Viasat Inc (VSAT) 2017 Q1 法說會逐字稿

Welcome to the ViaSat FY17 first-quarter earnings conference call. Your host for today's call is Mark Dankberg, Chairman and CEO. You may proceed, Mr. Dankberg.

Okay, thanks. Good afternoon, everybody, and thanks for joining our earnings conference call for our first fiscal quarter of 2017. I'm Mark Dankberg, Chairman and CEO. And I've got with me here Rick Baldridge, our President and Chief Operating Officer; Shawn Duffy, our Chief Financial Officer; Kevin Lippert, our General Counsel; and Bruce Dirks, our Treasurer. And before we start, Kevin will provide our Safe Harbor disclosure.

Thanks, Mark. As you know, this discussion will contain forward-looking statements. This is a reminder that 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. That said, back to you, Mark.

Okay. We will be referring to slides that are available over the web. And I'll start with some highlights and a top-level business overview, and after that, Shawn will go through the consolidated and segment-level financial results. And in this quarter, we thought we would spend a little time reviewing how we think about our R&D investments in satellite broadband productivity, and give some color about how that productivity manifests itself in our key broadband services businesses, including an update on the R&D status on ViaSat-2 and ViaSat-3. And then we will review our outlook and take questions.

Our first quarter was a pretty good start to the year. Overall, revenue was up 5% year-over-year, new orders were up 10%, and adjusted EBITDA grew 3%, all compared to the same quarter of last year. As we said last quarter, we had significant growth in R&D spending in our Commercial Networks segment on ViaSat-3 and success-based commercial airborne Supplemental Type Certificates or STCs. As a reminder, we are expensing the preflight payload engineering activities on the ViaSat-3 satellites as an artifact of building the payloads internally. Absent that growth in R&D expenses, adjusted EBITDA Company-wide would have grown 16% year-over-year, and Op earnings even more than that.

Our Satellite Services segment was especially strong. ARPU increased 8% year-over-year, due to our combination of packaging attractive, higher-bandwidth, higher-speed service plans, and increasing attachment rates on a growing portfolio of value-added services. Subscriber count was relatively flat. That cost has been trending down on a per-subscriber basis. Commercial aircraft in-service is growing.

And all that led to a 32% year-over-year increase in adjusted EBITDA in the Satellite Services segment. Overall, we continue to see strong demand for Satellite Services in target markets. And later on, I will go into more depth on our longer-term thinking on how we drive profitable growth there, leveraging ViaSat-2 and beyond.

Our Government segment also had very good results. Adjusted EBITDA and operating earnings for the first quarter increased 13% year-over-year on flat revenue, but with a higher proportion of services than last year. Overall, our government business seems well-positioned for growth in revenues, orders and earnings, due to favorable outlooks in satellite facility services, active data links and some recently certified additions to our network security appliances portfolio.

Government customers have been initially preparing satellite mobility terminals, and then have been following that with the ongoing services contracts. As those are executed, that helps to provide visibility to our growth trajectory there. The Handheld Link 16 Radio that's shown on the slide is a good example of the way we've been growing our government business. We've had end-user support to develop a product that has strong market pull, even without there being a program of record for it.

A lot of our growth in the Government segment, including in mobility, has come from creating unique and growing markets for products and services that aren't served by the DoD acquisitions (inaudible). We're very pleased and excited by two very significant aeronautical mobile contracts with American Airlines and the US government senior leadership fleet, including Air Force One. Obviously those are really high-profile customers, and we believe winning those awards speaks volumes about the market recognition and the quality and economics of services on our satellite networks.

But even more importantly, they are strong endorsements of our very simple, yet highly differentiated, approach to satellite broadband infrastructure. On that note, Shawn will give some more detail on the financial results, and then I will come back and give some more color on that differentiated broadband infrastructure topic.

Thanks, Mark. As Mark just highlighted, the Satellite Services and Government Systems segments growth overall increased in adjusted EBITDA, more than offsetting declines in Commercial Networks. Specifically, revenues were up 15% in Satellite Services, and flat in Government Systems. Yet adjusted EBITDA was up 32% and 13%, respectively, in each of these segments.

Once again, we reached record recurring adjusted EBITDA of $72 million in Satellite Services, with EBITDA margins topping 47%, which is a 300-basis point increase sequentially from FY16 Q4, and a 600-basis point increase year-over-year. And I'll talk a little more about what drove those increases in a little bit.

In Government Systems, our adjusted EBITDA grew on flat revenue. This reflects the growth of our higher-margin service offerings as they fill in this segment, and higher product margins. This service revenue expansion was multi-dimensional, with strong growth from managed Wi-Fi services in military locations worldwide throughout ViaSat Wireless Services Group, plus growth in government mobile broadband service revenue. We continue to make good strides in this part of our business, with our recent announcement around future services for Air Force One and other elite aircraft for the US government.

In Commercial Networks, quarterly revenues were down slightly from the prior-year period, with lower sales of ground Internet infrastructure products offset by ramping modem sales to [MV&Co]. Our adjusted EBITDA performance truly reflects the strong ramping of our R&D on our ViaSat-3 satellite project we spoke about last quarter. And in-flight mobility SEC activities which, in total, doubled year-over-year, and with the primary driver for our reduced segment profitability. So overall, we had a very good performance this quarter, with continued growth in adjusted EBITDA, despite the ramp in the ViaSat-3 program and SEC activity associated with our commercial business.

Looking to the next slide, our income from operations was down slightly, with higher depreciation and other non-cash expenses offsetting our overall increase in adjusted EBITDA. And our GAAP and non-GAAP net income follow the same relationship, both down slightly from the prior-year period. But again, keeping in mind our R&D trends I noted earlier, each of these measures would have actually grown very sharply year-over-year, at levels of over 50% on a constant R&D-level basis.

Regarding cash flow, we more than tripled our cash flow from operations from the prior-year quarter with another quarter of solid EBITDA performance. Plus strong collections on the large broadband infrastructure program wind-downs we discussed earlier, which reduced billed and unbilled accounts receivable balances by $23 million from the end of the last quarter. And if you look to our trailing 12-month cash flow from operations, you will see we generated $341 million of cash, which was over 20% higher than last year Q1 for the same period, excluding the $40 million one-time Loral settlement. And more than double all of last year's satellite CapEx investment. So clearly our operations continue to fund a meaningful portion of our strategic growth activities.

Our Q1 capital expenditures were up $55 million compared to last year, with the additional two ViaSat-3 satellites under construction, as well as higher quarterly expenditures on our ViaSat-2 project and related ground segment as we approach the beginning of the launch window at the end of this year. So we ended the quarter with $200 million outstanding on our revolver, and $264 million drawn on our $387 million loan commitment from Ex-Im Bank.

Overall net leverage was 3.1 times trailing 12 months adjusted EBITDA, up modestly from 2.8 times in the fourth quarter of FY16. And our liquidity remains very strong. Last quarter, we spoke about our Q1 revolving credit agreement upside of $300 million to a total of $800 million. This increase and the growing contributions from our operations brought a significant lift in liquidity quarter-over-quarter to $689 million, with combined availability on the revolver and Ex-Im loan commitment, and the $47 million of cash on the balance sheet.

This next slide summarizes our key Satellite Services metrics. As you can see in the chart on the top left, adjusted EBITDA increased pretty dramatically on a sequential basis, up 13% from the previous quarter on an essentially flat subscriber count. This earnings improvement was largely the result of another quarter of strong ARPU growth, up $1.50 per sub, per month from Q4 last year.

Since we launched ViaSat-1, we have grown ARPU nearly every single quarter. We have consistently grown our retail mix, as well as tailoring plans and value-added services that appeal in the dimensions our customers value. So in Q1, we saw evidence of this again, with more customers selecting value-added services, and more customer upgrades, especially in transition to our Liberty Plans. And remember, a large majority of that ARPU increase drops right to the bottom line. And we think these trends will continue throughout FY17, as we ready for our ViaSat-2 service launch next year.

Another important notable is related to our SAC per-sub cost, which this quarter reached the lowest point since our ViaSat-1 service launch, mostly as the result of lower advertising costs, which are expensed. So with the additional growth from our commercial air business, which I will hit on in a minute, our segment adjusted EBITDA grew sharply and reflected a year-over-year revenue conversion of 87%.

On the in-flight connectivity front, we added another 33 commercial aircraft during the quarter, ended the period with 509 commercial aircraft on our Exede network. We expect tail counts to continue to grow with our existing customers, and that will be further expanded with our newest customer addition, American Airlines, initially outfitting 100 new Boeing 737 MAX aircraft. We noted earlier that we are continuing to invest in obtaining Supplemental Type Certificates as airline relationships expand, along with the types of aircraft approved for Exede service.

So year-over-year, we continue to grow our consumer broadband and in-flight Internet businesses in top-line revenue and EBITDA, on an absolute and incremental-margin basis. These results would not have possible on our legacy WildBlue satellite, with their limited capacity and higher capital cost per gigabit compared to the much more efficient ViaSat-1. And our upcoming ViaSat-2 and ViaSat-3 platforms, which will take this efficiency to an all-new level. So with that, I will turn it back over to you, Mark.

Okay, thanks, Shawn. This next slide makes a relatively simple point, but it is one that is really important to understanding our thinking about ViaSat's businesses and how we build competitive advantage. The charts are a reminder that broadband users keep demanding higher speeds and more bandwidth, and that the revenues or ARPUs that service providers earn from their users isn't growing as fast as the speeds and volumes.

The service providers with stronger competitive positions, like cable companies competing mostly with DSL, have been able to earn ARPU gains. But customers still continue to get better and better values in terms of megabits per second, per subscription dollar, or gigabytes of consumption per dollar. In markets with greater competitive intensity, like, say, cellular services, ARPUs could actually decrease, even while speeds and volumes are increasing.

So this race to deliver more output with the same or less capital input is essentially the definition of productivity, and that is what we're going to focus on. The upshot is that service providers with infrastructure productivity advantages are going to do better than those with lower productivity. So cable hybrid-fiber coax networks improves speeds and volumes more efficiently than copper or telephone plant does, so they are winning those competitions.

It's harder when cable companies compete with fiber. In mobile cellular, the major providers are using industry-standard networks, so competitive intensity is a bigger factor than relative productivity. We'd like to compete using assets not like cable hybrid-fiber coax versus telephone copper, are much more productive in dimensions that matter to customers. Absolute productivity gains relative to demand are really good, but relative productivity gains compared to the next-best competitor also matter. So even in times when usage demand might go faster than productivity gains, you can still do well if your productivity is better than competitors'.

Also one of the big advantages of the types of flexible satellites we're building is that our bandwidth is portable. We can move it to those markets and applications where our productivity advantages are greatest. Productivity gains are valuable when per-capita bandwidth consumption continues to grow.

We believe that is absolutely the case, as more users spend more time consuming more video, in more ways, on better devices, with higher-resolution screens. Whether that's live-streaming, over-the-top, video-on-demand, embedded in social media or otherwise, if we can improve our bandwidth productivity faster than others, we can embrace bandwidth demand instead of fighting it. And satellite broadband is very good at video.

So when we decided to enter the satellite services business with ViaSat-1, our fundamental objective was to lead the industry in satellite network infrastructure productivity, measured in useful gigabits per second of satellite bandwidth, per million dollars of invested capital. Then what we want to do is use that ViaSat-1 bandwidth in bandwidth-intensive applications to turn that productivity into profits.

We wanted to redefine the satellite services playing field to be technology-driven, where end-to-end gains in productivity would yield significant competitive advantages in high-growth markets. And we believe we are demonstrating that in residential broadband, commercial aeronautical mobile and government mobile broadband. Finally, in cases where our infrastructure productivity is better than terrestrial alternatives, and that can be compared to DSL, for instance, or air-to-ground aeronautical networks, we can take customers from those markets too.

So this chart shows relative productivity of different classes of satellite networks in two ways. The right-hand or orange scale measures gigabits per second of bandwidth capacity, per million dollars of infrastructure investment. Or gigabits per megabuck -- that is the term we have used. The left-hand axis shows roughly how much a satellite operator would need to charge for that bandwidth in millions of dollars per gigabit per second, per year, to earn a 10% after-tax return based on a common set of assumptions on fill rate, EBITDA margins, and other factors.

And be sure to note that the left-hand scale is logarithmic, and that reflects the fact that there's orders of magnitude differences in productivity among these different classes of satellites. The least productive in pure bandwidth are our L-band mobile satellites. L-band enables handheld phones, and it is relatively immune to weather, but it is very expensive on a cost per bit.

Productivity has improved from there for traditional C- and Ku-band fixed satellite services, global mobile Ka-band, what is called high-throughput satellites for Ku-band. Then there is existing and planned A-band high-throughput satellites. And finally, the most productive is ViaSat-1, and then the two or three series beyond. You can do your own spreadsheets, but the lower your bandwidth manufacturing costs are, the lower the selling price you can offer and still make a pre-defined return on investment, other factors being equal.

We also show a column for resellers whose primary business is selling connectivity to users like airlines or maritime ships or government platforms, by packaging and selling bandwidth purchased from the satellite owners. We show a range of prices that they would charge, based on the type of bandwidth they are selling. Since their costs build on their prices that are charged by the satellite operators, you would expect that their sale prices reach asset class, would be higher than that of infrastructures' owners, in order for them to earn a profit. Though the charts are directionally really representative of bandwidth pricing among these asset classes, think of productivity as destiny for the satellite broadband services.

Using those same assumptions, a ViaSat-1 class satellite could deliver a comparable return at an order of magnitude lower pricing. Which is just another way of expressing its order of magnitude productivity advantage in bandwidth manufacturing costs. Of course, we recognize that the going-in risk profile for ViaSat-1 was higher than that of a mature fixed satellite services, or a global mobile provider. So we aimed at commensurately higher returns.

Our current revenue run rate on ViaSat-1 is multiples of the representative million dollars per gigabit per second per-year value, and that's directionally indicative of the returns that we are getting. Also note that, even at that market price point, ViaSat-1 pricing is much better than or competitive with other asset classes. So far as we are aware, we don't see any existing or planned satellite infrastructure projects with better productivity than ViaSat-1. So we can use the productivity gains of ViaSat-2 and ViaSat-3 to provide even better services to our customers, expand our addressable markets even more, and still earn better margins.

You don't have to understand all of this math to tell the difference between the in-flight experience on a JetBlue flight compared to an air-to-ground Wi-Fi service. We believe we are seeing the effects of our productivity advantage in key markets, and that we can scale them to create even more compelling competitive advantages. And that is why we are investing at the pace that we are.

So on the next two slides we'll show how those productivity improvements are manifested in the Aero Mobile and in the residential markets. This slide shows a relatively recent snapshot of our commercial aeronautical mobile service and spot EMAPs of two Ku-band high-throughput satellites and a global mobile Ka-band satellite network covering the continental United States. Our spot EMAP is a composite of beams from ViaSat-1, WildBlue-1 and [Anatec-2] satellites, each covering different parts of the US. And we can choose from about 140 different available user beams in total. The picture only shows the outlines of the larger beams around the edges of the United States, but there are lots of other beams inside throughout the country.

So productivity is the ratio of useful output and gigabits, divided by cost input in mega bucks. And since the cost in orbit of all these classes of satellite systems is relatively comparable to each other, then that means that orders of magnitude differences in productivity must be due to order of magnitude differences in the useful available bandwidth per satellite. And that affect is further magnified because our first three satellites are wholly focused on the US and Canada, and the other systems are distributing their bandwidth over multiple geographic regions.

So if you look at the pictures closely, you will see each of the other alternative's system only has about five or six beams in total available to serve the entire US market. And each of those five or six beams has a total throughput capacity in the range of about 50 megabits to maybe a few hundred megabits, depending on the satellite beam bandwidth when those bandwidths are delivered to an aeronautical mobile terminal.

In the upper picture, each of the green arrows in the ViaSat network image is one of hundreds of aircraft in flight at that moment in time. Our network has so many beams and so much bandwidth per beam that we are currently able to deliver speeds of close to 100 megabits per second, per plane, for those planes with that level of demand, at each point in their flight. ViaSat-2 will more than triple our effective network capacity, and we will be able to support substantial growth in the residential network, plus, say, in the range of 2,000 or more aircraft, at service levels as good as or better than we deliver now.

But now imagine a network with only five or six spot beams trying to support even, say, 1,000 planes, having to live with way less bandwidth per beam. When those aircraft congregate in high-demand markets up and down the East Coast or at hub airports, there will be beams with hundreds of aircraft in them. Even if an individual airplane could process 100 megabits per second, the amount of bandwidth per beam, divided by the number of airplanes per beam, will often result in single-digit megabits per second available per plane. Which is about the same amount that is currently delivered by the air-to-ground networks.

In that case, putting 100-megabit per second modem on a plane only moves the bandwidth bottleneck from the plane itself to the satellite network infrastructure that is serving those planes. And even that assumes that a single reseller can capture all the Ku-band high-throughput satellite bandwidth available in the US market.

Finally, it's really important to note that the failure mode here is not that the bandwidth is so expensive, it's that there is just not enough of it. And the result is the same thing that's created the opportunity for us today -- too much demand chasing too little bandwidth, resulting in congestion, slow service, limited number of passengers per plane and limited online activities.

So right now, competitive intensity in the aeronautical mobile market is really high. But in the long run, to the extent that passengers behave in the air as they do on the ground, we believe the battle will be won based on bandwidth productivity. While we've made great strides in improving the number of passengers using the Internet and the amount of bandwidth available to each passenger, we're still in the early innings of helping the airlines capitalize on this onboard Internet experience.

We can help the airlines embrace the trends driving Internet bandwidth growth instead of fighting them. We believe the airlines are beginning to see the effects of this in their own business as passengers express their preference for better and better connectivity. And we are confident of good growth going forward.

So this next slide is one we have shown in the past. It was used by FCC Chairman Wheeler in September of 2014 to illustrate the state of competition in the US residential broadband market as a function of four different speed tiers. Just like in the air and mobile market, we would like to compete where we can have a productivity advantage relative to the competition, and where competitive intensity is less. In this case, the bandwidth productivity advances of ViaSat-2 and ViaSat-3 are really important to separate ourselves from satellite competitors, and to position ourselves better against less-productive terrestrial networks.

The blue segments at the bottom of each speed tier column show the fraction of the market that has no terrestrial alternative at that speed. So if we can target the 25 and 50 megabits per second portions of the market, then we can be the only viable alternative in about 20% of those total markets, compared to only about 5% to 7% at the 4-megabit and 10-megabit per second segments where we compete today. Furthermore, at those speeds, about 70% to 80% of the market has either zero or one choice of provider, compared to only 25% to 40% of the market with only zero or one choice of the lower speeds. So clearly with the right assets, that's a better place to play.

At higher speeds, we will need allocate more bandwidth per subscriber. But our [PLUs] will be higher, competitive intensity will be lower, and our bandwidth flexibility and portability will help us find those geographic markets with the best distribution channels, subscriber acquisition costs and retention rates. We do expect that terrestrial speeds will increase, and that more choices will be available in more geographic markets. But we also expect net subscribers will continue to migrate upward in speed to 100 megabits per second and higher, as long as we can improve our productivity faster relative to terrestrial alternatives in specific geographic areas, we've got attractive growth and earnings potential. That's pretty much the opposite of conventional wisdom that satellite services are condemned to play at the low-end of the markets, at price points associated with the least attractive mass market -- terrestrial alternatives.

So since we are investing a lot in R&D and capital, we figured it would be good to give an update on progress in those areas. So ViaSat-2 satellite construction and the integration is essentially complete, and the spacecraft is in the middle of environmental stress screening. All functional performance testing has been completed, and the satellite meets its performance objectives. We've also invested a lot in the new ground network system needed to achieve the broad geographic coverage and bandwidth flexibility that we expect will further improve the value and productivity of the network, compared to alternatives. And that's also meeting its objectives.

Some of our R&D investments in ground gateway technology are reducing system capital costs. So total capital investment is at the low end of the range we anticipated at the start. That means our productivity improvement target of double ViaSat-1 looks really good. There is upside potential in capacity, so there is a good opportunity to do meaningfully better than our initial objective, just as we did with ViaSat-1.

The upper right image shows one of our random installations. We've been investing in STCs for retrofits on narrow bodies, such as the 737-800 and 900 we are beginning to serve in Europe with EL AL. Overall interest from both domestic and international airlines has been really good, and their interest has led to good progress on factory retrofit and line-fit options with both major airframe companies. And with the ViaSat-2 launch nearing, and ViaSat-3 under construction, we are seeing new and significant activity for installations on long-haul wide-bodies too.

The lower right pictures our new facility in Tempe, Arizona, where we are building the ViaSat-3 payloads. It includes a high bay clean room for final assembly and integration of the payloads, which will be made into the high-power volume platforms. At this point, we are building preflight engineering units and performing module sub-assembly and payload systems tests and characterizations. It is still early, but so far, things are on track for a 2019 first launch, and performance is consistent with our objectives.

The bottom left shows government and commercial versions of our new 100-gigabit per second network infrastructure encryption appliance. This generation is aimed at cloud and data center applications, and it's the first Layer 2 device to market at that speed. So last year, we invested in development certification of new infrastructure and edge devices, and we're seeing really good market interest in those products.

Okay, so finally, I will give some insight into factors that should drive our financial results going forward. Given the upcoming launch of ViaSat-2 and the substantial lift we expect that can create across our businesses, we will consider each segment in a pre-and post-ViaSat-2 environment. For Satellite Services, it looks like current trends can continue prior to the ViaSat-2 launch and in-service states. We aim to continue to grow ARPU through higher-value plans and additional value-added services.

SAC costs on a per-sub basis have been trending down, and churn is roughly steady. We will continue to add new aircraft, and would expect that trends driving greater passenger engagement and bandwidth would continue. We're working with existing and new airline partners to help them better monetize our best-in-class passenger experience, and also to use connectivity to help reduce IFE costs -- in-flight entertainment.

So with ViaSat-2 in service, we expect new high-speed plans without hard bandwidth caps, to enable continued growth in ARPU. We will be targeting faster subscriber growth, probably with per-subscriber SAC costs growing to more historical levels, and aggregate SAC expenses scaling with subscriber growth. And we will also have some migration from existing plans. We expect churn to remain steady, or possibly improve from current levels, with the new plan.

We also may see a higher installation rate for new aircraft once ViaSat-2 is in service, and increased service availability due to the greater geographic coverage. And ViaSat-2 can also benefit our business debt services, as we migrate more of those from Ku-band to Ka.

In our Commercial segment, there's opportunities for modest revenue growth, especially related to mobile terminal equipment. As we said last quarter, we plan on continued elevated R&D expenses for ViaSat-3, preflight engineering and aircraft STCs especially. Post-ViaSat-2, we see increasing opportunities for revenue growth, especially in the mobile terminals, and R&D expenses may flatten.

For our Government segment, we've seen long-term trends towards growth in both product and service revenues. But with the growing proportion of service revenues, we believe those trends would continue, and given the lumpy nature of new orders, could have opportunities for backlog expansion. We're also in the early stages of new product cycles on small Link 16 tactical data-link terminals and next-generation network security appliances. We believe ViaSat-2 entering into service lifts our prospects for growth in the Government segment in terms of revenue earnings and new orders.

Overall, we're really focused on what we believe is a very compelling longer-term story. I hope we've shown, is that, starting with ViaSat-2 going into service, and accelerating as ViaSat-3 comes into the picture, that we are building a sustainable competitive advantage in terms of asset productivity, that greatly expands the opportunities for us, both in our current businesses and a much larger addressable market.

You'll hear us talk more in the future about opportunities we are just now beginning to get into, in areas like maritime and global Wi-Fi access, that have exciting potential for us to deliver the growth that our groundbreaking technology can enable. And that's why we're investing at the rate we are. And we will look forward to sharing more as those [developments] mature. So that's it for our prepared remarks, and we will be happy to take questions.

Thank you.

(Operator Instructions)

And our first question comes from the line of Rich Valera from Needham & Company. Your line is now open.

Thank you. Appreciate the color on the pre- and post-ViaSat-2 with the different businesses marked. Just wondering if you thought that migrations might become a factor again post the launch of ViaSat-2, as they were when you initially launched ViaSat-1?

Yes, we do. We listed that as one of the factors. I think that it's going to be much less of a factor than it was with ViaSat-1, for the main reason that the difference in the satellite productivity is narrower than it was between WildBlue-1 and ViaSat-1. So they were looking at a 10x, 20x increase in productivity. And this is more like double, or maybe a little better than that. So there will be some, but I would think it's -- it ought to be more manageable, based on the way we define the services.

Okay. Is there any way we could think about that? I think you peaked at like [60,000] a year in the first year of that transition. So sort of a factor -- significant factor less than that is how we should think about it, when we think about the first year of being here?

The way I would think about it is, one of the ways -- and we talked about this a couple quarters ago, is think about it as we improve the productivity, we can use some of that bandwidth gain to improve the service. That's one of the big factor, and another part to improve our margins. So think of it as, if we gave it all to our customers, we would have half the customers per unit bandwidth. If we kept it all for ourselves, we would have double the customers in total, per satellite, and we will be in-between.

So what it really will mean is that the yield on ViaSat-1 will go down a little bit, but will grow subscribers on ViaSat-2 at a faster rate. And because a lot of the terminals are capable of the services that we offer, ViaSat-2, we won't have the SAC costs associated with the migrations that we did before. So that is a big benefit for us.

So when we model migrations, think of it -- it's a little more of a normalizing productivity than it is SAC expenses, as it was last time. There will still be SAC expenses associated with it, but those will be moderated.

Got it.

Does that make sense?

Yes, that does. And just wanted to get your sense on how your message is resonating in the market. You presented a pretty compelling picture there of the relative bandwidth/cost benefit that you have, versus pretty much anything else out there in the market. And it seems like even in the worst case, ViaSat-1, you are at least 10x better than pretty much anything else out there. And then you can -- it goes up to kind of 100x.

So when you present this to your prospective in-flight connectivity customers, how is this resonating? Or are they getting it? And what would be a pushback they would have? It seems if they're looking at that, it would be a no-brainer to go with you guys, to go with the guy with 100x bandwidth cost advantage. But what are some of the things you are sort of hearing, and how is that resonating in the marketplace?

One thing that anybody can observe is, just look at the language that is now around what is going on in the airplanes, right? People are always talking about, oh, we need 100 megabits per second per airplane, right? Or 200 megabits per second per airplane. So people are -- one is, you are seeing an acknowledgment that the demand for bandwidth is going to be a lot higher than it was, So that's a good thing.

I think another thing that you're seeing is more people talking about the things that drive bandwidth consumption, which is video streaming, as an example. So you can say, well, we can do video streaming. Or you don't need video streaming, look at our IFE, who needs Netflix? I think one is that you are seeing more and more conversations around this.

Another thing that I think is actually indicative of what is going on in the market is that more and more satellite operators or re-sellers are talking about capacity in satellite and gigabits. One of the things that we tried to emphasize in our productivity chart is that you're going to see -- when people know that gigabits are important now, I think they are inflating the gigabits rankings of satellites, compared to what they were talking about originally.

One of the really good ways for people to reconcile that is to look at illuminated gigahertz of spectrum, and then to think about, you're not going to get 10 bits per second per hertz, or four bits per second per hertz, onto an airplane with a small amount of illuminated gigahertz. So what's happening is, I think that the playing field is tilting to the points that we have made. I think that the competitors are trying to frame their offers more in the ways that we've framed ours.

I think that I'd say industry consultants and customers are still wrestling with all the technical details that go into making that chart, and people are, I think, still a little bit confused about it. But I would say, when we talk to airlines, they are more and more interested and concerned with, how can their providers ensure that they'll deliver the amounts of bandwidth that they say they will, much more focused around that. And that's why I think you're seeing people put out one airplane or two airplanes and saying hey, look, we can stream Netflix, we can get 100 megabits.

But I think the real issue is, what happens when you have hundreds or a thousand airplanes in a small number of beams? I think people are little bit confused about that, so that is our current campaign. The flipside of that is, especially when we've gotten some of the customers that we have, there are a number of airlines that, especially these little-known international ones, that are just saying, we just want the best one. And right now, I think there's a lot more acknowledgment that that's us.

And so that's one of the reasons that we've been so busy in the STC space, and then working with the air framers on the widebody for a line fit and factory retrofits. And you are covering the space too, so I think you can weigh in and (multiple speakers)

Yes. Oh, fair enough. I appreciate that color. That was helpful, Mark. Thank you.

Thanks, Rich.

And our next question comes from the line of Andrew DeGasperi from Macquarie. Your line is now open.

Thanks. Macquarie. First question, could you maybe clarify for us what the SEC special frontiers order impact is to ViaSat-2 or ViaSat-3, potentially, I think, it's in the 28-gigahertz band?

Sure, yes. We obviously follow that, and we have paid a lot of attention to that. We have been probably the leading proponent of spectrum sharing in the satellite industry. I think that put us in a leadership position in the discussions around that. And also we've been probably ahead of the curve in taking advantage of spectrum sharing in that 28-gigahertz band.

So at the end, I think that the statements that the SEC made around what their intentions were, are good for 5G and for the satellite industry. They acknowledged the issues around grandfathering the teleport gateways for existing satellite systems, including for ViaSat-1 and ViaSat-2. And there are provisions for many thousands of new gateways throughout the US for ViaSat-3 and beyond, using that 28-gigahertz spectrum band.

The SEC also acknowledged that there was need for further study to deal with both the local interference levels associated with those gateways, and the aggregate interference that may arise from 5G use of the 28-gigahertz band on the spacecraft itself. So I think we believe that there is a framework in place for a good outcome, but there's still a lot of work to be done. Overall, we think the outcome is good, and consistent with our plans.

Great. And if you could just remind us what the backlog is currently on your commercial aircraft? I see that you keep adding planes, but where are you essentially adding these from? Are they existing customers that are just giving you more aircraft, or are they new fleets coming online? Can you maybe just comment on that?

Yes, so we have basically, primarily adding aircraft from the fleets that we currently have. So that is adding aircraft for JetBlue -- for instance, JetBlue completed its regional expansion, JetBlue has also talked about adding planes to its fleet. We've been adding to the United fleet, Virgin America, and EL AL. And then we have -- the single biggest lump of backlog is the American Airlines 737 MAX award.

If we think about the rest of the year, [tell] ViaSat-2 goes up from the swing in American, that is when it will hit. But as far as your existing customers, how should we be thinking about net additions?

I would say maybe not quite at the levels that we've been at in the last quarter, probably going down a little bit from that, based on what the current situation is. That could change, but I would say for the next couple of quarters or so, I'd think about it in those terms.

Got it. And last question for me. Retail business, based on DISH net numbers, it seems that your retail business has more than made up for the weakness in wholesale. Can you maybe tell us what you are doing differently there?

I think we're doing the things that we've described over the last year or so. I think we've got -- we are very happy with our retail distribution network. I think we've expanded that. I think that the plans that we're offering are good. They are more attractive. I think we are just grinding it out. I think we're getting operationally pretty good, and I think those are all good factors pending the launch of ViaSat-2.

Got it. Thank you.

And our next question comes from the line of Mike Crawford from B. Riley. Your line is now open.

Thank you. Can you give an update on the status of your JVs that you're forming with Eutelsat?

Okay, yes. I think last week, week before, they had their earnings conference call, and that was one of the topics there. And I think our view of it is just like theirs, that we've got good top-level agreement between us, there's a pretty fair amount of detail to go through to reach closure on the agreements. We don't see anything that is jumping out as threatening that, but there is work to do. I think they said, and would we agree, that probably this year is a reasonable target for completing it.

Also, one thing I would like to point out, which is good in the meantime, is that we're cooperating with them on helping to think about the KA-SAT services. And one point we mentioned is that we now have our first joint European aircraft on KA-SAT with EL AL. So that is exciting, and I think it's good progress.

Just one more question in that regard is that, Eutelsat sold its majority stake in this maritime service provider. So does that mean that you will be able to serve as the maritime market with Eutelsat, with one of the JVs?

Mobility services is definitely on of the targets in the JVs, so that would be through KA-SAT and through the next-generation satellite. And I think we are a really good mobility partner. I think we would have to talk -- you would have to ask Eutelsat what their overall plans are in serving that market. But I don't think that is -- I'm not sure I would read too much into that one particular issue.

Okay. And if I could just turn back to this productivity issue, Mark, one thing you stated on this call, I believe you said ViaSat-2 will more than triple effective network capacity for airborne mobility?

If you think of ViaSat-1 having unit capacity of one, and ViaSat-2 has double that capacity, then between the two of them, that makes three. And that's (multiple speakers)

I thought you were referencing the more flexible beam architecture.

No, I don't want to be at all ambiguous about that. Basically, what we've talked about is, double the throughput. So that means we've got a lot more illuminated gigahertz on ViaSat-2 than on ViaSat-1. And we've gone through this in the past, the infrastructure cost is like 20% to 25% higher, so the total throughput must be more than double in order to get double the bandwidth economics.

And we have not counted flexibility as part of that. It is just raw bandwidth. Now, we think that flexibility and portability is going to increase the yield on the satellite, but we haven't factored that into that number.

The other thing is -- this is Rick. With the increased coverage, I think that we'll still have more time of the aircraft that are flying, and more time within the coverage area. And so that is going to be a factor as well.

Yes. I think one of the things I've said also is, we think there's upside to our current numbers. So we will report on that as some of the elements of that fall into place. But right now, we're sort of -- we think we are doing well, compared to what our original target was. Because the capacity numbers are holding and the capital costs are at the low end of the range.

Thank you. And then just to try to be absolutely clear on this, so with the Ku-band satellite that has a wide beam, maybe only five beams over North America. So what, in theory, could you do with things like antennas and modems and compression software, to improve the volume of bits that you can deliver to one aircraft, and then to hundreds of aircrafts? Like to use your example, up and down the East Coast?

What you look at is illuminated -- you look at illuminated gigahertz. And there was a time, maybe they are not advertising it so much. But there was a time when you could look on the web, and you can find the illuminated gigahertz for each of these satellites, and then you divide that among the beams. And it's not necessarily uniformly allocated among the beams, but it gives you a sense of the order of magnitude of the amount of gigahertz that you have to work with in each of these beams.

And then from there, what you do is, you just turn that into how many bits per second per hertz you can get, and that is a function of the link budget. The antenna on an airplane is a factor, but it's basically a two-way link budget, so you have to look at -- you can improve with the bigger antenna, you can improve the down-link portion of the link budget a little bit. But that's not the total link budget. You would have to put that in context of the total link budget. And it is 20%, 30% increase in capacity relative to a smaller antenna.

You can make bigger and bigger antennas, and you can get a little bit more gain and a little bit more throughput, but you are not going to go from 1.5 or 2 bits per second hertz to 4. You might go from 1.5 to 2, or 1.8 to 2.4, or something along those range. That is what you get. The rest of the stuff, things like compression, those are pretty much available to those people.

We think we have really good compression technology. A big part of that has to do with the way content is distributed by media companies. So there is limits to compression that are driven by end-to-end encryption that are available to any ISP. So our point is, there's not a lot of maneuvering room, and it is, like we say, it is like no bandwidth productivity is the destiny of these systems.

Okay, thank you. And then just switching to commercial networks, you're talking about some opportunities to grow, even though the types of gateways and architectures you are building and designing for yourself are different from what others seem to be considering, unless that's changing. So is this more of a modem-type opportunity with NBN and potentially others? Or what is it specifically that you are targeting for potential growth in commercial networks, given the way that your business has shifted?

Yes. One of the decisions that we had to make was whether we wanted to build network infrastructure or garden-variety, quote, hyper throughput satellites, or whether we wanted to invest in those things that would drive productivity. And what we have chosen to do is to invest in things that would drive productivity. I think if we were only in the business of building and selling products, that would be a tough decision. But when we are in the business of selling services, and we have these enormous productivity gains, that justifies the R&D investment that makes it an easy choice.

So what we're doing is, we're really building ground infrastructure for satellite networks that are using our unique technologies, and we haven't seen any other operators really build for those. So NBN was a good opportunity for us, because they use basically ViaSat-1 technology. What they want to do is a ViaSat-1 in their market, and KSAT was the same way. So those are good fits for us.

But we're not investing in that type of infrastructure anymore. So we don't really model selling infrastructure, going forward, as a standalone product. What we are seeing, though, more and more, are outside satellite operators saying, wow, I think about this productivity issue, we would like to be on your side.

So we are seeing more opportunities for things where in specific regions, satellite operators would partner with us, building on our network infrastructure. Eutelsat and KSAT, that's a really good example of that. So some of that might lead to network and infrastructure sales, like for instance, what we did with XEI on ViaSat-2. We have other opportunities for infrastructure product sales, but they are more based on our own types of satellites and on partnerships.

Okay, thank you. And then a final question relates to government systems where you have STT and this [BATS] terminal that represent potentially quite large programs that, as you mentioned, came in not through normal programmer record. Are there other opportunities like that, like say with that information assurance products, that you are targeting right now?

Yes, definitely. More and more, we are seeing a disconnect between the procurement system and what users want. And so that's what we are aiming for, especially things that are adjacent to or variants of stuff we already do. And a great example of that is the STT terminal, which started out as a, basically as a R&D program with us, with some customer support, but then turned into a program of record for Apache helicopters and others. Quite a big market for helicopters, both domestically and internationally and other small form factors for that are handheld.

Really good sign recently was given the radio designator by DoD, which is indicative of its desire to turn that into essentially a new program of record for that type of application. And we see more of those, not only in the technical data links and security space, but also in the satellite space. I think that is a big part of why our revenues are still growing, and we are diverging from some of the other big defense aerospace command and control communications companies.

Okay, thank you.

Our next question comes from the line of Andrew Spinola from Wells Fargo. Your line is now open.

Thanks. Mark, I wanted to ask you about, on slide 10, when I'm looking at the coverage maps of ViaSat-1 versus your competitors, the one thing that is clear to me is that you've obviously concentrated your capacity with ViaSat-1 over the United States, while your competitors have opted for a lot more coverage. And I'm wondering, as you opt for a lot more coverage of ViaSat-2 and ViaSat-3, why won't you suffer the same or similar productivity declines that your competitors have with the existing satellites? Or will you?

That's a really good question. So there's really two factors. One is, the total illuminated gigahertz that we will have with our satellites is much, much larger than the total illuminated gigahertz for those other ones. So if you look -- if you go through and do some research, you will see that the total illuminated gigahertz for those three others ranges from 5 or 6 gigahertz to maybe 10 or 12 gigahertz. ViaSat-1 was 100 and ViaSat-2 is quite a bit more than that. ViaSat-3 is way more than that.

So one is the illuminated gigahertz, and that is really the factor that we have focused on. And by illuminating gigahertz, what I mean is, take the bandwidth per beam, multiply that by the number of beams, and you will get what's called total illuminated gigahertz. So number one is, that is a really important metric of total productivity over the entire footprint.

But the other really big factor is the one that Mike Crawford was bringing up, which is this flexibility. And so one thing that would be nice with the satellite -- and this is what we do with ViaSat-2 is, you could say, hey, boy, the US market for in-flight Wi-Fi is way bigger than all the rest of those coverage areas. How about if we move all of those illuminated gigahertz into the US market? That would be really good, if that's where the demand is. And none of those other satellites have the ability to do that.

That is a really important thing that we are doing on ViaSat-2, where we have a very big footprint, we cover Latin America down to South America, we cover the Atlantic Ocean, we cover the Caribbean. But if we wanted to, we could bring all that bandwidth back into the US. So that doesn't increase the illuminated gigahertz, but it makes it a lot more useful when different markets develop at different rates. So we think those are the two really big advantages that we have.

Got it. And I just wanted to ask one quick follow-up. You mentioned the maritime market is something that might be interesting to you longer term. And it's a unique market, given the predominance of L-band and the strength of one competitor. And I'm just wondering, it seems like ViaSat likes to target areas that can be disrupted and markets that have characteristics that would lend themselves towards being disruptive. And just wondering if you have a similar thought process about the maritime market? And if that's one of the reasons why you think that is an opportunity?

Yes, exactly. The thing that is really interesting about the maritime market now is, lots and lots of ships, tens of thousands of ships in total, but not very much bandwidth is used. So the bandwidth per ship is not that high. And essentially, what a lot of those ships are paying for is the ability to be connected if they need to be, but they are not using a lot of bandwidth, because there is not a lot of bandwidth to be had.

At the other extreme of the market, there is a small number of ships with very large bandwidth demands. And for those, it doesn't mean that they would have to cancel their L-band subscription. They could still use that for safety and emergency communications.

But they have very high bandwidth needs, because they have a lot of passengers, or they have a big crew, or they have an exploration mission that requires collection of enormous amount of data. Those are the ends of the market that we are going after, the small number of ships with very large bandwidth departments. Because that is where we think most of the money is, and that is what we're going to target.

Got it. Thank you.

Okay. Well, I think that covers it. My voice is gone, and that covers a lot of the questions, so I think that will be it for this afternoon. Thanks a lot, everybody, for joining our call. Look forward to talking to you next quarter.

Ladies and gentlemen, thank you for participating in today's conference. This concludes today's program. You may all disconnect. Everyone, have a wonderful day.