Welcome to Viasat's FY 2021 third quarter earnings conference call. Your host for today's call is Rick Baldridge, President and CEO. You may proceed, Mr. Baldridge.
Okay. Thanks for joining us today. We've tweaked things a little bit in response to some customer input. We've adjusted the timing of our call so that people would have a little bit more time to review our shareholder letter and prepare for the Q&A session. We released our shareholder letter earlier today, and hopefully, you've had a little time. Today's call will primarily consist of Q&A. First, let's have Robert Blair provide our safe harbor disclosure.
Thanks, Rick. 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 report on Form 10-K and Form 10-Q. Copies are available from the SEC or from our website. With that said, back to you, Rick.
Thanks, Robert. Hello, everyone. Thanks for joining us on the call today. In addition to Robert, we've got Mark Dankberg, our Executive Chairman, our CFO, Shawn Duffy, and Paul Froelich from our Corporate Development and Investor Relations teams. Like I said, we're going to do something a little bit different today with our Q&A. We've had some inbound questions from investors in advance. In addition to the questions from analysts, as usual, we'll review some of the investor questions to the extent they haven't already been covered during the Q&A. We thank those investors for submitting the questions. Before getting to the Q&A, we'll do just some very quick highlights from the letter. First, another really solid quarter performance-wise, continuing our really strong EBITDA growth throughout the fiscal year 2021.
In fact, we anticipate the full year will be another record for the company on the EBITDA and operating cash flow front. The performance drivers remain pretty consistent with what we've been focusing through the pandemic period. The Fixed Satellite Service results were excellent, driven by strong demand in the markets we're in. Our Government franchise performance was solid, but it's being impacted by the logistics issues associated with COVID, like we've talked about before. This is pretty much expected, and it reflects some of the inherent lumpiness we usually see in several categories and in awards. Overall, we feel really good about our outlook on the government, because we've had a big order backlog, really good first half of orders, and our backlog's growing, and still growing in that area, which helps underpin our confidence in the future.
IFC remains impacted by the travel volume. We did see a sequential pickup in the quarter. We've had a really strong quarter in our commercial networks business. We generated $224 million in operating cash flow in the quarter and over $700 million on a trailing 12-month basis. Net debt was flat. Our leverage ratio declined, helped by our continued growth in EBITDA. Book-to-bill was positive. Backlog grew to $2.4 billion, not counting over $3 billion in unawarded delivery indefinite quantity contracts, IDIQ contracts in our government business. Numbers aside, during the quarter and early here in the new year, we've announced some important execution milestones and wins.
The agreement to purchase the remainder of our EBI business in Europe from our joint venture over there, which we think helps us establish a good market position and helps reduce the risk in our fixed broadband business as we prepare for ViaSat-3. The announced acquisition of RigNet, which we've been looking at for a while now, and that helps jumpstart our growth into several key remote enterprise sectors ahead of our ViaSat-3 launch. Our new IFC win with Delta was pretty exciting, and it's an entry point with a major new, really high-class customer. A new distribution partnership in SKY Brasil as we complete the rollout of our fixed broadband offering in South America's largest market. These developments all underline our focus on execution ahead of ViaSat-3 as we grow our global footprint and build out a diverse set of existing and new markets.
Speaking of ViaSat-3, we continue to make really good progress on the program, and we're moving closer to delivery of the first payload. The record COVID spike seen in the quarter, especially in our Arizona, where our payload facility is located. Those things haven't really worked in our favor. They've caused some intermittent work delays, both internally and with our supply chain. Here in this recent quarter, mostly internally. That's unfortunately moved our first launch into the first quarter of 2022. Us and Boeing will be hunting for schedule efficiencies to make up for as much of that as possible in the growing months. We had some progress that performed ahead of our schedule. Had the COVID stuff not impacted, I think we would have had an improvement in our schedule.
We really remain excited about the big picture opportunities for ViaSat-3 program, and we'll open up the company for our shareholders. The performance on that satellite is we're well over 95% complete with just a couple remaining things to get done, so we should be done here in a few weeks for the delivery to Boeing. With that, let's get started to that first question.
Thank you as quicky remind to ask question you need to press star the one on your telephone to withdraw please press the pound key. Our first question comes from the line of Phil Cusick with JPMorgan. Your line is now open.
Hi, guys. Thanks a lot. In your shareholder letter last quarter, you highlighted the strong government backlog to drive sequential revenue growth into year-end. Now we're talking more about friction with the administration. How should we think about that? In the past, under administration changes, you've seen a sequential decline from December to March, but nothing that's outside of your normal year. Since you're calling this out, should we be looking for something stronger?
No, I don't think so, Phil. We're seeing kind of some of the same trends. I think without the COVID thing, it would've been quite outstanding. That's why we say it's just a little bit impacted by the COVID stuff, and we're seeing that. An example would be COVID's impacted people from actually showing up and doing live witness of testing. That's an example where that test milestone is required for us to actually ship a new set of products. It's things like that that we're seeing that are unusual, but the rest of the business cycle is pretty similar. I think the only reason why we want to point it out is we just had such a great order flow. That order flow is not going to create an unusually high level of sales through the December quarter or through the March quarter.
Nothing really outside of the normal.
Is there any kind of product cycle in your government sales, or is this relatively the same amount of products, and any sort of ins and outs quarter to quarter is more driven by the vagaries in procurement than it is in what you're shipping?
Yeah, I think that's a fair statement. We haven't seen any weaknesses in any of our specific business areas on the government side, so I think that's a fair statement.
Okay.
The administrative transition is just more a normal course for what happens during a transition from one administration to another that we anticipate.
Okay. One more, if I can. Margins in sat services and commercial networks continue to trend really well. Can you unpack some of the underlying trends in those segments? There's been a lot of upside to margins lately. How should we be thinking about that going forward?
On the satellite services side, this is what we've seen every single time, ViaSat-1 and ViaSat-2. That we have some upfront costs that we experience as we launch those new satellites. When we cross that line of covering those costs, the margins have expanded the last couple of times, and that's what's happened here on ViaSat-2 where we've covered all the costs. The incremental margin, if you look at it year- over- year, is actually over 100%. More revenue has flowed through to margin than just the revenue increase. We've had a little delay in ViaSat-3, and so what we've done is we've been able to reduce cost in that area and generate operating cash. That's a normal trend that we've seen. On the commercial networks piece, it's a little bit different. There, we're getting ready to ship our payloads.
We had a bunch of R&D driving the margins in that business down during that development cycle. Those are phasing off. Is that right, Shawn?
Yes.
Should we be looking for that?
Yeah.
Should that R&D sort of bounce back up, or is this a good line for a while?
Yeah. I would say that I think we've kind of come down to a steady state from a percentage of revenues. It's been around this 5% of revenues. I kind of expect those levels.
Great. Thanks, guys.
Thank you.
Thank you. Our next question comes from the line of Ric Prentiss with Raymond James. Your line is now open.
Hey, good morning or good afternoon, depending where you're at.
Good morning, Ric.
Hey. Couple questions. First, as you manage capacity with the ViaSat-2 filling up and getting good margins, as Phil pointed out, you're managing capacity, but you're also starting to earmark some for the in-flight connectivity ramp. How should we think about the trend for broadband net adds versus how much capacity you need, and what kind of pacing should we think on that in-flight connectivity order flow?
We've had some additional orders subsequent to the quarter, pretty good orders from existing customers. We're starting to see that come back. The airlines are way more in control of that. We're obviously a follower of that. We have some models, some high and low and mid models. I think I don't think you'll see a lot of input, a lot of increase until we get past a certain point just in terms of the vaccinations and that. We're seeing it come back, for sure. I think on the consumer residential broadband side, Mark, you might want to chime in here. We're not seeing a slowdown in demand. What we're seeing is people are choosing to upgrade services and expect that to actually slow a little bit as well.
We're making trades for higher and upgraded plans for subscribers in some of those areas while preserving capacity in all those markets for IFC growth. We'll make a trade. We'll probably see slight reductions in subscribers for a little while. We've got some ideas and things we're trying that could offset that, I wouldn't want you guys to put that in your models here in the near term. We have some ideas about improvement. For right now, I think we expect some continued slight decline in subs.
Okay. The ViaSat-3A is slipping out to calendar year first quarter 2022. How fast can you get that in service? Remind us, is it still like a six-month lag for the 3B, and how about 3C?
The first one's going to be probably the longest cycle of launch to in-service, because this is a very complicated overall system, and we've got to go through pretty thorough testing. We'll launch with a minimum amount of subs for a while and test, but it'll be several months before we have a real full launch on that satellite after it gets in orbit. We expect a pretty short cycle time to get in an orbit after we launch versus what we had last time. Very short. On the order of 30 days type thing before we're ready to go. That helps that overall period. We'll be launched and in service faster than we were launched on orbit last time, is what we expect. The second one, it's in the order of five or six months behind the first one.
That's what we're expecting. The third one may be a little closer than it was before behind the second one. Mark, do you want to chime in there at all?
No, I think you covered most of it. The main thing I'd say, Rick mentioned at the beginning that we're trying to do things to recover schedule. Those areas where we can do integration and test in advance, which is what we're doing now, give us opportunity to bring in that schedule.
Okay. One bigger picture question. I got to tell you, probably the top question we get, and you guys probably are, too, is the whole LEO question. SpaceX Starlink, you've got now AST & Science out there talking with folks. The RDOF auction's now over, quiet period, thank God, finally ended. What can you share with us as far as your view of the addressable market, the competitive landscape, and what the future kind of looks like? I know it's a big picture question, we get it a lot.
Okay. One part we can deal with really quickly is the AST thing is direct to mobile. It's not a competitive service to what we're doing. We don't see that as impacting us, and we'll just put that one aside for a second. On the LEOs, and the RDOF, I think we anticipate a trend where there is more build-out in the market, more due to RDOF than to other satellite competition. One thing, with RDOF, there's still a lot left to be determined because there's a lot of concern that a lot of the money was won by technologies that are unproven and may not be able to meet the obligations that are associated with them. There's still some steps to go before those things are finalized.
I think from our perspective, to the extent that they are awarded and don't fulfill their obligations, that certainly creates more. It just leaves a bigger market for us. On the other hand, the one thing I would say is we have had a service offering that's been positioned relative to the market the way the market has been and the way it is now. We have a lot of maneuvering room in terms of how we do go to market, and there are many people that have terrestrial services that are not happy with them in one way or another, whether it's price, their ability to support certain types of applications, or the service that goes along with them, where we believe we can compete.
The main things I would say is our expectations, our plans for service, are in the low single-digit percent of the U.S. market in those markets we go into. Here in the long run, we don't see those types of addressable markets changing for us that much as a result of these things. The way we go about those markets, that might be the service plan, the specific service plans we have, the specific price points, or the way we go about it certainly could evolve. I think you'll see that evolve. It makes complete sense for us to do that in response to changes in the market as they unfold over what we think is a pretty long period of time, three to five years. That's kind of what our view is. We see that we'll evolve. We always have.
We've evolved from where we started.
Right.
[crosstalk]
I'm sorry.
Okay, good.
Yeah.
Mark made a really good point there that I think shouldn't be lost as the reason why we've been going after higher ARPU, and higher quality services is, I mean, we have limited capacity, and we have more demand for capacity, so that's what we did. One thing is, if the situation is different than that, we'll have a different tactic. People shouldn't just expect we're just going to go after higher ARPU things all the time. We're going to do what makes sense.
Right. Yeah. One other thing I'd add is right now, we're in a very intense capital spending campaign relative to our revenue and earnings. Optimizing for cash flow makes a lot of sense. We don't expect to be in the same position after ViaSat-3's launch, so that creates more maneuvering room as well.
Yeah.
Great. Thanks, guys.
Thank you. Our next question comes from the line of Louie DiPalma with William Blair. Your line is now open.
Great. Rick and Shawn, good afternoon.
Good afternoon.
Hey there, Louie.
Also good afternoon to you, Mark, as well.
Thanks.
I was wondering, you announced the Delta win, and for the economics of the hardware installation for Delta, are those economics similar to the economics for your other recent customer wins for the in-flight connectivity?
I mean, you could assume that. In other words, if what you're asking for is are we selling hardware to them and then providing services? Yeah. I mean, we don't go into the details of every customer, but the answer would be yes. It's fairly representative of our offerings. Each offering's unique, but the overall thing you should assume, yes.
Great. That is helpful. On the same in-flight connectivity topic, do you have an update in terms of the total number of aircraft under contract that you have now, if you include the Delta planes? You also mentioned subsequent to the end of the quarter that you won other orders from your existing customers. I was just wondering if you have a count of your total number of aircraft under contract.
Yeah. Through the end of the quarter, it was a little over 1,000 backlog.
Great.
You could assume it's grown a little bit since then.
Yeah. We have about 1,500 now, so the total under contract would be close to 2,500.
Great.
More or less.
Thanks. Thanks, Mark and Rick. I was also wondering, do you have updated thoughts on what free Wi-Fi could mean for average annual revenue per aircraft relative to what your current average annual revenue per aircraft is for your existing base? I know this is a future catalyst, but investors are wondering what could be the implications post-COVID as airlines switch to a free Wi-Fi model.
Basically the way to think about it is that the variable revenue that we get really is going to be related to the number of passengers that use it, and the types of activities that they use, and the way the airlines choose to deal with that. A lot of that discretion goes to the airlines. The general trend, and we've seen this trend anyway, is that if you put a really good Wi-Fi service on, engagement tends to go up, and that tends to drive additional revenue, in one form or another. The exact amounts will vary depending on the airline and the type of service they choose to present and the other types of partnerships that they or we create to engage passengers. It's very hard to be specific, but the general trend is up.
Thanks, Mark. I have a question related to your government business. You have referenced that you have $3 billion in government/defense IDIQs. I know that $3 billion, I think it represents the ceiling value. I was wondering, based on your conversations, how much of that ceiling value do you expect to convert the actual task orders and contracts over the next several years?
There's a saying that past performance doesn't guarantee future performance, so I'll start with that. Our history is essentially we consume the ceilings of the IDIQ contracts we have. Very rarely do we not. We're expecting that. That doesn't guarantee that's going to happen, because they're a little bit unique, and they're different than each other. We've had a really good history of fulfilling those. Like I said, that doesn't guarantee the future. That's kind of what we expect.
Thanks, Rick. One last one. Mark just talked about the intense CapEx cycle that you're going through right now as it relates to ViaSat-3. I was wondering, can you give us an update in terms of how much of the ViaSat-3 CapEx you've completed, and how much remains?
Yeah, sure. I can take that, Louie. Right now we're over halfway through the program. You can think of that as over $1 billion is behind us. We're pretty far along that. I think one of the other things to kind of put around that is if you look at our liquidity position today, strong operating cash flow is like $700 million the last TTM, that we're in a really good position. We're basically fully funded to finish the build-out.
Great. For CapEx, like this quarter, hypothetically, are you able to break out what the CapEx would have been this quarter if there was no ViaSat-3? In other words, what is the CapEx that you incurred this quarter related to the other portions of your business, whether it's the customer premise equipment or maintenance CapEx? Are you able to break that out?
Yeah. I think the best way to maybe look at that, Louie, is just kind of look at that TTM cash flow that we had. If we had taken our ViaSat-3 spend out of that, our free cash flow for that same 12-month period would've been about $275 million. That's probably the best way to kind of size it up.
Yeah.
Again.
Yeah, that's what I was looking for. Thanks, Shawn.
Yeah.
Thanks, everybody. That's everything that I have.
All right. Thanks, Louie.
Thank you. Our next question comes from the line of Mike Crawford with B. Riley Securities. Your line is now open.
Thank you. With the ViaSat-3 constellation and the terabit per second capacity of the three satellites, I think you're going to increase your total bandwidth you have to sell about eightfold. What would be some rough approximations of what that selling those bits could mean to revenue and EBITDA for Viasat today, where you're looking a little over $2 billion in revenue and around $500 million in EBITDA today?
Go ahead, Mark.
Yeah. It creates a lot of opportunity, right? One of the things that we've said in the past is that part of the way that we've been successful in competing is that we get these productivity gains, right? We'll have eight times the bandwidth at a lot less than eight times the cost that we've put in place so far. It'll be like double the cost, or double the CapEx that we've spent so far. We'll end up offering a fair amount of the productivity gains to our customers, and that'll vary depending on the market and the way that we bring the bandwidth to market, the terms of the contract, all of the support that goes with them. It's very difficult to give an answer.
One of the things I'd throw out there as ballpark, if we gave substantial productivity improvements, we can still increase our revenues pretty significantly. I don't really want to. It's not fair or meaningful to put out a single number, but you can just look at what's happened over the last eight or nine years where bandwidth consumption has grown by close to a factor of 10 per capita consumption over about the last 11 or 12 years. We need to be prepared for that pace of growth over the next decade as well. Actually, we look at that as an opportunity, not a problem, because as long as our productivity improves faster than other options, then we compete really well. Then, it just depends on what goes on in the market, how we deal with it.
I don't think we want to put out any specific numbers. The one thing I would say is, back in September when we did our annual shareholders meeting, we did put out a five-year forecast which showed that in aggregate, we think we can around double revenue over five years. That'll give you a sense of how that product, to the extent that some of that's satellite services, not all of it's satellite services, we did give a breakout of that. You can get an implicit view of how we think we're going to apply the bandwidth and what that means for revenue growth. That's probably the best way to interpret for now.
Okay. Thank you, Mark. It seems, under that construct, that after the ViaSat-3 EMEA satellite starts to load, that that's around the time when you flip into generating positive free cash flow, even if you build another satellite every year and maintain this CapEx regimen around somewhat just south of $1 billion a year. Can you remind us what types of capacity we might see in ViaSat-4 and ViaSat-5 types of satellites that you're currently contemplating?
Yeah. That's a really good point. I think the point that this per capita consumption's going to continue to grow means you have to have a plan if you want to maintain competitiveness, and just to price your service offerings on a go-forward basis, you have to continue to improve productivity. We've talked about ViaSat-4 a couple times over the last few quarters, and we're working on the detailed design of that, and what we think is somewhere in the range of five, six, seven terabits is what's possible there. There are still trade-offs associated with that's kind of the range. Think of it maybe a factor of five-ish, 5+ improvement relative to ViaSat-3. The next generation it's a little bit harder to have clarity on it, we think that, and ViaSat-4 is based on design work. That's where we are now.
ViaSat-5, it's based more on conceptual stuff, we think there's another factor or two that can be gained, at least beyond that. Altogether, that means we've got an order of magnitude of productivity gains in front of us that we're working on.
Okay. Thank you.
Mike.
Yeah.
Mike, one of the questions, one of the points I just don't want to go unnoticed is, you're right. After we launch the second one over the EMEA, it's two or three quarters after that, and we think we'll go free cash flow positive, and as a company, we expect to stay there after that. I think it's a really important point regardless of our build-out, unless we do something completely different that I don't know about.
Okay. Thank you, Rick. Last question is, if a LEO constellation of, let's say, 2,000 satellites, was circumnavigating the globe at 550 kilometers, like Starlink, how many of those, at any given point in time, would be over the U.S., and then how many of those would be visible from one point on the ground, say, over Chicago O'Hare Airport?
Okay. The exact answer to that question depends on the orbit selection. What Starlink and Amazon have done, and not everybody has done, and not all of their satellites follow this, basically, they've tried to incline their orbits in a way that the satellites that don't cover the poles spend more of their time over the mid-latitudes, where the U.S. is. Think of it as 6%, 7%, 8% of the satellites would be within sight of the U.S. Even that depends a lot on the specifications of the ground terminals, because the fact that the satellites are in sight of the U.S. doesn't necessarily mean that the ground terminals are in sight of the satellites, and can do that with the latency specs that they have. There's still some complications there.
Just the fact that they're over the U.S. doesn't mean that they're all over places in the U.S. where there's demand. That's the other factor. For your question about a place like O'Hare, there are regulations around what's called EPFD, which is equivalent power flux density, and the whole point of that is to prevent non-geo satellites from putting an amount of power into a specific location that could cause interference to geosynchronous satellites. The answer to how much of a bandwidth they could bring to a particular place depends on the specifics of that, but it definitely puts a bound on the amount of bandwidth that can go into small locations.
The specific answer, again, depends on the orbits and the look angles of the satellites, but with 2,000, you can think of it as on the order of 10-ish satellites that could be in sight of a particular place in the U.S., but they may not all be usable for a variety of reasons, depending on some of these specifications.
Okay. Thank you, Mark. Thank you.
Thanks, Mike.
Thank you. Our next question comes from the line of Chris Quilty with Quilty Analytics. Your line is now open.
Thank you. Wanted to follow up first on Delta. I know with American, you guys had a crazy fast rollout of antennas to the aircraft. Should we expect that type of installation, or will this one be more gradual?
It's fast, though. They want to get the ones they've identified outfitted quickly. We won't have at the same level. You remember we had quite a few more aircraft initially awarded by American than we do here. The rollout was very rapid. We'll go as fast as Delta wants to go, but it won't be quite as fast, but it'll be definitely a pretty big increase from where we are.
Okay. I guess part of that is, of the 300+ aircraft, what is new aircraft versus retrofits, and presumably they've got new aircraft that are parked that are easy. Have you given a breakdown of what you expect the penetration to be of new versus retro?
We haven't. I mean, we have that, Chris. I'm not sure that we've got that cleared with Delta to release.
Okay. Final question on Delta. Somebody kind of knocked around it, but the ARPA for the aircraft, given that this is going to be an all-you-can-eat plan, is it going to overall impact your reported revenue per aircraft substantially from what you're Well, okay, right now the revenue per aircraft is down because of COVID, but if you compare it on sort of a nominal, regular basis, is it in the same ballpark of what you've been generating historically?
Mark said earlier that it depends on what type of third-party deals are done on either side. It depends on if they go free and how fast, and there's a lot of factors in there, Chris. The other thing is the airlines are very sensitive to us talking about ARPA. As opposed to some of our competitors in the past that have single point businesses, that the only way they kept investors happy was focused on ARPA, this is part of our satellite services business, so we blend that. I think our goal is to work with the airlines in a way that either going free or having really high penetration and high passenger engagement is a good thing for them and is worth either in a way that generates additional revenue for the airlines or something that makes this a net very positive deal.
Yeah, I think that's kind of all I want to say about that.
Okay. Significantly, I got it. Oh, I guess on that same note, you guys had historically provided the backlog, and I didn't see it in the shareholder letter. I guess either that or the consumer ARPU. Are those numbers you intend to give on a go-forward basis?
Things are changing pretty fast, especially with the onboarding of the European business. That's definitely at a lower ARPU than kind of where we are currently in the U.S., bringing Brazil on. We're working through how to best disclose those things. We've said many times. For instance, back to your previous question, we think satellite services revenue and EBITDA is going to grow next year, and it's going to be a blend of things as in-flight connectivity comes back and also said kind of expect our residential broadband in the U.S. to decline somewhat, slightly. Things are changing pretty rapidly and just like we did this last time, we were really focused on subscribers. Subscribers didn't grow that much. ARPU grew a lot.
We'll make whatever changes make sense so that the overall economics are we're getting the most out of it that we think we can get. We're sorting through what type of disclosures we're going to have in the future.
Okay. Because that was a question I was going to ask Shawn is once you take on the European business, you're going to bring on a slug of subscribers and obviously a very different ARPU, and how are you going to report that?
Yeah.
Hopefully when you start the new fiscal year, you'll have an entirely new set of aligned metrics, possibly with RigNet folded in as a different business segment or something that we can follow on a go-forward basis.
I think that timing is appropriate, Chris.
Okay. One question on the commercial networks business. You guys have been really strong in the antenna business, kind of a boring large diameter business. What's going on there?
We had another actually just really good new business quarter in that business. It's usually through another prime, but we do a reasonable amount of government business in that segment, too. That's helped contribute to future growth there. We have some very good opportunities in front of us as well. It continues to just be a good growth area for us with very little invested capital.
Chris, keep in mind, that is where we report the product sales for the airlines as well within that segment looking forward.
No, I know. The language in the last Q, or maybe it was in your letter here, talks specifically about some of the large diameter stuff. I thought it said the Earth observation customers.
Yeah.
Right. Yeah. Basically, one of the things that's driving interest in space is, besides communication, is Earth observation. Generally, that's kind of the way you get really high throughput on these Earth observation satellites is by having large apertures on the ground. Because you want the satellites themselves to be inexpensive. Unlike in the communication space, where you have way more Earth terminals than you have satellites, in this case, a lot of times you have more satellites than you have Earth terminals. It's a really interesting field. I think there's lots of opportunity there, and I think we've got a leadership position in that. That's part of what's driving the results there on both the commercial and the government side. There's more to it than just the large apertures as well.
I think there's more stuff that we're working on, and we have also a shared service initiative that's doing really well. That's we call Real-Time Earth. I think that this notion of real-time access to Earth observation data is going to be a growing phenomenon as well.
Good. Speaking of little antennas, when you look at the ViaSat-3 rollout, you don't call them gateways, you call them something else now.
Yes. Satellite Access Nodes or SANs.
Yeah. SAN. Okay. When do you begin the rollout of that infrastructure in anticipation of ViaSat-3? Does that show as a significant CapEx movement in any way?
Yeah. This is Rick. We've already started. As you can imagine, the first thing you have to do is locate those points, make sure you have fiber there or get fiber brought there, get power and get leases executed, get all the regulatory approvals for each one of those sites. I mean, there's a process to go through here. We're well into that process here in the U.S., and we're actually well into that process in Europe and have begun that in the Asia Pacific region. We've already begun to incur expenses associated with that in front of including fiber leases and that. In addition, we actually have some more gateways up and running, and we're doing some tests of various elements of the ViaSat-3 network over some of our existing satellites in preparation.
Like Mark talked earlier, to the extent that we can do some concurrent integration and testing, we can help accelerate those things later. We're currently doing some of that today. We're well on our way here for the rollout of ViaSat-3. We've begun to incur expenses in the low millions of dollars per quarter in that, and that's going to increase quarter by quarter as we prepare for ViaSat-3.
Got you. Final question on commercial networks for Shawn. I mean, the margins in that business have improved by 30 points since the beginning of the year, and I know you've said the R&D is going to flatten out, the internal R&D. That element we kind of know. Modeling that on a go-forward basis, can you drive that business to an actual profit on a go-forward as you begin to ship all the mobility antennas and all these ground equipment, or should we still model that as a loss leader?
Yeah. Probably a couple things.
Well, first of all, loss leader isn't a good example.
Yeah.
That's a whole different type of-
Where we put our R&D. Really.
Yeah.
Yeah. I think there's a couple things as you're looking out. One is, as I talk about, that is where we put the terminal deliveries for the IFC. As Rick was talking about on the Delta program, we'll expect to see a lift there. That'll help that segment. We think our R&D is hitting kind of a baseline rate, but I'd expect that to stay there as we grow. I think we'd expect improvements on a year-over-year basis looking to next year. We're going to continue to invest, too.
Got it. If I can ask one final question on the consumer business. Most of your growth, your incremental growth in subscribers has been in Latin America and Brazil. I know SKY Brasil is talking, you're looking at expansion into other parts of South America. What do you have for capacity in the region right now? Either on satellites or in terms of incremental subs that you think you can grow in the region, before ViaSat-3 hits the market.
It's going to be limited. I mean, we're working with, obviously, with a Brazilian satellite down there today. In Mexico and Central America, we're working on ViaSat-2, all the way down, a little bit in Colombia, all the way down to kind of the top part of South America. It's reasonably limited until we get ViaSat-3 for the Americas on board.
Yeah. The other thing is, we have to portion that bandwidth among all the different markets that we have. We have in-flight customers. We'll have remote enterprise customers as we get RigNet. We just have to allocate the bandwidth. We're just in a situation where there's a lot more demand than there is supply.
Very good. Thank you.
Thanks, Chris.
Thank you. Our next question comes from the line of Giles Thorne with Jefferies. Your line is now open. Giles Thorne, your line is now open.
I'm just happily sitting here talking to myself on mute. Sorry about that. It happens a lot. My first question is, would you be willing to tell us, back in early 2017, when you were putting the investment case together for ViaSat-3 and ViaSat-3 EMEA, what your target subs were for the point at which that satellite came online? Where are you today on the sub count?
One is, like I said, it's a dynamic marketplace, and it would be very presumptuous for us to pick a subscriber number. What we're more focused on is how much revenue can we get in total and what's the return on that revenue, and does it justify the CapEx. One of the things you keep working on is having multiple markets. I just described sort of what's going on in South America, where, because we have multiple markets, the aggregate demand for bandwidth is greater than the supply, which is a good thing, right? That's what we want. That's just what we're doing in Europe, too, through our government work, through the remote enterprise stuff that we can do with RigNet, through the in-flight connectivity market.
A lot of those markets are doing quite a bit better than what we would have projected back in 2017, as an example. That's the situation we're trying to create. The short answer is, we don't have a subscriber. We don't have a residential subscriber target, and we will dynamically set one in the context of all the demand among all these different services. That's basically how we've done things.
Okay. Just to push on that. The success of ViaSat-1 was having WildBlue there ahead of it, and the objective of the venture with Viasat was to have as many subs as you could. I don't know. It doesn't feel like you didn't have a subscriber number in mind. That's fine. What is the subscriber number now?
Okay. I just want to go back on that and the history on WildBlue. What WildBlue really got us was distribution, back office. It enabled us to do retail. Once we had that retail, one of the main reasons we did the acquisition was because we wanted to rejigger the way that we went to market, based on the things that we learned about the market from having the WildBlue experience. Remember with ViaSat-1, I have to go by memory here a little bit, but we basically had 10 times the bandwidth. We probably ended up with two to three times the subscribers and 3x the revenue. The margins were much, much better because of the productivity of the satellite.
The things I'd say that will apply to Europe is these are the things that we're doing now and will do with the KA-SAT, and the capability we have there is all the back office stuff is more complicated. The support stuff's more complicated because of the diversity of countries and languages. We're going to be able to test that. We also need to make sure we have the distribution and support channels we want in place, and we'll be able to do a lot more experiments regarding the service offerings and the way we bring those to market. We expect to get the same types of knowledge and experience, and information through this, as we did with WildBlue in the U.S., although we're starting from a different place. It still doesn't make sense. It doesn't make sense that we have a subscriber target number.
The way I describe it is we're going to do it in the context of all the demand for all the services.
Okay. It sounds then, Mark, like you are going to be building out what it is that you thought you were going to get by working with Bigblu . You're going to be doing that yourself, so that there are aspects of the customer journey that you weren't going to do, but you are going to have to build that out now.
Well, that was our plan. If you look at the reason for doing a joint venture with Eutelsat was that we were going to lead the retail part of it. Bigblu , they were working with Eutelsat at the time. They would've been a component of that, but there's other ways to achieve that as well. We don't look at that as a decisive issue.
Do you have any mechanism to prevent?
I would say, as a matter of fact, we have been investing in this entire customer journey side. This is a good way to put that to work ahead of the ViaSat-3 launch. To me, this is a really good way to put our own tools in place, which we've already begun to do in a couple of those markets. Spain is one and Norway is another. I think this is just a really good opportunity for us to get those things in place prior to the ViaSat-3 launch.
Rick, is that what you're talking to when you say buying in Big Blue should accelerate and improve momentum around fixed broadband in Europe?
No. I think that we've been developing end-to-end customer journey tool set that's far more digital and being able to roll that out, for instance, in some markets that like we have limited capacity in Europe on KA-SAT, but ahead of ViaSat-3 launch, where it's going to be a big launch, is a much safer way to begin that process in a way that we can rapidly roll things out, learn, and iterate.
Do you have the subs that you bought in with EBI, is there a mechanism to stop Eutelsat taking them from you once Konnect VHTS comes online?
It's a little hard to hear you, [crosstalk].
Couldn't understand that.
Could you say that one more time?
Yeah. Sorry, forgive me. I'll come a bit closer to the phone. If I understand correctly, with the purchase of EBI, you now own completely a set of subscribers, and by the way, it'd be really useful to know how many subscribers. A set of subscribers that are legacy wholesale to Big Blue. I'm wondering, with Big Blue now firmly in bed with Eutelsat for Konnect VHTS, do you have a mechanism to keep hold of those subscribers and stop Big Blue migrating them at some point in the future onto Konnect VHTS? Is it like a non-compete, something like that?
There are elements in our agreement with Eutelsat that deal with what you're talking about specifically. We talked about a little bit of it in the way we announced the deal, but there are more embedded elements that deal with that specifically. Yes, we have a structure that I'm sure will be imperfect that deals with any type of poaching of those subscribers. We also have economic remedies that deal with those types of things. Honestly, it's a component of this. It is by far not the most important thing. I'd say that by far the most important aspect of this is getting a running start for ViaSat-3 and accelerating ramp into a region where we think the demand is going to exceed the supply of our capacity over there.
Yeah. I would not interpret this as us battling with somebody else to keep subscribers on comparable plans.
Right.
I think the real battle here is going to be what types of service plans can you offer, how do you distribute them, what are the pricing of those plans, how much demand is there, and actually the other really big thing is going to be what's the trajectory of your future supply of bandwidth, because that has a big influence on the types of services you can offer. This is kind of a set of initial conditions, and to Rick's point, there's economic consideration that's baked into the agreement that deals with the issue that you said, but it's not the main event at all. It's a sidelight.
Final question, sticking with the same topic, what is all going to be for the go-to-market strategy going to look like? Just what's the latest thinking there, please?
What is the go-to-market strategy? Are you talking about on the residential side?
Yeah.
Okay. Yeah. Remember, residential is only one component. Basically what we're going to end up doing is we're going to define very different service plans. I think we'll be able to deliver a more consistent quality of service for the plans that we do have. We'll come up with pricing and terminal strategies that are consistent with those. It's a little bit premature to say what those are, because we want more experience in the market, and this is what's going to give us the opportunity to do that. Clearly, the European market is very diverse, more so than the U.S. market, and there won't be a single point solution that we apply all across the continent. It's going to vary by market.
One of the things that we're getting better and better at in the U.S., and we're applying the same things to Europe as well, is really marketing on a more localized basis. Because we can. Because we can deliver different types of services and different price points into different specific market sub-segments pretty easily with our system.
Yeah, the last thing we're going to do, Giles, is tell the people that we compete against exactly what we're going to do right now. That's [audio distortion] .
Yep.
Fair enough. All righty. Thanks. Thank you very much.
Okay.
Thank you. Our next question comes from the line of Mathieu Robilliard with Barclays. Your line is now open.
Yes. Good morning, good afternoon. If I could start with a question on the government side. As you pointed out, and we've been used to it is a volatile segment, but the trajectory has been very positive over the last many years. I was trying to see how we could frame a little bit from our side, on the analyst side, what is the potential market you're going after. I don't know, is it a market where you expect to gain market share from other players? Is it just new application, new services that are developing? If you could frame in any way, how big is the opportunity there, that would be super helpful.
One thing, again, no guarantee that past performance equals future. We've had a lot of years of really good, consistent growth. I think we've only grown, maybe not grown a couple of those years. It really is by consistently going into new and bigger markets and environments. We've moved from Tier 2 to Tier 1, to moving to tier 0 in some of those applications. We've been able to develop and rapidly deploy products that are around the programs of record in a way that's really helped our growth. We're working on the things security and secure tactical communications and things that are absolutely critical to actually accomplishing what the warfighter needs. That market is expanding globally through joint warfighter concepts. Just from a position standpoint of needs, we think we're in really good markets.
Security, tactical comms, strategic comms, hybrid networks that integrate line of sight radio and SATCOM networks and multiple satellites in that same network. The cloud really is just very coming into play in that which we think only strengthens what we're bringing. We really like this government communications market, and it's been a good one, and I think the ViaSat-3 constellation only enables further growth, not just in the global SATCOM piece, but in the other pieces that we bring.
Yeah. I think just another way to think about it is what's driving the growth in the defense business is very similar to what's driving growth in commercial businesses, which is what used to be just voice communication now is imagery and computer to computer. There's more things computerized. One of the ways to look at our growth is, can we get satellite antennas on more platforms than we used to? That's happening for sure. You can see we've announced lots of different platform wins, whether they're ground or helicopter or different types of aircraft. The other is the amount of data that those platforms get. The other dimension that Rick described is, where it used to be acceptable just to encrypt a link, encrypt a communications link, now you've got to worry about trust and cybersecurity. That's another component of our growth.
The other thing, the third area is the terrestrial radio links, whether it's tactical data links for others. Those are the three big thrusts are satellite, cyber, tactical radios, and then moving a lot more data. Those are really big picture trends that we have good competitive positions in.
Thank you. That's helpful. I had a follow-up question on the comments you were making about ViaSat-4. If I heard correctly, you were talking about a capacity of 5 TB/s-7 TB/s . I wasn't entirely clear if you were talking about one single ViaSat-4, which would seem like a material leap, or if you were talking about a couple or three satellites for this kind of capacity.
That's per satellite. One individual satellite. Then we could have multiples of those satellites if that's what made economic sense. Yeah, it is a big leap, and I think that one of the things that everybody needs to take into consideration when they look at all these different approaches to broadband from space is how each of these things compare relevant to the others. I'm going to just throw out one other thing, which is that when you're doing geosynchronous satellites, it's very straightforward to increase frequency reuse and increase capacity on individual satellites without affecting either our other satellites or anybody else's satellites in the geo arc. One of the big issues with non-geosynchronous is that every non-geosynchronous satellite can interfere with every other non-geosynchronous satellite, whether your own or somebody else's.
That's one of the things that's really important in those filings is understanding how much spectrum they include in their filing and the way that that spectrum can interfere with other systems. That's one of the biggest issues that's going on now in these non-geosynchronous ones. It's very, very difficult to increase your capacity without impinging on some other filing. What we think is not only do we have a productivity advantage now, but that we'll be able to expand that productivity advantage over time.
Great. To follow up on that, it is a question that was asked before, but obviously I think for a lot of investors, a question mark in this industry is where does the CapEx race end? Technology changes, you have to adapt, you're looking to lower your costs, increase your market potential. That always has to be balanced against free cash flow at some point, right? The way you think about it, as you expressed it before, which is really from once ViaSat-3 effort is behind us and you gain scale, really your objective is not to be free cash flow negative anymore, and you think you have enough scale to continue to invest as you should?
That's right. Yes. Yes, that's the answer. Just to be clear, and you sort of pointed this out with the ViaSat-4 thing, customers don't care how much CapEx we spend or how many satellites we have. What they care is how much bandwidth we have. The way you manage your CapEx is to get as much bandwidth as possible per CapEx dollar. That is the entire point of what we're doing. ViaSat-4 is a really good example of that.
Excellent. If I may ask just a last one, just on the airline side. I hear people talking more and more about free Wi-Fi. I didn't fully understand if in the case of Delta that was going to be the offering, but in any case, that seems to be a big discussion, free Wi-Fi in airlines. At the same time, you look at the industry and obviously it's offering, it will go back on its feet, but it's going to take some time. I'm wondering how realistic is it to assume that all the airlines that started are going to offer free Wi-Fi in the next few years when they have so many headwinds to face? Is that your working assumption when you model-
First of all, we should state that in no way did we say today that Delta is going free or anybody else is going free. We're not stating what the airlines are going to do. I just want to make that clear.
Sure.
Yes. Totally agree. The other point is, really, in a lot of ways, all this is up to the passengers, right? If the passengers express a preference for airlines that have free Wi-Fi and some airlines have free Wi-Fi and can leverage that preference in a way that's more profitable, that's what will drive the adoption of free Wi-Fi, right? That's what's really at issue. We've worked with airlines that have been able to do that, and to do it in a way that's economically profitable for them. I'd say it's a tricky thing to do, and it will evolve as airlines compete with each other. We're going to take our lead from what the airlines do.
What we think is we're learning a lot and that we can offer suggestions to the airlines that will help them drive passenger engagement at lower cost to themselves, or even in a way that drives enhanced revenue. The other point is, the way we've been able to do that is through the same argument that I just described before, which is high productivity, which means that we can give them a lot of bandwidth value. Those are the ingredients, and now we have to see how the market responds.
If you're an airline and you don't have a choice because your provider actually can't provide the capacity to enable that's not a good position to be in.
Right. Yeah.
Yes. Makes a lot of sense. Thank you, guys. That's very helpful.
Great. Thanks, Mathieu. We got last question for this. Anyway, last question.
Our last question comes from the line of Simon Flannery with Morgan Stanley. Your line is now open.
Great. Thanks for fitting me in. If I can continue on the IFC, obviously nice to see the Delta win. Could you just talk about the opportunities that you see out there in the medium term, to continue to take share in that business? Expanding aviation into, I think you've talked in prior earnings calls about the opportunity with government in the military to do more on this front. I think you had an announcement about some new antennas on the business aviation side. Any color about the kind of the opportunities there to grow into that TAM? Thanks.
Yep. We've talked a little bit before about that, a little bit surprising, but the amount of discussions we're having globally didn't really decrease during the pandemic. Some of the actions people were willing to take immediately, there was a pause in that, obviously. A lot of ongoing discussions. As that business took a pause last year, the point of time at which ViaSat-3 is going to come on globally gets a little closer. As people are making decisions to buy airplanes, that's falling closer into the time in which we'll have global coverage. That's helping the discussions, and I think that's where a lot of the growth will come from, is in international over the next few years. We know kind of the most of the growth in the airline industry is going to come in some of the international markets.
Being there with global coverage is going to be a really important factor for us. A lot of that isn't really just taking share. It's connecting new aircraft that have never been connected in some of these airlines that they're ordering and in the growth of aircraft that will be ordered in some of those new emerging markets. A couple of things we're doing is we're continuing to make real progress in, you mentioned one, there was an announcement out about our demonstration of our phased array antenna on a business jet. We continue to invest in new technologies that will hopefully allow for shorter installation cycles, lower cost, better quality, all those things. We share those roadmaps with our airline partners and potential new airlines.
Just continue to lead investments in this area, I think, in a way that enables not only-- because right now we have airplanes flying on our satellites and lease satellite capacity, so allows for roaming into other networks. That's an important aspect in what we're able to offer.
Yeah. The other thing I would add to that is we've been very successful in the North American market. I think that's been our best market, partly because that's where we had the most coverage to start with. That's also the market where penetration is the highest. That is, more planes have Wi-Fi in that market, the domestic U.S. market, than any other market. I think that the airlines have become the most knowledgeable about the role that in-flight connectivity plays in their overall competitive strategy. The other things I would mention is that, the times when we don't win, when we're not successful in the market, it's usually because somebody discounted equipment or an install or dangled some offer that they weren't able to fulfill because they, as Rick mentioned before, they didn't have enough bandwidth to really scale the offer.
I think that the fact that we've been very successful in the North American market is going to help us in these international markets where the airlines have less experience about the role of in-flight connectivity in their overall value proposition. I think people are noticing that. I'd say we're really optimistic. A lot of growth is going to come in that Asia-Pacific market. The thing that's really notable about that is that so much of the travel routes are over oceans. That's where the ViaSat-3 value proposition is really important because that'll give us much more bandwidth over oceans than anybody else. I think that's what creates optionality for the airlines. We're really optimistic about the growth prospects, and so far, the responses that we're getting back from the market are supporting that.
Thank you.
Okay. If there's no other questions, I think, Paul, that most of the questions that we had from investors were actually covered just in today's conversation. Thanks everybody for coming in. If you guys have some feedback on this format, providing the letter a little early and giving some time or any type of other timing feedback you want to have, if you could reach out to Paul Froelich on our end or Jeff Matika. We'll take any comments and consider it into consideration. Thanks again, everybody.
Ladies and gentlemen, this concludes today's conference call. Thank you for your participation. You may now disconnect.