Viasat Inc (VSAT) 2017 Q3 法說會逐字稿

Welcome to ViaSat's FY 2017 third quarter earnings conference call. Your host for today's conference call is Mark Dankberg, Chairman and CEO. You may proceed, Mr. Dankberg.

Okay, thanks. Good afternoon, and welcome to ViaSat's earnings conference call for our third quarter of our fiscal year 2017. I'm Mark Dankberg, Chairman and CEO, and I've got with me Rick Baldridge, our President and Chief Operating Officer, Shawn Duffy, our Chief Financial Officer, Keven Lippert, our General Counsel, and Bruce Dirks, our Treasurer.

So before we start, Keven will provide a safe harbor disclosure.

Thanks, Mark. As you know, this discussion contains forward-looking statements. This is a reminder that certain factors could cause actual results to differ materially. Additional information concerning these factors is contained in our SEC filings, including our most recent reports on Form 10-K and Form 10-Q. Copies are available from the SEC or from our website.

With that said, back to you, Mark.

Okay. Thanks, Keven. As usual, we'll be referring to slides that are available on the Web. And we'll start with some financial and business highlights, and then Shawn will go into more depth on the financial results. Then I'll come back and give a little more color on our business and discuss our outlook, and then we'll take questions.

I want to start with a little bit of a recap on revenue and EBITDA margin trends over the last ten years. We've got a chart at the top of this page that shows that, including year-to-date for fiscal 2017. We excluded the non-recurring portion of the Loral ViaSat-1 settlement in fiscal year 2015.

So we've had steady revenue growth even as our business mix has evolved, and after we first sold the bandwidth on ViaSat-1, we've become more services-oriented and we've invested in vertical integration that puts us in close contact with our end users over a bigger portion of our customer base. That gives us greater control over the quality and value of our offerings and more sensitivity to the continual changes that are inherent in rapidly-evolving Internet-driven markets.

Until this year, when we ramped up work on ViaSat-3, we've been steadily growing both EBITDA and EBITDA margins on a Company-wide basis. So just to recap, we believe that anticipated bandwidth productivity of the ViaSat-3 satellites is pretty transformation and should enable pretty substantial competitive advantage.

To achieve that potential, we've developed a new and different payload architectural framework that we think has a faster and steeper learning curve than the industry's been on. One of the artifacts of our choice to build those payloads is that the pre-flight portions of the design and construction work is expensed as R&D instead of capitalized, and that's what shows in our results for the quarter and year-to-date.

Even with the decline in EBITDA margins, though, due to the incremental R&D shown by that orange line in fiscal year 2017, our year-to-date revenue growth has yielded record absolute EBITDA dollars for the period, so that's translated into over $300 million in operating cash flow year to date, with satellite services contributing the most. We've reinvested all that operating cash flow and more into our ViaSat-2 and ViaSat-3 programs.

Our ongoing assessments of our target markets and the feedback we've gleaned from those markets, especially consumer broadband, in-flight connectivity and government mobile broadband, has only increased our confidence in the opportunities created by these new satellites. And I'll give a little more color on that later in the call.

We're aiming to achieve our growth by growing the broadband verticals we've already successfully entered by adding new verticals and by extending the existing and new ones to new geographic areas. We've shared our view of Europe as the next best market after the US, and that's why we've worked with Eutelsat to create a joint venture to grow the broadband markets there. We expect the first step in that JV to leverage Eutelsat's existing KA-SAT satellite to close by the end of this month, and we both intend to extend that agreement to our upcoming European region ViaSat-3 later this year. I'll give more color on that later too.

Finally, the ViaSat-2 satellite is complete, and it's waiting launch via Arianespace on April 25. I'll give more color on ViaSat-2 and the progress we're achieving on the ViaSat-3 satellites after Shawn's financial presentation.

Thanks, Mark. We're seeing excellent top-line results in our satellite services and government systems business segments. Slide 4 shows revenue and adjusted EBITDA by segment for the third quarter of fiscal 2017 compared to the same period last year. Revenues were up $33 million, or about 9% from the prior period, with growth in service revenues slightly outpacing the growth in our product revenue base.

In satellite services, revenues reached another record of $160 million due to a combination of higher RPU in our residential subscription business and higher (inaudible) counts and revenue in the commercial air business. These growth drivers, along with the existing scale of our consumer service business, allowed us to convert roughly 69% of our year-over-year top-line revenue growth straight into adjusted EBITDA.

In our residential business, our mix of premium subscribers increased by nearly 60% from the year-ago period, which helped drive RPU to a new record in addition to showing higher customer satisfaction trends and corresponding lower churn on a sequential quarter-over-quarter basis.

In commercial air, we have 109 more aircraft in service compared to the same time last year. And just as we've seen increasing bandwidth usage in our consumer business, we're seeing similar trends in IFC as our partners broaden the way they utilize our high-capacity satellite solutions for their airline customers, such as JetBlue's recent addition of gate-to-gate Internet service.

In commercial networks, our revenues were flat, with good opportunities to resume top-line growth in the second half of next year reinforced by our recent commercial air wins and more than 750 future air terminal ship sets now under contract.

Looking to the bottom line, our segment adjusted EBITDA loss widened, reflecting the higher expense to R&D primarily related to the ViaSat-3 program Mark just discussed as well as investments in expanding our mobility products portfolio.

As discussed in our previous calls, we expect to see R&D continue to grow a bit in the next few quarters as we accelerate SDC activities in support of the recent commercial air wins and continue our payload development activities.

Our government systems segment also delivered a strong quarter despite the uncertainties surrounding the continuing resolution environment. New orders came in at 152 million, and our backlog now tops out over 650 million, a 60% increase year-over-year.

Revenues grew 10% year-over-year to $166 million, and was heavily driven by growth across our satellite networking, data links and encryption product base businesses, coupled with continued strong performance from government mobility service offerings, altogether generating $40 million of Q3 adjusted EBITDA, a 7% increase year-over-year, which was slightly lower than the revenue increase due primarily to the higher portion of lower-margin product sales this period.

Overall, we're very pleased with our Q3 results. We had good top-line revenue growth, contributing an additional 15% to year-over-year adjusted EBITDA performance offset by a $15 million increase in R&D versus the same period last year.

As we look forward, we are sitting in a very strong position with Company backlog of over $1 billion, up 23% from the same period last year. And keep in mind, this excludes the recurring monthly services we also get from our residential and mobility broadband customers.

Slide 5 shows revenue and adjusted EBITDA performance for our fiscal 2017 year-to-date period compared to the same period last year. Many of the key drivers are similar to the third quarter results I just discussed, so I won't spend too much time on that slide, but I would like to highlight a number of new records we achieved for the one-year period, including record awards of $1.3 billion, record revenues of $1.1 billion, and record adjusted EBITDA of $257 million.

As we look to our segments, performance to-date mirrored our Q3 trends, with our commercial segment reflecting the significant R&D spend offset by record performance in both revenues and earnings from our government and satellite service segments.

Turning to slide 6, our income from operations for the quarter was down, due primarily to the $50 million increase for the R&D in the period, partially offset by the strength of our underlying business segments. On a year-to-date basis, operating income was essentially flat, but, again, our top-line revenue growth contributed strongly to our operating income year-over-year, nearly doubling amounts for the same period last year, offset by the increase in the R&D investments I spoke about earlier.

Looking at our net income and EPS performance for both the quarter and year-to-date periods, we see the flow-through impact in operating income I just discussed. Plus, in Q3 of last year, we also had a $6-million, or $0.13 per share, fed R&D credit benefit due to the retroactive reinstatement of that legislation in December 2015. So in summary, taking that catch-up tax benefit out, our non-GAAP earnings per share was up slightly in Q3 and up 11% on a year-to-date basis.

Regarding cash flows, our year-to-date cash from operations hit a new record of $313 million, reflecting growth of $93 million due to positive working capital changes and improved adjusted EBITDA performance.

With respect to CapEx, expenditures were up approximately $124 million from the prior-year period, with the majority of this increase attributed to our ViaSat-3 satellite and the ViaSat-2 program, including the associated ground segment. Total spending in the period across the three projects was about $209 million so far this year, or roughly double the level spend during the same period last year.

Our November 2016 equity offering of approximately 7.5 million shares generated net proceeds of $503 million after expenses. We've used $225 million of the proceeds to pay down the outstanding revolver balance, and the remainder remains fully available to fund our corporate activities such as future satellite construction payments and our European broadband investment with Eutelsat.

The financing has also significantly lowered our leverage position, and it's given us quite a bit of flexibility surrounding our ViaSat-3 global network buildout, especially the pacing of our future Asia-Pacific ViaSat-3 satellite. So we ended the quarter with $285 million of cash, no borrowings under our revolver, and $275 million outstanding on our $387 million Ex-Im Bank loan commitment.

Overall net leverage was 1.8 times trailing 12-months adjusted EBITDA, and was nearly half of the 3.1 times in the immediately preceding quarter. Our overall liquidity position is really strong, with combined availability of over $1 billion, which includes revolver capacity, remaining Ex-Im loan commitment, and the cash on our balance sheet.

So with that, I'll turn it back over to you, Mark.

Okay. Thanks, Shawn. So I'll go into some more detail on some of the points we talked about upfront. This slide has a set of charts that both illustrates how we've continued to drive revenue and EBITDA growth on our existing satellite fleet and the exciting growth opportunities enabled by the upcoming ViaSat-2 launch.

We've said over the last several quarters that consumer satellite services' growth has been driven by offering higher-priced, higher value premium plans and by non-bandwidth-consuming value-added services such as Wi-Fi, voice-over-IP and easy care service offerings.

The single most popular option has been an upgrade to 25-megabit-per-second speeds that we've introduced in selected geographic markets. While bandwidth constraints still limit the availability of this option, you can see that growth there is accelerating.

The chart on the lower left is one we've used before, and it, for context, showed, as of September 2014, an estimate used by the FCC chairman of the competitive environment across the US for different offered broadband speed tiers. While only about 7% of the market had no other option for a 10-megabit-per-second service, you can see that about 20% of the market had no other choice for a 25-megabit-per-second service. That's like triple the opportunity there.

Speed continues to be the single most desirable attribute of broadband service. At 25- to 50-megabit-per-second speeds, 75% to 80% of the market was estimated to have either zero or one choice for an ISP, compared with only 40% of the market with zero or one choice at 10 megabits, so clearly, moving up market created the opportunity for more favorable competitive dynamics. Our success in doing so is shown in the upper right chart.

We've steadily increased the proportion of our subscribers on premium and mid-tier plans. That's what enabled us to generate growing revenue and EBITDA from our existing fleet, and our focus on providing better service to fewer customers increases our confidence that we can use ViaSat-2 to drive attractive net subscriber growth when that enters service. There's just no reason to sacrifice profitability now when the resources to be more aggressive on subscriber growth will shortly be on the way.

We constantly monitor our subscribers to measure customer satisfaction and its relationship to voluntary churn. Not surprisingly, those two are very correlated. Happy customers churn much less often.

The chart on the lower right shows the relationship between bandwidth volume caps and churn. Our Freedom plan is like an uncapped plan. It has the greatest customer satisfaction and the lowest churn rate. So the way we think about it is that high speeds are the curb appeal that attracts new customers. Then taking away fear of usage caps is the key to having them stay longer.

The lower right-hand chart shows why we're so focused on bandwidth productivity. With better bandwidth economics, we can demonstrably deliver more value to our subscribers by sharing our productivity gains with them. With higher speeds, we can operate in a more favorable competitive environment. The combination is really exciting.

And finally, while going to 25 megabits is already delivering better results for us compared to our 12-megabit original ViaSat-1 service, we're really excited by the ViaSat-2 network that enables speed of 50, 75 and up to even 100 megabits per second on a scalable basis.

So turning to commercial air, during the third quarter, we increased our backlog of commercial aircraft substantially, especially through our expanded American Airlines relationship. We now have 555 aircraft in service, and a total backlog of 750 more to be installed through about the middle of calendar 2019.

Our most fundamental strategy is to enable our airline partners to use an on-the-ground broadband experience available to every passenger and even the flight crew as a competitive weapon. We aim to help them leverage connectivity to improve passenger satisfaction, to drive preferences on competitive routes, and to craft unique and compelling opportunities to increase revenue, as well as offset or avoid costs in other areas such as in-flight entertainment.

That was the logic behind our acquisition of Arconics, a small but growing player in automating flight operations and in wireless in-flight entertainment. Arconics has been an early leader in automating the connected aircraft, but their customers were lacking the cost-effective connectivity to maximize the value of those capabilities.

Arconics also brings a proven, scalable wireless in-flight entertainment system for both stored and live content that we can seamlessly integrate with broadband-connected Internet media offerings. That combination is enabling American Airlines to realize substantial cost savings by delivering state-of-the-art in-flight entertainment from any of a number of sources directly to passenger devices on their new 737 MAX suite. It's an example of how we can create greater variety, optionality and opportunities in passenger entertainment while meaningfully reducing capital and-or recurring costs.

That's a unifying theme among the airlines that have chosen us as their in-flight connectivity partner. They're interested in creating and monetizing greater passenger engagement using the whole Internet and all of its exciting new media services and business models, versus merely offering limited text, email and Web to a small faction of passengers per flight.

Remember that those airline passengers are basically the same people that use the Internet on the ground, so it's not surprising that the same formula that works for consumers at home is what drives user satisfaction in the air. Higher speeds and more bandwidth at lower cost is the only way to meet steadily rising expectations. We believe satellite productivity, high-capacity density in key markets, and geographic coverage flexibility are key factors creating enduring competitive advantage.

2017 should be a good year for us as we see more airlines introduce new and creative ways to use our connectivity in the US, Europe and Australia. The launch of ViaSat-2, our European joint venture and our partnership with the national broadband network in Australia all will expand the reach and resources for on-the-ground connectivity.

We've got additional opportunities to capture airlines that see in-flight connectivity in this fight. We believe that segment of the market is expanding. Our ViaSat-3 constellation is pretty compelling, and we're committed to making our airline partners successful.

Our government business continues to perform exceptionally well. It's setting new records for orders, revenue and adjusted EBITDA. While timing of individual programs and product lines remains somewhat lumpy, the aggregate is performing pretty well. We believe we have a leadership position in most of the markets in which we participate, and we continue to build on a track record of delivering profitable growth and order momentum in a market that's been a challenge for many others.

We continue to see strong, multi-year growth opportunities in mobile broadband services, tactical data link and cybersecurity. Our mobile broadband services' value proposition is certainly enhanced by increased geographic reach. Our international partnerships help in the short term, with the global ViaSat-3 constellation driving substantial interest for both existing and new government customers. Our recent capital raise is an important factor in being able to complete the global constellation with an Asia-Pacific satellite sooner than we otherwise would have been able to.

And we're also pleased to be earning follow-on business from existing customers as well as initial tests and trials with new customer organizations and user groups in each of the main growth areas of our government segment. The substantial backlog we have built helps lend confidence to government systems being a significant contributor to growth for the balance of fiscal 2017 and on into fiscal 2018 too.

So we're really pleased to have reached the agreement to close the previously announced joint venture with Eutelsat. At the closing of the joint venture, we'll create two new entities. ViaSat will invest about EUR132 million to acquire 49% of the KA-SAT-based infrastructure entity, including KA-SAT's existing wholesale business. We're also creating a new retail entity, 51% owned by ViaSat, to drive end user retail business modeled after our successful US venture. We're both focused on growing KA-SAT revenue and earnings via residential and mobile broadband.

We both consider this the first step of a broader agreement in which Eutelsat intends to invest in ViaSat-3 as their next KA broadband satellite for the Europe, Middle East, Africa regions. We expect to conclude that step later this calendar year. We've got a long partnership history with Eutelsat. We believe the JV agreement will help us each achieve our business and financial goals. Getting a running start with KA-SAT and partnering with Eutelsat will be two powerful ingredients in helping us achieve greater success with that portion of our global constellation.

To give a little update on the satellite platform, we've got high expectations for a competitive advantage due to these new platforms. We're happy to report that the ViaSat-2 satellite construction and test is complete. It will ship to the launch site next month for a scheduled April 25 launch date.

We've also been making excellent progress on integrating and testing our new advanced ground network with the satellite payload. Just like with terrestrial mobile systems, the end user throughput of broadband satellite is constrained by the number of network access points times the bandwidth per access point. So think of mobile wireless -- that's the number of cell sites, and for satellites, it's the number of gateway teleports.

Just as mobile operators are driving to smaller, less expensive cells to improve coverage, total capacity and capacity density, satellite systems must to the same thing with gateways. There's no way to get terabits-per-second throughput from a broadband satellite without also multiplying the gateway access network. Current gateway architectures make that prohibitively expensive. That's an example of one of the system problems we've addressed with ViaSat-2 and that we're extending with ViaSat-3.

The blue outline box in the lower right-hand corner shows the relative size of ViaSat-2 gateways, compared to the rest of the industry's state-of-the-art further to the left. ViaSat-2 gateways are much smaller. They only require utility cabinets instead of dedicated buildings for their local hardware. They support more spectrum and are much less expensive to maintain and operate. They're also designed for higher reliability and tolerance to terrestrial network outages and weather effects. Since last quarter, we've made great progress in testing and demonstrating all these critical network features.

We're also making good progress on ViaSat-3. We're still in the pre-flight hardware phase, but we're progressing well towards the actual payload construction. The successful preliminary design review confirmed compatibility and system-level performance of the spacecraft subsystems. We've been building and testing pre-flight versions of hardware and testing them for system and environmental performance. So far, so good. We've retired a significant amount of program risk while still maintaining our overall performance, schedule and budget objectives.

Okay, so I'll finish up with a brief update on our outlook and business drivers. Overall, there's not much change in our business outlook for the fourth quarter of our fiscal year 2017. There will be some new business dynamics starting in our fiscal 2018.

Overall, we see good prospects for continued revenue growth, that we've already described, across all of our business segments, comparable to what we've achieved this year -- maybe even better. Some of that is due to the backlog and competitive success we've had in the government area and in the in-flight connectivity market.

Of course, we anticipate that the launch of ViaSat-2 will result in some good satellite services revenue growth and indirectly catalyze growth in government and our commercial segments too. Although the incremental revenue growth should carry margins that are comparable to current levels, the aggregate adjusted EBITDA margins will see some counterbalancing pressures from a couple of factors, including the completion of payments from Space Systems Loral related to our ViaSat-1 settlement as well as the timing and transition effects of launching the ViaSat-2 services.

We expect ViaSat-2 to enter commercial service in the last calendar quarter of 2017, which is our fiscal year 2018 third quarter. We also expect adjusted EBITDA margins to recover and grow nicely as satellite services continue to grow in fiscal year 2019.

So that's it for our prepared remarks, and we'd be happy to take questions now.

Thank you. (Operator Instructions). Rich Valera, Needham & Company.

A question on the consumer business. You've had a very nice trajectory on the RPU there the last few quarters, up about 2.5% quarter-over-quarter, each quarter. I was wondering how you think about that RPU longer term. First, sort of the next few quarters before ViaSat-2 gets into service, and then how do you think it can trend once you get ViaSat-2 into service and you can offer, say, 50- and 75-megabyte plans?

Yes, so let me -- basically, we expect the current trends to continue, driven by the same factors that have. There's going to be a bound to it. I think we're not there yet. When we get to ViaSat-2, we do think there's going to be demand for the really high-speed service, and that'll introduce a little bit more upside to it.

One other factor I would say is that we've migrated our mix from retail to wholesale, so we have a much greater proportion of retail than we used to. I think when we launch ViaSat-2, we may end up with faster-growing wholesale, and that wouldn't affect the RPU of the retail component, but the wholesale retail mix may moderate overall RPU.

Got it. And then a similar question, but on the in-flight business. The first several hundred planes that you took onboard or through LiveTV and, presumably, they're getting a cut there and are at certain terms, and now you're going to start adding a lot of planes from American and others that are -- you're the service provider, there is no sort of cut being taken, presumably. So I'm just wondering how that is likely to affect the average revenue per plane. Should we expect that to be trending up as you start adding planes that are where you're the primary sort of service provider to the end customer?

Without going into the details of our contract, that's not -- I don't think that's really going to be the driving factor. I think there's going to be a couple of balancing -- a couple of offsetting trends. One is that we think the passengers really like the service. The airlines are learning more and more what the value -- how to capitalize on the benefit that they bring to those passengers. And besides that, you've got the general trend on the Internet, which is the same activities you do this year use more bandwidth than they did last year. There's just more video embedded in things like social media and Web pages and email attachments. Everything is trending up for bandwidth.

On the other hand, one of our main advantages is that our customers get to tap into our fleet, which has improved productivity and lets us reduce the bandwidth, and so I think that there will be a trend probably towards more revenue. I think it'll be mostly driven by greater bandwidth usage. That overcomes kind of the other factors. I think that's what the driver will be, and I think that greater bandwidth usage will come from more penetration of a little more bandwidth per user.

And then also we think that more and more, the airlines are seeing the benefits of tapping into Internet-based media as opposed to purchased broadcast media or storage media and that that's going to enable them to save money overall, but it'll probably increase bandwidth usage.

Got it. That makes sense. Thank you, Mark.

Thanks, Rich.

Ric Prentiss, Raymond James.

A couple questions. First, congrats on getting the first step of the Eutelsat joint venture. Obviously, some questions, then, surrounding the second step, where you both intend to move forward with it. Can you help us understand -- they said on their webcast this morning bright and early that they were anticipating maybe contributing about half of the space and ground component and that there might be a total cost of that for the ViaSat-3 being of about $650 million. How should we think about the timing and the size of kind of what their contribution might be? And it was asked this morning if there would be any non-cash component to it.

Okay. So, yes, I think what you heard was them responding to specific questions. What I'd say is that, overall, it's a little bit of a complicated agreement. There are a lot of dimensions to it, and I'd say it's not -- it doesn't really make sense to just single out that one capital component dimension for a discussion.

I think overall, they talked about what their objectives are, which was to fit within their capital budget, to be able to preserve their EBITDA margins with an infrastructure-based business, to get to a really attractive -- the number they cite is $1 million -- or EUR1 million, I think, per megabit. So all those things -- so we can help them achieve all those things that they're going to -- it's going to be an overall comprehensive agreement, and I'd prefer to wait until we have that agreement to go into details on what each component will be. I think that'll -- but I think it's going to be a good -- we're confident it'll be a good agreement for us and will be a good agreement for them.

Right, and they this morning thought it could be, by quote, year-end. You're just saying later this year. Should we think about roughly when you might be able to update us on that?

Yes, I think that's the plan, is by later this year, this calendar year, that we should be able to complete that agreement.

Okay. And my second question is, on slide 12, you talk about kind of the R&D levels pre ViaSat-2 and post ViaSat-2. On the post ViaSat-2, it says on the slide network R&D will stabilize relative to overall business. How should we think about that in terms of dollars or percent of business? Because, obviously, it's elevated now, but just trying to think of scaling it looking forward as the company moves beyond the ViaSat-2 days and even beyond ViaSat-4 days.

Hey, Ric, this is Shawn.

Hi, Shawn.

One of the things I kind of mentioned as we were talking through the script is that we expect R&D to uptick a little bit over the next couple quarters, and then as we go into next year, on the back end, it should level out a bit. I think one of the things we're looking at is opportunities and timing of the APAC satellite, so we'll update you if there are any drivers there, but -- and we'll expect it to grow for the next few quarters and then maybe even out, probably on the next -- on the back half of the year.

Perfect. That helps.

Ric, this is Rick. It should come down after that. I mean, what we're -- we're doing two things -- one, the pre-production hardware that's for ViaSat-3, and we'll have some more, some additional non-recurring for the third APAC satellite, like Shawn said. We're also spending quite a bit of money on the ground network, as well. So what we're doing on ViaSat-2 is useful in ViaSat-3, and then a lot of what we're doing on ViaSat-3 will be useful on all the future satellites, so once we get through that phase, the percentage of R&D kind of to revenue line should come down.

Makes sense. Thanks, guys.

Thank you.

Andrew DeGasperie, Macquarie.

First, I was wondering if you could maybe comment on (inaudible). Do you see any potential disruption to your (inaudible) aircraft agreement there? And then, secondly, Shawn, can you maybe walk through the guidance for this system here (inaudible)?

Okay, I got the first part. I think the first part was a question about [Talus] and [FDS17], and we don't see a disruption to our existing in-flight business based on that. The second question, I didn't hear. Could you repeat that?

Sorry. I was asking about the guidance for fiscal 2017. How is 4Q likely to shake out given the last few quarters, on EBITDA specifically?

Yes, I think as Mark mentioned kind of during the call, that we think it's holding pretty steady to what we told you guys before for the year, for 2017.

Okay.

Our fiscal 2017 ends in March.

March.

Yes.

I don't know if that was the year you're talking about.

This fiscal year, specifically. I just wanted to know, based on the trends in the businesses and the uptick in R&D, do you think you're comfortable meeting your year-end guidance?

We're not giving specific guidance. We haven't for a while.

Okay. Thank you.

Giles Thorne, Jefferies.

I had a couple of questions, please. In the first one, I wanted to play devil's advocate for a bit, so bear with me. So given your evident, incredibly high conviction around the relative competitiveness of ViaSat-3, it seems that you just had -- Eutelsat is walking away with more from the JV than you are, and the kind of thinking behind that is that once they've entered into the JV, they're going to have control over the EMS satellite from the ViaSat-3 constellation for roughly half the CapEx cost of that satellite, but they'll have control over about how and where it gets sold.

Now, I appreciate you control the retail co, but at the infrastructure co-level, Eutelsat can kind of do what they want and, in effect, they could even go so far as to sell to your competitors should they want to do so. So I know you've laid out previously why you chose this path as a bridge head into other markets and why Europe is attractive, but, again, why did you not back yourselves just to go it alone, I suppose? Sort of just revisiting that thinking. And then after the US one, I'll leave my second question.

Okay. Well, just on your first question, okay, so it's a good question. I think that that's -- your question sort of highlights why we think it's really important to consider the ViaSat-3 portion of the joint venture once we've reached agreement on all of the terms. We have a preliminary agreement with them, but it's different than what you described. I think rather than go through all the differences, I think it's best to wait until we have a definitive agreement, and then we'll both talk about it then. But it's not the -- I think that your question would be a fair question if it were the way you described it.

Okay, so we will wait for a bit more detail.

Yes. Yes.

Okay, fair enough. Fair enough. Second question was around HTS LEO and the risk from HTS LEO. Should it happen has been discussed more robustly of late, and I don't think you've ever really addressed it in this forum, so it'll be interesting to hear your comments on the technical feasibility of HTS LEO. And then assuming all the various hurdles do get cleared and some of these constellations do come to pass, it'll be interesting to hear your perspective on actually how much latency really does matter as a buying decision in your chosen application areas of residential broadband and mobility.

Okay. Yes, I've talked about this somewhat in the past. I'll give the highlights. One is there are a lot of challenges with building the LEO systems. There are a lot of different LEO systems that are proposed. Probably the one that is closest to reality is OneWeb, and I think OneWeb faces a bunch of challenges, but they're not insurmountable. I don't think I would bet against -- it's not that we would say it's not going to work or we'd bet against it. I think what we're doing is a better investment. That's our view. We've had a lot of involvements with kind of every flavor of non-geosynchronous satellite systems, and I think we have a good understanding of them.

And officially, just to put things in perspective, in 2020, you're looking at probably well over 1000 terabits per second of access bandwidth demand globally, and both we and OneWeb would have single-digit amounts of terabits, so I don't think it's a fight to the death between us, as an example. I do think that we have some very fundamental differences in our go-to-market approaches. Number one is our own experience in selling consumer broadband is that latency is not the -- it's not even one of the top two dominant performance attributes for consumers. The top two are clearly speed and volume of bandwidth, and it's really driven by video.

So for us, the thing that we've been very focused on is driving down the cost of bandwidth so that we can offer high speeds, and especially, essentially, uncapped plans. As I've described before, that's a big factor in customer satisfaction.

But the other really big differentiator between what we're doing in a low earth orbit system especially is the geographic distribution of bandwidth. So with a low earth orbit system, your satellites are distributed uniformly in orbit space. It's not exactly the same as uniformly on the surface of the earth, because they have lots of bandwidth at the poles, but the bandwidth as you get closer and closer to the equator -- bandwidth available in any geographic market becomes closer and closer to the geographic -- to the fraction of the earth that that market represents.

So as an example, the US is about 1 -- the continental US is about 1% of the surface of the earth, so it turns out, depending on -- a lot of this depends on the details of how they measure capacity and the services that they deliver, but it's on the order of 1% to 2% of the spacecraft, and in a low earth orbit constellation, that can serve the US. Likewise, the same for Western Europe or other high-demand markets.

So that really informs the type of businesses that you can go after, and not only is that true for the initial deployment of constellations, it's true for all of the expansions as well. So, I mean, the example I use, it's a little bit like -- let's say Verizon or AT&T says -- hey, I've got this really interesting breakthrough in cellular technology. The only requirement is that I have to distribute my cell towers uniformly geographically across the US. So for every one I put in New York City, I've got to put a corresponding number in Montana, Wyoming, Idaho and all those other places as well.

And that's really what happens with these low earth orbit systems, so for us, what we've seen -- and we've shown this multiple times -- is that geographic demand for bandwidth is very localized, and something like 95% of all the people in the world live on 5% of the surface of the earth. So that's what I think you'll see, is that we're going to be able to go after markets where there's high demand and then reinforce those markets very cost-effectively with additional spacecraft. That's a lot harder to do with GEO, and it doesn't mean that you can't be successful. It just means that you have to have a very different go-to-market strategy. So we like ours, and we don't -- it's not that they can't co-exist, but we certainly like ours better.

Thank you very much.

Louie DiPalma, William Blair.

Mark, Rick and Shawn, thanks for taking my questions. Since you have made progress on a commercial ViaSat-3 capacity agreement with Eutelsat, I was wondering if you could address the ViaSat-3 Department of Defense opportunity. And based on preliminary tests, how does ViaSat-3 compare to the US government's legacy wideband, extremely high frequency and MUOS systems? And what can those systems do that ViaSat-3 can't do, and what can ViaSat-3 do that those systems can't do?

Okay, so that's a good one. So one is we're very optimistic about our prospects with the US DoD. So far, all of our business with DoD has been conducted through individual service agreements. So generally, we've made agreements with specific organizations who have purchased broadband terminals from us and then gone on to purchase individual service contracts for those platforms that are tailored to their operational requirements. And I'd say in the big picture, that's probably the way things will continue.

We do -- what's interesting in DoD is that all of the defense organic capabilities are procured against very specific operational requirements. So if you look at -- you've mentioned kind of the advanced EHF. So advanced EHF has very specific nuclear strategic functional capabilities, which ViaSat-3 wouldn't necessarily address.

Right.

Now, it also has some applications for lower-end technical uses that we feel we would address probably more effectively. And so while we wouldn't displace an EHF constellation, we may be able to absorb some applications that are more tactical away from its primary mission, and so we've had ongoing discussions with users about that.

If you look at the other organic systems, like the WGS and MUOS, they're really designed for very different things. WGS is already well over-subscribed, and, again, the DoD has not yet really identified ongoing requirements beyond WGS that they've funded other than the ongoing purchases of commercial bandwidth.

And what we've been able to convey to our customers is that ViaSat-3 compared to your other commercial options, and actually even compared to WGS, has all these features -- much higher speeds, more bandwidth per platform, more simultaneous platforms in the air, probably greater resilience against accidental or intentional attacks, and actually a much lower ongoing cost.

The MUOS segment was originally really intended for ground users, but it also has chipboard and airborne applications. Our services are -- where those are measured in kilobits, ours are measured in megabits. So for applications that are things like imagery, intelligence, battle management, a lot of our customers are -- even in the case where they do have MUOS capability, are augmenting it with ours. So overall, we've had really good reception, and we don't really see on the horizon either written requirements or alternatives for the features that they've been buying from us.

Great. Thanks.

Thank you.

About one more call.

Okay, one more.

Mike Crawford, B. Riley & Company.

Just to continue on this low earth discussion, so the -- you and 10 or 11 others provided these proposed constellations that would share spectrum that OneWeb is requesting. How many of these constellations do you think the FCC would give spectrum to and might get billed?

We can't speak for the FCC. We've had only preliminary feedback on our filings. It's a little bit complicated because of the change in administration and change in management at the FCC. Overall, I think the main pieces of feedback I'd say we've gotten are -- one is that our filing was thoughtful and that there was a range of thoughtfulness, I'd say, and completeness in the filings, and ours was probably at the higher end of that. And then the other one is what are they going to do with that, and right now, the feedback is the FCC doesn't necessarily see themselves in a position of picking winners and losers, and so they're going to -- they're trying to wrestle with what is the best way to deal with that.

I think one way to sort of understand what might happen would be to go back and look at what happened in the past, in the 1990s, when there were these multiple LEO constellations for L band and S band and how the FCC dealt with that, which was really kind of dividing up the spectrum and allowing the market to sort through what the implications were. You might see something similar this time.

Okay, thank you. And one, I thought, unique aspect of your proposed constellation was this feature where you would have antennas pointing not only down to earth on your proposed constellation, but also pointing up towards ViaSat-3 to use your global ViaSat-3 network as a form of backhaul and a means of getting data from the satellites back down to the ground and to users. To what extent might you be able to do that for others' low earth orbit or MEO constellations, including the DoD?

Yes, that's a really interesting capability. Basically, what you need to be able to do to do that is you need -- on the non-GEO, the lower earth orbit, whether it's LEO or MEO constellation, is you need to have antennas that look up, or you need to have some way to look up as well as down. I think right now, most of the LEO communications constellations that you're seeing think that latency is the primary value proposition for their system, and so they're not really that interested in looking up.

On the other hand, there are a whole bunch of lower earth orbit earth-sensing and observation satellites that see that as a really, really interesting opportunity, so we've been working with them, and we see that could be a really good feature. For us, we're kind of a little more focused on speed and volume of bandwidth. We think it makes sense even for communications satellites, but for systems that are going to have more expensive bandwidth, they're probably going to be more focused on the latency.

Okay. Thank you very much.

Thanks, Mike. Oh, I guess we have one more question that we'll take.

Andrew Spinola, Wells Fargo.

Just a pretty simple question -- could you walk through the comments you made about fiscal 2018 EBITDA impacts? If I'm correct, the settlement goes away in 2017. You made the comment that the R&D expenses go up for the next couple quarters, so I would think that it would be over $25 million a quarter, maybe something like $100 million or $105 million in fiscal 2018. And then the comment about transition expenses with ViaSat-2, is that just simply increased SAC because of stronger gross adds, or is there something else in there?

Okay. Yes, I think the Loral SSL is about $7 million --

A quarter.

A quarter, yes.

Six and a half a quarter.

$6.5 million a quarter, so that'll go away. The R&D (inaudible)?

A little bit higher than that, but not much.

A little bit higher than this year?

Well, a little bit higher than the number Andrew gave.

Oh, okay.

It's up a little bit from there.

Okay. And then on the ViaSat-2, basically the two main effects will be that once the satellite goes into service, we'll have a little bit higher fixed costs on the ground network, and we had that same effect when we launched ViaSat-1. It'll be moderated this time, but we'll have a little bit higher fixed cost, and then the other is the SAC expenses.

To think about it, you have to spin the fiber up and the other things to be able to support the satellite, Andrew, to start with.

All right, got it. Got it.

And then the success base (inaudible).

Understood. Appreciate it. Thank you very much.

Thank you, Andrew. Okay.

Thank you. I show no further questions at this time. I'd now like to turn the call over to Mr. Dankberg for closing remarks.

Okay. Well, thanks, everybody, for tuning in. That concludes our presentation for this quarter, and we look forward to talking to you again next quarter, right after our ViaSat-2 launch. Should be fun. Thanks.

Thank you. Ladies and gentlemen, this does conclude today's conference. You may now disconnect. Have a great day.