German Security Policy and the Future of the German army referencing satellite capabilities as a “fundamental component” of the security and defense policy. And the European Commission in its recent European Defense Action Plan foresees a substantial increase of funding in Security and Defense.
Over the past two years we have stood up market-solution centers and government was one of them. These centers are responsible for developing products and solutions customized to market segments. With this knowledge, we decided to invest more time to help leaders in governments and armed forces to understand the capabilities we can offer.
While the European Commission is the gatekeeper on the topic, the onus is also on the industry to raise its game. We were not there two or three years ago. I’d like to think we are there now.
The Mobile World Congress just ended in Barcelona. Is it your impression that the satellite play on 5G is now better understood by the terrestrial side?
Yes. We moved from no conversations to very engaged conversations. EU Commissioner Günther Oettinger was very engaged and has facilitated the dialogue between the stakholders in the 5G ecoystem including satellite operators. In fact, he championed a series of roundtables including manufacturers, satellite and mobile operators as well as telcos. Commissioner Oettinger has been driving this initiative until he took over the responsibility for the EU budget.
It is going to take time. For decades the terrestrial and space industries didn't talk to each other and this cannot be fixed in a single conversation. Two years ago I went to Barcelona and I was the only satellite executive there. It felt odd. It’s not odd anymore. We are going to have some tensions over the question of spectrum, and it is important that we talk about this matter openly and aim to resolve it in a constructive manner.
I believe in the benefit of sitting around the table and discussing the challenges, use cases and opportunities. Only when discussing use cases we will understand how 5G can work and the impact it can have. The use case that is the most interesting to me and SES is the media one.
Peter B. de Selding