Iridium Communications Inc. (IRDM) Earnings Call Transcript & Summary

Hi, everyone. This is Anthony Klarman from Deutsche Bank. I want to thank everyone for joining us here for the presentation for Iridium Communications. I'm glad to be joined today by the company's Chief Financial Officer, Tom Fitzpatrick. Tom, thank you for joining us here today. And I know that you had some high-level remarks you wanted to make before we start the Q&A. So I will turn it over to you, Tom.

Great. Thanks, Anthony, and good afternoon, everyone. I'll just give by way of background, description of the company Iridium. We are -- we operate the world's furthest-reaching communications network, satellite or otherwise. We cover 100% of the globe. And so our network reach is differentiating. The other element of our differentiation is our network architecture, it too, is one of a kind. It's described as low earth orbit mesh, that is the nature of our architecture. Like I said, it's one of a kind. It does things for our customers that others can't. It's what enabled us to our subscriber base to 1.5 million subscribers. And in our last -- most recent year, we had EBITDA of $355 million. Our -- the way we go to market is unique as well. We use some 450 partners to distribute our service around the world. This partner ecosystem is -- has deep domain expertise in the vertical markets that they cater to. And so they act as not just a very efficient way of distributing our services, but also as kind of a product development operation for us, where they recommend new products that we should bring to market based on the feedback they get from their vertical markets. So in 2019, we completed what was a 9-year construction project, where we completely ripped out and replaced our old satellite network with the brand-new satellite network. So the construction project was, like I said, 9 years in the offing. Well over $3 billion of capital expenditure to give us our brand-new network. We took our leverage up over that construction period to 5.8x. And so with the final launch in early 2019, it brought in a financial transformation at the company, where we went from a profile of significant use of cash to a profile of significant generation of levered free cash flow. And that, we think, is going to be the situation of the company for about 10 years. We think there's about a 10-year CapEx holiday, wherein our CapEx will average about $40 million a year over the 10-year period. And so when you consider that our EBITDA, which was sort of on LTM basis, around $360 million. We have no cash taxes until 2024. You put up that all together and you see a profile of significant levered free cash flow. We estimated the midpoint of our guidance for this year, it's like $1.67 in levered free cash flow per share. That represents about a 4% yield depending on where the stock price is. And conversion rate in excess of 60%. So we would encourage investors to benchmark us against the tower sector and other communications infrastructure firms, and we think we stack up very well against them in terms of valuation considerations. So with that, I think that's just the summary. I'd like to turn it over to Anthony for Q&A.

Great, Tom. Thanks very much for that. And by way of instructions, for the folks listening who are on the call, if there are topics that you would like to put forward for me to ask Tom, given that it's a -- just a one-on-one conversation here in listen-only for the participants. You have a text box, hopefully, on your screen, next to the video screen of Tom and I. You can enter some questions to me in the text box, and I will try to address as many of those as time allows. Tom, it seems hard to start any place other than COVID here. And it was an incredible year in 2020. You were off to a great start in January and February, and then the world came to a halt in March, just about a year ago this time. And when it first happened, you certainly dialed back the expectations of the market a little bit given the headwinds that you were seeing in aviation and maritime and some other areas. But if we fast forward to the end of 2020, you actually wound back basically where your guidance had started for the year. And I thought maybe a helpful perspective for the folks on this call would be to talk about what some of the early headwinds were that you saw happening in COVID in your businesses, particularly around maritime and aviation, even some of the personal use? How and what to extent you saw some of those headwinds abate throughout the year? And what your guidance for 2021 informs about a little bit of the return to normalcy a bit?

Right. So yes, when COVID hit, it hit us all kind of very abruptly in mid-March. What we saw initially was just basically the channel seized up. So what we saw in our legacy commercial voice and data business, our satellite phone business, was a significant drop-off in the period from mid -- March 15 to when we did our first earnings call, which was, I think April 20-something. We saw a significant drop-off in the rate of activation. And we had to make a call at that point in time what the implications of that were if that kind of relative performance to the comparable period in the prior year, if that recurred. And we did that. And now even assuming that and assuming what we were seeing in aviation [ we it ] acutely in our IoT business in -- because of aviation, basically traffic near-stop, we provide safety services in aviation. And we saw that revenue dry up overnight. And so based on that, and just -- we made the call that even with all that, we thought we would grow our EBITDA over the prior year, even if we assumed kind of -- that what we saw basically in the month period between our first earnings call and when COVID hit, we still thought we would grow. And then things got better from there. As the summer came about, we started seeing not normalcy, but better activations in our legacy business. Certainly saw increased activations in IoT, most notably in personal Communications. In March, we weren't sure what to make of that, how much personal communications were would snap back. And it did quite well. We had a record quarter of net activations, largely on the back of activations in personal communications and IoT. So I would characterize it, Anthony, as we kind of, in April, when we had very limited data, we kind of called the year, what we saw with that limited data. And then as the year progressed, we were able to kind of -- we saw not normalcy, but improvement to the point where we did bring in the full year on at the very end -- very low end of our guidance that we issued in the period prior to COVID.

Great. And maybe just sticking with that concept for a moment. If we look at your guidance for the year, whether it be service revenue or EBITDA, you're certainly projecting year-over-year growth in '21 versus '20, which maybe distances you a bit from your satellite peers who are struggling with some things, particularly notably in the FSS space. But the service revenue growth and the EBITDA growth is not back to the historical trends or the exit run rates that you had once talked about once the new fleet was installed. I guess, what have you assumed in terms of return to normalcy in 2021 as you think about how to forecast and think about the trajectory and the run rate of the business and the pacing throughout the year?

Right. So let's just take IoT, that's the easiest. If you look back in time, IoT has been a solid double-digit grower for, I don't know, 10 years. And we eked out 1% growth in 2020. That was COVID-related, principally aviation. And what we said in our guidance in '21 is that, that growth rate will be up materially. What we know is with the significant headwinds, some $4 million of headwind and that we endured in 2020, is not going to recur. So we're not going to have a headwind. Not sure how much of a tailwind we're going to have, but that's going to depend on how much air traffic comes back. But we're not going to have a year-over-year decrease in aviation. So that's the easiest to understand. We think there's continued strength in personal communication net additions. So we're quite confident you're going to see a nice bounce back in IoT. I would characterize our voice and data business as should come back. We don't think it gets quite back to normal, but you should see improvement there. Our hosted payload business, which grew in 2020 because of a step-up in our Aireon contract, that is not going to see that step up, the step-ups are over there, and we're now in the steady state sort of $47 million worth of hosted payload revenues where we see the steady state. So that's going to be down about $3 million from what we posted in 2020 because there's some true ups associated with the Harris payload. So they're the kind of the main points, Anthony. In our guidance. What we said, what we've gone on to say is, we don't think you should think about us as a 3% grower in terms of service revenue. We think we have better prospects than that. We continue to see an atypical environment in maritime, though COVID continues to cause our installers, they can't get on ships. It's hard to get the mind share of ships' captains to sell -- for our distributors to sell service product into them. We think that changes. We think as maritime gets back to normalcy, it feels like as we exit this year, we should have a more normal environment that we can start selling into and installing. So we think 2020 gets better, but 2020 -- excuse me, 2022 gets better. But we think 2023 and 2024, then you're going to see the effects of the full sales cycle. We're very confident in our Certus product. We hear nothing with good feedback from the channel. It's faster, it's smaller, it's cheaper. So it should sell. It's just -- we had the unfortunate circumstance that COVID hit just as we were putting that product in a big way out for sale. So that's kind of how we think about our kind of growth trajectory.

Great. And I want to stick with maritime maybe for a second. That was a sector and a vertical where obviously the belief that there was and still is a very big growth opportunity, given that there's really one kind of big incumbent there in Inmarsat, who has a lot of share, both from the small vessel and the large vessel piece of the market. Obviously, you just laid out what 1 of the big challenges is for you in terms of just getting access to ships. There's a big risk, I guess, for shipping companies to allow outside people on. You never would want to have a breakout or an infection on a ship because that would take it out of service. What are the -- aside from sort of a COVID normalization and getting access to ships, what are the sort of big sort of sell-in factors in the market that you think give you a big differentiated advantage and allow you to take some share in maritime, given that, that was a pretty decently sized pie that you had talked about at your prior Analyst Day as a real robust opportunity in front of the company?

Right. So I mean, our certification in GMDSS, safety services, we kind of -- we are now a certified GMDSS provider. That unseats a 30-year essentially monopoly that Inmarsat had. They were the only provider that had that certification, so that basically meant that they needed to be on any so-called SOLAS-class vessel, that's a certain tonnage vessel, of which there's about 60,000 ships that had to have Inmarsat on them. Well, those days are over. And so we now approach those same shipowners with a solution that we think our service product is better than the fleet broadband alternative offered by Inmarsat, and you can rip their GMDSS gear off that ship as well. So they don't -- they no longer have a kind of sole rights to providing that safety service. We think GMDSS is a game-changer there. And not so much in the discrete revenue we get from selling the service, but rather, it's sort of -- there's -- the bona fides that come from the fact that we have that certification gives us entrée to ships that we sometimes would not have had before. And we think that's very beneficial to our ability to sell a product we're very confident in, which is Certus. Similarly, while GMDSS is only required on ships over a certain tonnage, the so-called SOLAS ships, there's no reason that smaller ships -- safety is a good thing, and there's no reason they shouldn't want an Iridium GMDSS solution on their ships just for safety services, notwithstanding the fact that it's not required by law.

So I want you to correct me if I think about this incorrectly. But GMDSS, I think it was pretty late in the year. I assume there's not a lot of GMDSS inherent in the guidance. And that one of the other long-term benefits potentially could be the ability to sell in other products and services if you were able to actually use GMDSS to get in the door on vessels that otherwise perhaps were a little outside your remit.

Absolutely. That's our Certus plan presupposes GMDSS certification. So the way we always thought about Certus was, hey, we will sell this product. It's better than the competition, but we will also have the same kind of stripes as Inmarsat has relative to safety services.

And will that -- will they both show up in the service revenue line? Or is GMDSS -- okay.

No, that will all be in service revenue, yes.

Got it. Great. And then I did want to -- you mentioned aviation slowdown. Obviously, that the cockpit product that you have is obviously a safety feature as well in the cockpit of planes. How much of a sort of straight derivative is it off of just increased travel? In other words, are there kind of flat fees that go along with your aviation product? Or is it purely usage, such that if we look at the TSA or the airline data and ultimately the volume of flights pick up, that there's a pretty high, almost 1:1, correlation between the revenue, the usage-based revenue that you get from aviation and what's in the run rate today?

So the aviation shows up in 2 places. It shows up in IoT, and it shows up in commercial voice and data. I'll just describe what shows up in IoT, and that is the safety services for commercial airliners that -- and that is very much usage-based. There's very -- the access fee is low and the usage fee is high. So when -- in that instance, when traffic ground to a halt like it did, that's why you saw our revenue dry up pretty much overnight when traffic halted. And that, we estimated that at about a $4 million impact in 2020. Conversely, in our commercial voice and data business, that is not commercial airliners, that is general aviation. And that model is different. That is characterized by higher access fees. Much like our satellite phone, it's higher access fees, lower usage because it's incidental use, that sort of thing. And so we had an impact in our -- in that business, but it was much more muted than in a -- in IoT because it's not usage-based, a; and b, general aviation held up pretty well. It was really commercial aviation that experienced a big impact. So the answer to your question is, in IoT, you should see the tailwind as, the faster commercial travel gets back to normal, the more we get a tailwind from that.

Great. Yes, that's perfect. And maybe sticking with airplanes for a second. Obviously, Aireon has been a piece of the story for some time. You're an equity owner in Aireon. They're a hosted payload partner. Obviously, their business would have seen some very significant impact, given that they are very levered to the commercial aviation space. And I guess the big question for you on Aireon is the timing of Aireon was actually coming close in terms of when they were to be hitting various milestones. Those have all been pushed out as the commercial aviation slowdown happened. What should we expect from Aireon as it relates to either hitting certain milestones that they have in your agreement that would facilitate step-ups in their agreement, or obviously also be ultimately the back-end sell-down of a portion of your equity stake, which you had previously agreed upon?

Right. So there's not going to be any step up. They've hit their -- the last milestone. So the -- our revenues from Aireon are $39 million, and Harris is about $8 million. Think of that as the steady state, and that's where we are. The kind of the events to occur in our capacity as a vendor to Aireon. Aireon pays us data fee that they owe us annually. They just pay that regular way. And the hosting fee was levied upfront. It was a $200 million hosting fee, which they paid us something like $50 million. I think they owe us -- or $50 million or $60 million. They owe us around $150 million. But there's a minimum that, basically think of what they paid us as prepaid, and they owe us at least $16 million a year, which is our revenue recognition of the hosting. So to the extent that they haven't paid us the lump sum there, there's a contractual minimum that they have to pay us. So we always -- our cash will always equal our cumulative revenue recognition. What they're endeavoring to do is to pay us the $150 million or so of lump sum payment for the hosting fee. And all that is tied to their ability to refinance their existing credit facility. And the events of COVID, all that did was push that event to the right, all right? Because it's all about Aireon's credit statistics, and they're signing more customers and getting their revenues to increase. Well, COVID dealt them a kind of a detour, if you will, but it's a detour that's going to be circumvented in time when air traffic comes back to normal. So Aireon, probably in sometime late 2022 or 2023, will refinance their facility, and we'll get that lump sum payment, right? So that's 1 event. That has very little impact on our financial results because we're booking the hosting revenues at $16 million a year whether they pay us the lump sum or not because of this guaranteed minimum. It is important for -- it will be a nice slug of capital when we get it, and that is important. But it doesn't impact kind of our annual financial results, if you will. And then following that -- go ahead. Do you have a question, Anthony?

No, no. Go ahead because you're probably going to answer what I was going to ask.

Yes. So following that, the sequence is, is that will be the first event. Then as Aireon continues to build their revenue base, they can do a tack loan to whatever facility they put in place to take out their current facility. And the tack loan will be used to buy back certain of our interest for $120 million, leaving us with 22% of the company. And so the things that we think about is Aireon, even with COVID, made great progress with the FAA this year, they signed up some new customers. You saw India lit their airspace. Their non-air navigation data sales are going well. And they, as they exited last year, 2020, were cash flow breakeven. So they -- we think of Aireon as the exact story that we've told everyone it was. It just probably pushed to the right, I don't know, 12 months or something like that, because of the event of COVID.

Now I wanted to just be clear on that because we wrote something about it in our update, and I think we actually misstated what was happening. But to the -- I think you also provided, and your equity partners provided, some bridge financing for Aireon so that they could use that as a capital bridge to help get to the other side of COVID. How does that exposure work in the confines of potentially the broader refinancing that Aireon approaches at some point down the line? Does that get repaid along with the rest of the debt as it's refinanced into a more traditional facility?

Yes. So the bridge facility will be -- to the extent that it's drawn at the time that they do a refinance, will be paid off. I think they're going to leave the bridge facility in place, but it will be undrawn at the time of the refinancing, and just as a backstop. But the answer to your question is yes, that will be repaid in full upon the refinancing.

And so look, the credit markets are a pretty healthy place. I guess, to the extent Aireon is able to do a refinancing sooner than the time frame that you laid out, whether it be '22 or '23, does that pull everything back forward with it? In other words, everything is kind of centrally tied around the recapitalization transaction that Aireon pursues at some point down the line?

Well -- and so it could be incremental, right? It's unclear. While it may be easy for Aireon to get financing of a size sufficient to take out their current facility, which is -- was a range before they were operational and is very expensive and has very kind of limiting covenants, it would be in Aireon's interest to take that facility out sooner rather than later, even if they can't arrange a big enough facility to pay us our hosting and then double back and do a second refinancing to pay us our hosting. So they may very -- they may very well do a refinancing that makes their cost of financing more economical, given the buoyant debt markets, even if they don't do -- repay us coincident with that. But we're going to [ see it some day ]. If you say, when is it going to be? Late '22, maybe '23, depending on Aireon's rate of ramp.

And then how do we think about your equity stake in Aireon long term? There's an agreement to sell it down. It's been prewired as the $120 million, so there's a set valuation for it. Is that something where you see the strategic value to remain an equity investor long-term, even when we fast forward and get back past the refinancing and any other payment or monies that you're owed from Aireon? How do we think about the stake that you have long term?

So once we're bought down to 22%, then it's going to be, we will essentially be clipping coupons then. They will be paying us a dividend. They will be dividending out their free cash flow. We're 22% owner. The way we've sized it is, eventually, that -- those dividends will be bigger than the $39 million we get out of them as -- in our capacity as a vendor to them. And we -- that was our prediction in early 2019. We don't see any reason that's not the case. All that has happened is there's been a 12- to 18-month setback as a result of COVID. But the fundamentals of that business remain unchanged.

Perfect. I think we've kind of covered the Aireon topic well. I know it's one that comes up a lot in discussions. I did want to segue a little bit to the broader competitive environment and competitive intensity. We do get a lot of questions from time to time about just competing LEO products. And I know that you've spoken in the past about differences in technology, differences in the spectrum usage, Ku/Ka versus L, LEO versus MEO. Again, could you maybe just highlight for us the big technological differentiators that Iridium has with L-band LEO? And how you view the competitive forces coming in the LEO market to the extent you think that some of the incremental supply that's coming online, from whether it be Musk or Bezos or the big tech entrepreneurs who are pursuing LEO constellations, how you feel they potentially cross over into some of your markets over time?

Right. So what I would say is, and I said this in my prepared kind of introduction is, our network architecture, so it's low earth orbit, that's LEO, mesh, in the L-band, right? It's that combination that's unique. So Musk is -- he's low earth and Bezos thinks he's going to be low earth. But they're not -- they're in not in the K-band -- or excuse me, not in the L-band, they're in the K-band, and that's a vastly different spectrum. It has different properties. It's very good for high throughput. And it has the same limitation as our friends in the FSS space, the Intelsats and the ViaSats of the world, which is its K-band, susceptible to rain fade, much more typical uses, point-to-point, and that's what Starlink says they want to do. They want to do -- they want to compete with the cable companies and stream video to a house in rural America, right? Well, that's point-to-point. That's different than mobile communications for satellite phones or IoT devices that have very small antennas and are highly mobile. It's a different use case. And the way we've won against our existing competition, Inmarsat, is yes, they're in the L-band, but they're a geostationary satellite. So that's 50x, 5-0-x further from the Earth than we are. And they have issues with latency. I mean, the clearest example is they have a satellite phone that they brought to market a decade ago, and it pretty much hasn't laid a glove on us because it's got issues. I mean, it works, but it's got issues. And we charge at a much higher price point than they do. We've never flinched in terms of our pricing strategy as to them because we just think ours is a better mouse trap. And so that's in our kind of legacy telephony, satellite phone business. But that -- those same attributes transfer very elegantly into IoT. And so why has Garmin centralized on our network? Why is Caterpillar picked our network? These kind of very large names have picked our network because it's both low earth orbit and L-band. And it's those things that represent the kind of the competitive moat that we -- have stood us well against existing competition. And we think, as we see new entrants coming into the space, we think it bodes very well for us. We even think that, just like we are -- we're kind of synergistic with ViaSat in their VSAT terminal -- or VSAT, very small aperture terminals. We go on ships behind them as backup because the K-band is susceptible to rain fade and we're not. The same applies to a Starlink, who's operating in the K-band. So we think we can play nice with all of those new entrants just like we do with existing FSS players.

You're teeing up my industry consolidation question, but I'll save it for a minute or 2 from now. I guess I wanted to keep exploring something that you said because it is a question that comes up, is how important is form factor -- sort of end-user form factor equipment on the back end for the subscriber. One of the concerns people have had is that whether it's Musk or Bezos, they can perhaps afford to do economically unreasonable things to gain market share. And how big of an advantage is it, the fact that you'll have much smaller kind of form factor equipment given L-band and LEO, relative to what the power needs might be if they were trying to get into some of your markets with Ku or Ka in the LEO environment?

Well, right. So just think of in IoT, right? I mean, our modem is matchbook size, a very small antenna, very low power requirements. You can't -- you can -- no matter how much money you can throw at something, laws of physics apply, right? And so Garmin is not interested in something that draws way more power, requires a bigger antenna. They want one that's small, if you look at their inReach device. Caterpillar is not interested in it. It's sort of -- it's -- the K-band is great. It has some great attributes and we see the attributes. And I just look at -- if you take a plane, for example, a commercial plane. There's a whole bunch of people that are going after in-flight connectivity to stream video from the cabin. We're not in the cabin. We have no designs in the cabin. We want the cockpit because that's where we have our niche, which is safety services. And we cede the cabin to all those players who are using the K-band to go against that -- go after that opportunity. So it's like that. It's -- the spectrum dictates kind of how we think about the market.

Perfect. And I guess in terms of spectrum, you're obviously pure-play L-band in LEO. You have significant unused capacity in the network because you've just launched it and you have plenty of ability to scale capacity. Are there future needs for other spectrum bands over time? Or are there other things that you think would make sense in having alongside the spectrum holdings that you have and the L-band that perhaps makes sense in the future?

Right. So we are not spectrum-rich. I mean, we're fine for -- as we look out 10 years, but we are not spectrum-rich, and we are definitely keeping our eyes open for logical extensions of our spectrum footprint. And as you think about -- I get questions about M&A, et cetera. And so what we can say is, look, there's nothing on our shopping list, right? What we've told the market and our partners is we have no interest in vertical integration. So as companies in our partner channel or on the market for sale, we're not interested in that. We're not going to compete with our partners. But an economical acquisition of logical spectrum that will be a logical kind of bolt-on to our existing position is something we'd definitely look at.

Maybe that's a good segue to talk a little bit about M&A. You -- I guess you've taken great pains to have a capital structure -- as the CFO of the company, you've taken great pains to have a capital structure that doesn't have a lot of breakage cost in it. You completed a pretty significant refinancing when the constellation was up and the government contract was re-signed for another 7-year term. You now have a lot of balance sheet flexibility. Your leverage is going to be at your guidance range, end of this year, beginning of next year, whenever you wind up hitting it. How important is it do you think to preserve the flexibility in your capital structure long term? And does your capital structure allow you to have that as an asset as you think about potential either M&A opportunities or spectrum opportunities or any other forms of sort of inorganic growth that might come in the future?

Yes. So we have the most flexible. It's a Term Loan B. We can prepay it as we want. It's about as flexible as you can get, Anthony. So if there's a compelling opportunity, we feel like our capital structure is poised to avail ourselves of it.

And you made a little bit of comment about it in your answer to the last question, but you've also gone out of your way to avoid direct competition with your VAR and VAM partners, and you've largely stayed in more of a wholesale model. Obviously, you mentioned that, that was the game plan for the foreseeable future. But you have had some of your partners and some of your potential customers go through their own form of restructuring during COVID. There has been some consolidation of partners. Are there risks to Iridium to the extent they are vertically integrated with perhaps one of your competitors as you think about potential exposure to particular VARs and VAMs in the channel?

So -- we don't think so. Inmarsat has been vertically integrated. They own Stratos. Stratos sells our product. That hasn't impacted us at all. They bought Ship Equip and others. We're able to sell around that. In fact, we think we gain points with our channel partners because we [ agree ] with them versus Inmarsat. And maybe we get some business as a result of that, in that they'd rather have us on a ship, for example, because they don't have to worry about us unseating them in front of the customer.

Got it. And you just started your first shareholder return program in the company's history. I know that it was well telegraphed and it had been talked about for a long time. I think sized at about $300 million over 2 years. It should allow you to continue to buy back shares and to continue deleveraging. But it's still a relatively modest percentage of your market cap. The market cap has expanded significantly in the last 12 months, the recent history, notwithstanding. What were the capital allocation parameters that you and the Board talked about in terms of setting both the timing and the size of the shareholder buyback? And what would be the priorities around the cash over and above what you would use for the buyback given that, at some point, you run into an inefficiency from a capital structure perspective about deleveraging much further than where you're probably going to be levered by year-end?

Right. So this share repurchase, which is just the first. It's not the final, it's the first. It was done during -- it was announced during a period when we were not at our target leverage, right? And so why did we do that? Well, if you look at our stock over the last 52 weeks, it was $17 for the low and $54 for the high, and we didn't want to be caught flat-footed to the extent that we thought, hey, deleveraging is Plan A. That's what we're about. But if there's a really compelling opportunity because of some market dislocation that we have -- that has nothing to do with us, we wanted to be able to be ready to seize that opportunity. So that is why this first share buyback was announced when it was announced, prior to us getting into the target range. And so we will execute on that as we see fit and always balance the merits of deleveraging into our, and within our target range against the -- how compelling we view the buying opportunity is of our shares. What -- in stating a range of leverage that fits our target, below 2.5x, we -- to your words, we think that is inefficient, and we would intervene before we would allow that to happen with something shareholder-friendly. Is it going to be a buyback? Will it be a special dividend? Will it be a recurring dividend? I don't know one of those, but we like leverage between [ 2.5 x ], and we have the ability to position it there, depending on which shareholder-friendly, equity-friendly kind of vehicle we use at the time.

Understood. And in your prepared comments, when we were starting off, you talked about sort of the characteristics of the business more broadly and how you felt it created more of a peer comparison to telecom infrastructure versus just pure-play satellite. And I guess the one question that comes on the back of that is that those companies tend to operate with much higher leverage. You mentioned the tower sector, they have between 4 and 7 turns of leverage between where you're looking, but they trade for 20 to 25x EV to EBITDA. As you think about how to position the company more to telecom infrastructure, what do you think about the potential for leverage actually being kind of above the range as you've achieved what you need to achieve from a capital structure perspective? And at some point, do you think, if you try to position more to telco infrastructure, that the market could theoretically ask for a higher form of leverage than what you've had today because you've been in a pure deleveraging mode post the launch of the constellation?

Right. I mean, so we think 2.5 to 3.5x leverage is right for us, right? We draw the analogy to telecom infrastructure as to say -- more say, we are not like our satellite peers. Don't use valuation metrics of our satellite peers with their cash flow attributes and their growth attributes because we're different from them. And so telecom infrastructure, we think, is more like us. We're not saying we're a dead ringer from them, and so we don't have to have their exact leverage profile. But it's kind of a balance. It's more -- we point to them as, look at our levered free cash flow statistics, look at our yield, look at our conversion. Don't we look a lot like them? Isn't the fact that we have less leverage kind of a good thing when you look at the risk profile of the enterprise?

Yes, understood. I guess where I wanted to head next was more about the medium and longer-term views of the business. Both you and Matt gave what I would call intermediate-term guidance on the call, albeit anecdotal, on the last earnings call about where you saw the business is going even post-2021. Respective of the fact that you have an Analyst Day planned at some point in the spring, so I'm sure you don't want to preview too much. I thought maybe it will be helpful as we kind of approach the end to leave folks with what you thought were sort of the good medium- and longer-term views of the business that you started providing really on the last call as to what you saw is more normalized growth opportunities and run rates for the business.

Right. So what I would say is the [ plan ] of what an investor needs to understand about this business is we're going to throw off a lot of levered free cash flow over the next 5 years. We introduced that topic in our Analyst Day in 2019, we'll reintroduce it here. That's the plan. Even if you assume no growth, which we're going to -- we think that's highly unlikely. Even if you assume no growth, there is a substantial amount of levered free cash flow that comes off of this business that's going to get returned to shareholders. But we think we're growthy. And we think we're more growthy than you can observe in 2021 for the reasons that we laid out. We kind of are operating with this brand-new, shiny product that we think is going to really resonate. We're operating in an environment that is very atypical and not conducive to us selling. We think that abates. It abates in 2021, towards the back half of the year. We start selling more in '22. We think that hits the numbers in '23 and beyond. We're introducing new products in the mid band. That means pictures over IoT devices, not just text, we think that is very accretive to the ARPUs and makes the product resonate even more so. And so -- and we think our government business, once Certus is built into the government gateway, we think we can sell that in as well. So we think that there's a number of, call them, accelerants on top of the growth rate that we see in '21 that will cause our growth rate to increase in '22 and beyond. And so that's kind of the profile. And that's what we look forward to going in a fair amount of detail at our Analyst Day here in the spring.

Got it. And I guess sort of the final topic as we get close to the ending time is around 5G. And I guess, if you look at the basic premise of Iridium as the company it is today, your value proposition is really to provide superior capacity and coverage in places where traditional networks can't go, be they terrestrial wireless or be they other forms of satellite where they just don't reach efficiently or effectively, whether it be price or throughput. What role does Iridium have in the world of the 5G ecosystem as we see 5G become a more material contributor in the U.S. over the course of the next few years? And how that translates -- how that potentially translates into some of your businesses, like maritime or cruise, or even pieces of some of your IoT business. What is the value proposition for Iridium in the growth of the potential 5G market, meaning the next few years?

Well, I mean -- so 5G is going to get built up where -- built out where the POPs are first. You know that, okay? And so the rollout beyond that is going to take years. We think -- just think we're a logical extension. The more IoT offerings that get built out on the back of 5G has to be good for us because we think we're the satellite provider of choice to the whole IoT phenomena. Like how we'll play in 5G? Our Satelles business is -- in, in-building coverage, they use the GPS clock to get into buildings, and that is more so in the 5G world. Satelles can provide that offering because our paging channel is a lot stronger than the GPS signal and can get into building. So we think with the advent of 5G, that could be a shot in the arm to Satelles, Anthony.

And really, so I guess when I hear you describe it that way, it's really not too much of a stretch to say that the 5G solution in the world that you're in is most likely probably going to be a mesh solution, where there's some form of terrestrial or traditional satellite coverage at various points. But it's very unlikely that, really, competing technologies or competing networks would get built that would be able to offer the same value proposition in 5G relative to what you're able to offer today?

No. Because look, 5G is -- doesn't extend the footprint of -- there's 90% of the globe that isn't covered, right?

Uncovered, right.

And so the day is going to go where the POPs are, and the POPs are in the 10% of the globe that's covered. And they're going to have greater bandwidth there, but it's -- you get away from that and there's no connectivity. And that's where we come in.

Yes. And then final question. It's -- you had mentioned the CapEx holiday that you have, you've always talked about it as a 10-year CapEx holiday. But the existing network actually, until you depleted it when this new one was done, actually lasted much longer than that. When does the plan -- given that you had mentioned being perhaps opportunistic around spectrum, when does the next planning phase start to occur about how you think about the evolution of the constellation, past NEXT and Certus constellation?

So I mean, I think we're thinking about it now. So the 10-year CapEx holiday is, thinking about it and spending on it are 2 different things, right? Last constellation lasted 20 years. When we replaced it, it took 9 years. So it feels like, well, we're not -- we don't see ourselves spending conservatively for 10 years. Thinking about it and thinking about what is our spectrum footprint that it's going to be built around, that's occurring currently. It is what -- we know that we're -- our spectrum, we want to augment our spectrum, and that's going to influence how we think about the next-generation constellation. And so thinking about it like is ongoing. Spending on it, we think, it's years away.

Thank you. And that's probably a good place for us to leave it off. Tom, I want to thank you for joining us. I'm sorry, we're not together in Palm Beach, but hopefully next year, we'll get together in Florida.

Next year, Anthony. Let's just keep our fingers crossed that, that happens. It would be great to be with you in Palm Beach next year.

Well, and if it does, I would imagine your aviation businesses, your maritime businesses will have felt the tailwind from that if we're all moving around globally then.

Yes. Yes.

And I want to thank all the investors for joining us here today. This has been the Iridium Communications presentation with Tom Fitzpatrick, the company's CFO. And Tom, we look forward to hearing more from you and your team at the upcoming Analyst Day in the spring.

Great. Thanks, Anthony.

Thank you very much, everyone. Thanks. Have a good day.

Take care, everybody.