Q4 2023 Iridium Communications Inc Earnings Call

Participants

Kenneth B. Levy; VP of IR; Iridium Communications Inc.

Matthew J. Desch; CEO & Director; Iridium Communications Inc.

Thomas J. Fitzpatrick; CFO, Chief Administrative Officer & Director; Iridium Communications Inc.

Christopher David Quilty; Research Analyst; Quilty Space Inc., Research Division

Hamed Khorsand; Principal & Research Analyst; BWS Financial Inc.

Michael Louie D DiPalma; Analyst; William Blair & Company L.L.C., Research Division

Richard Hamilton Prentiss; Head of Telecommunication Services Equity Research & Research Analyst; Raymond James & Associates, Inc., Research Division

Simon William Flannery; MD; Morgan Stanley, Research Division

Walter Paul Piecyk; Partner & TMT Analyst; LightShed Partners, LLC

Xin Yu; Research Analyst; Deutsche Bank AG, Research Division

Presentation

Operator

Good morning, and welcome to the Iridium Fourth Quarter 2023 Earnings Call. (Operator Instructions) Please note that this event is being recorded. I would now like to turn the conference over to Ken Levy, Vice President of Investor Relations. Please go ahead.

Kenneth B. Levy

Thanks, Megan. Good morning, and welcome to Iridium's Fourth Quarter 2023 Earnings Call. Joining me on this morning's call are CEO, Matt Desch; and our CFO, Tom Fitzpatrick. Today's call will begin with a discussion of our fourth quarter results followed by Q&A. I trust you've had the opportunity to review this morning's earnings release, which is available on the Investor Relations section of Iridium's website.
Before I turn things over to Matt, I'd like to caution all participants that our call may contain forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Forward-looking statements are statements that are not historical facts and include statements about our future expectations, plans and prospects.
Such forward-looking statements are based upon our current beliefs and expectations and are subject to risks, which could cause actual results to differ from forward-looking statements. Such risks are more fully discussed in our filings with the Securities and Exchange Commission. Our remarks today should be considered in light of such risks. Any forward-looking statements represent our views only as of today, and while we may elect to update forward-looking statements at some point in the future, we specifically disclaim any obligation to do so even if our views or expectations change.
During the call, we'll also be referring to certain non-GAAP financial measures, including operational EBITDA, pro forma free cash flow, free cash flow yield and free cash flow conversion. These non-GAAP financial measures are not prepared in accordance with generally accepted accounting principles. Please refer to today's earnings release and the Investor Relations section of our website for further explanation of these non-GAAP financial measures as well as a reconciliation to the most directly comparable GAAP measures. With that, let me turn things over to Matt.

Matthew J. Desch

Thanks, Ken. Good morning, everyone. As you saw from our release this morning, we finished out 2023 very well. Service revenue and operational EBITDA both were strong, and we generated more than $300 million in pro forma free cash flow for the full year. That last point, growing cash flow is the real storyline for Iridium. It supports our growth and investment and allows us to return capital to shareholders through dividends and share buybacks.
Since turning cash flow positive in late 2019 after completing our second generation constellation, Iridium has generated approximately $1 billion in free cash flow. And as we discussed with you at our Investor Day in September, we expect to have the capacity for approximately $3 billion in shareholder returns between 2023 and 2030. In fact, Iridium returned more than $310 million to shareholders last year. As part of this program, we initiated a quarterly dividend, which paid $65 million to shareholders in 2023. And as we announced in our 8-K today, our Board plans to increase the quarterly dividend to $0.14 per share starting in June. This will result in a full year increase of approximately 6% to Iridium's dividend in 2024.
Our Board has also authorized $1 billion in share repurchases since February 2021. Of this amount, about 2/3 has been utilized and 1/3 a bit more than $330 million remained available at year-end 2023. Iridium's ability to generate and grow free cash flow has been fueled by robust commercial service revenue. While equipment and engineering revenues can vary year-to-year, which they've been doing in the last few years around the pandemic, commercial service revenue growth has proven quite resilient. If you look at our commercial service revenue since the completion of our second-generation network in 2019, it has grown at an 8% CAGR. Subscriber growth has also been robust, growing at a compounded annual rate of about 15% over the same 4-year period.
2023 marked our network's 25th year of operational service which means we've had a lot of experience successfully navigating changing business environments and the evolving needs for satellite connectivity. The key reasons for our success have been Iridium's extremely reliable and truly global network our globally allocated and coordinated L-band spectrum position and probably most important, our laser-like focus on personal communications and doing what others can't do as well as we can. Our network has become the cornerstone of communications for safety services and mission-critical applications. This reliability is what truly differentiates Iridium from other service providers.
In recent months, investors have asked us about the impact that Starlink is making on the satellite industry and if we are being affected, whether from the existing broadband service or from the direct-to-cellphone service that we're -- that they're in the early stages of developing. Starlink has been very disruptive to the satellite industry, but we're fortunate that Iridium has carved our own unique path, which has allowed us to continue our growth even with their entry, and we are confident that we will continue going forward.
With regard to high-speed broadband, we've never really competed with the Ku and Ka-band satellite operators on this front, and that remains the case with the entry of Starlink. We positioned Iridium Certus as a reliable companion to all of the VSAT offerings in Maritime, and we're often a companion to Starlink installations as well. We also continue to be differentiated by our L-band spectrum's ability to support safety services in both maritime and aviation. That remains an important focus for us. It's a long-term niche where we'll continue to see growth at other broadband LEO services, including Starlink can address.
On the direct-to-device front, we're also following a unique path and expect to coexist with Starlink and others who are each delivering a very different service propositions to the market. Starlink is planning to offer a more regional solution using terrestrial cellular frequencies, which will need approval on a market-by-market basis. Iridium on the other hand, is planning to offer a standards-based solution that will leverage our existing global MSS L-band frequencies. There's a lot of varying opinions about the potential size of the D2D market, but I think most would agree, it could eventually be a sizable opportunity. Though I think everyone is beginning to understand it will take a long time to really mature.
We're now just in the very early stages of development, and we expect Iridium to be a core long-term player with a very strong offering. Being standards based, we don't have to be first to market, we just have to be one of the best and we've had success doing that in the mobile satellite industry since our founding. In the case of directed device, we believe our plan, which we're calling Project Stardust, will be one of the most practical solutions because it's an incremental development on our very flexible and robust network. It won't require launching new satellites every few years and will be global from day 1. It will use our existing spectrum position rather than relying on regulatory bodies for approval market by market. We're also focused on a very specific set of use cases for complementing terrestrial narrowband IoT and smartphone messaging, both of which will attract new partners like the mobile network operators to sell our services.
You probably know I'm a bit skeptical about the business case for D2D broadband from space, not because I don't think there will be some user demand and interest. But because our experience at Iridium shows that a 2G or 3G like service offering, that only work outside or inconsistently inside buildings will not translate well into the mass market experience that customers now expect from a 5G, 6G world. We believe that the foundational capabilities and capacity of these D2D broadband networks will not attract the customer use or price premiums necessary to sustain the high ongoing capital costs for continued satellite replacement and network operations to deliver them, certainly not for many years and without massive continued investment, if at all. We believe that Iridium's project Stardust strikes the right balance for a high-quality user experience and will be the kind of service that smartphone makers and MNOs will be proud to offer to their roaming customers at an affordable price.
Iridium's strong position in IoT and use in small form factors and mobile device also positions us well. We have an extensive partner ecosystem and are familiar with operating through this wholesale model and most importantly, believe that Project Stardust will expand and extend our existing leadership in IoT to lots of new use cases and applications. So 2024 will be a year of investment for Iridium. We are increasing our spending on R&D a bit and moving forward in a few product areas that position us to capitalize on longer-term growth opportunities. We want to maintain our lead in traditional areas of growth like IoT, mid-band data and U.S. government services, and we also plan to undertake initiatives that augment future business growth. Together, we believe these efforts will allow us to reach our long-term service revenue guidance of about $1 billion in 2030.
We highlighted our financial outlook and growth drivers during last fall's Investor Day including how we expect incremental growth from voice, IoT and broadband, which will complement our expanding relationship with the U.S. government. Initiatives like many of our partners mid-band service expansions this year, will drive new IoT and data applications in emerging industries like uncrewed aerial and maritime vehicles and support higher-speed consumer products that can send pictures and other important information for users on the move. Our partners are really excited about our new IoT transceiver that we'll finalize this year. It's natively IP and cloud-based and will enable even more compact designs for their applications. We're also excited to expand our safety service offerings in Maritime and Aviation in 2024.
This is the year Aviation will complete certification of the Iridium's end-to-end safety offering using Iridium Certus technology. Iridium Certus will become the most cost-effective solution for cockpit voice and data solutions, delivering increased value through safety and analytics to airlines, business jet and general aviation operators for the next generation of airspace. On top of these items, we're investing in incremental R&D with Project Stardust, because we believe it will position us well for standards-based narrowband IoT growth, by which I mean expanding the opportunities we are already addressing with our proprietary offering today and also helped to expand our business to MNOs and others delivering service to smartphones, laptops, watches and tablets.
Together, these investments give us great confidence in Iridium's cash flow and growth plans through 2030, all themes we highlighted to you last September. Shifting gears a bit, I want to share an update on Iridium's constellation health. When we launched the Iridium next constellation between 2017 and 2019, we were required to make an initial estimate of the network's life. As a starting point, we pegged it to the manufacturers assessed design life. However, we were and are hopeful to eventually get the same life from the Iridium next constellation that our first network achieved.
It ended up lasting over 20 years by the time it was retired. But we had no practical information at the time of launch to make our own assessment. Based on the manufacturer's assessment, we've been using a 12.5-year useful life for accounting purposes since 2019. We reevaluate that assessment periodically and our most recent review in the fourth quarter, prompting us to update the constellation's estimated life, which we now believe will perform well through at least 2035. That's great news and continues to support our expectations of a CapEx holiday through the rest of this decade.
The updated assessment also reflects the successful launch of 5 additional spares in 2023 and the overall performance of the constellation since its initial launch. These incremental data points led us to formally change our forecast for satellite life from 12.5 years to 17.5 years. While this change doesn't impact cash flow, it does have some accounting implications that Tom will discuss in detail, including impacts of how we recognize hosted payload service revenue and depreciation expense and the resulting effect on our operational EBITDA guidance. Overall, there are a lot of moving parts in our 2024 guidance. While we are still forecasting strong subscriber and service revenue growth, with our increased investments, our equipment shipments normalizing and some challenges to comps from 2023 on the surface 2024 doesn't look like a normal year of operational EBITDA growth.
If you look deeper, though, our growth profile and overall outlook really hasn't changed much. Tom will go through that analysis, along with the peak into 2025, where most of those onetime effects normalize. To me, the bottom line for Iridium continues to be free cash flow growth, which will continue in 2024 to support our ongoing investor-friendly activities. We expect share repurchases and dividends to continue and we're as excited as we've ever been about business opportunities, growth and cash flow we will generate. So I look forward to another exciting year for Iridium.
So with that, let me turn it over to Tom. Tom?

Thomas J. Fitzpatrick

Thanks, Matt, and good morning, everyone. With my remarks today.I'd like to recap Iridium's full year results for 2023 and provide some perspective on our fourth quarter performance and our change in accounting estimate to reflect the extended book life of our satellite constellation. This morning, we also released our outlook for 2024, and I'll provide some color here especially in context of Iridium's longer-term growth expectations.
As Matt noted, we increased the estimated useful life of our satellite constellation by an additional 5 years during the fourth quarter which extends the life of our satellites to 17.5 years for accounting purposes. The updated useful life affirms our confidence in the health of our constellation and the duration of our CapEx holiday through 2030. However, the accounting change, which has been reflected in our fourth quarter financial statements has a couple of implications. First, extending the time over which we depreciate the book value of our satellite reduces Iridium's depreciation expense by over $100 million per year. This will have the impact of both increasing our net income and earnings per share. Second, the accounting update reduces the annual revenue we recognize from our hosted payload contracts by spreading those fixed revenues over the satellite longer expected useful life.
This extension of estimated useful life had the impact of reducing hosted payload revenue by $2.3 million in the fourth quarter of 2023 as compared to prior quarters. Going forward, the change will cause annual service revenue to be approximately $9.1 million lower each year through 2029 than had we not updated the estimated useful life of our constellation. Hosted payload revenues will now be recognized through 2035 versus the previous schedule, which concluded in 2030. As you can see, this creates a comparability issue when considering our growth projection between 2023 and 2024 that we introduced today.
To assist investors with these changes and facilitate an apples-to-apples discussion, I would identify the effects of this accounting treatment on Iridium's expected growth when I get to our guidance. As for our full year results, Iridium performed well in 2023. New contract wins, strong equipment sales and favorable pricing all supported top line growth. We delivered strong commercial service revenue growth and had another good year of subscriber additions. Pro forma free cash flow was $303 million in 2023. Iridium shareholders were the direct beneficiaries of this performance. Dividends and share repurchases totaled $311.8 million during the year. In the fourth quarter, operational EBITDA rose 7% from the prior year's quarter to $114 million and total revenue grew to $195 million.
Strength across all commercial service lines and continued growth in engineering and support offset reduced equipment sales. Within the commercial business, we reported service revenue of $121.5 million in the fourth quarter, which was up 10% from a year ago. Revenue from commercial voice and data rose 12% from the prior year period and continued to reflect the benefits of the price increase we enacted earlier in the year. This discrete price action supported ARPU growth of 10% during the quarter and has been easily digested by our channel partners, evidenced by continued subscriber growth in our voice business.
In commercial IoT, personal satellite communications continue to fuel double-digit revenue and subscriber growth. Subscribers were up 18% from the year ago period and we ended 2023 with more than 900,000 personal satellite communications devices on our network. We believe that more retail customers are just now becoming aware of satellite connectivity and that these low-cost consumer devices will fuel demand for satellite messaging and SOS services for years to come. As awareness of these consumer-friendly products grows, so too do new applications using Iridium's global network. As Matt noted, our IoT partners continue to invest in new retail-focused products with new functionality supporting higher ARPUs, we believe this market segment will serve as a catalyst for IoT revenue growth moving forward.
In broadband, we reported revenue of $114.6 million in the fourth quarter, up 5% from the year ago period. Iridium service continues to be adopted as a companion to VSAT services in Maritime. We are, however, also seeing some competition from low-cost VSAT alternatives impacting certain vessels where Iridium service serves as a primary satellite connection. We expect lower billable usage on some vessels to pressure ARPU and in turn, revenue growth rates in our broadband business for a few more quarters. Once the lower usage rates normalize into our ARPU base, we expect revenue growth to accelerate on the back of subscriber gains. The vast majority of Iridium's broadband customers are already using Iridium Certus as a companion service. So this usage impact is limited and should normalize relatively quickly.
In all, commercial subscribers grew 15% year-over-year with IoT now representing 80% of the total at year-end, up from 78% in the year-ago period. Revenue from hosted payloads and other data service revenue, rose to $15.2 million in the fourth quarter, principally due to higher precision location service revenues, of which $2 million was nonrecurring and resulted from an updated estimate on a customer contract. This increase was largely offset by the $2.3 million impact on revenue recognition resulting from the change in useful life of our satellites. Government service revenue was stable in the fourth quarter at $26.5 million, reflecting the terms of our EMSS contract with the U.S. government.
Subscriber equipment, which reached record sales in 2022 and for much of 2023, decreased materially in the fourth quarter to $15.7 million. While full year 2023 finished as the second highest equipment sales in company history, we expect demand for satellite handsets and other Iridium hardware to decrease materially in 2024 and normalized to be more in line with periods prior to 2022 before we and our competitors began to experience the effects of supply chain disruptions due to the pandemic. Engineering and support revenue grew 74% from the prior year period to $31.1 million in the fourth quarter as Iridium ramped up work with the Space Development Agency. While this work has lower margins than our commercial business lines, the FDA contract remains highly strategic and aligns Iridium closely with the U.S. government's long-term space priorities.
Moving on to our 2024 outlook. We anticipate service revenue growth of between 4% and 6% in 2024 and are forecasting operational EBITDA of between $460 million and $470 million. In order to appropriately evaluate our 2024 guidance, I want to highlight several factors that are relevant. As I noted previously, we increased the useful life estimate for our satellite constellation as of October 1, 2023. Since we recognize the fixed portion of our hosted payload revenues over the life of the constellation, this had the effect of reducing hosted payload revenue by $2.3 million in 2023. It will also reduce 2024 revenues by $9.1 million, resulting in $6.8 million less of service revenue and EBITDA in 2024 compared to 2023 than if we had not updated our estimate. To reiterate, there is no operation or cash flow impact from the change in our satellite's estimated useful life, only the length of time over which we recognize the fixed revenue.
For illustrative process, at the midpoint of our 2024 guidance, EBITDA would be $465 million. This would represent about $2 million in growth from the $463.1 million in EBITDA we posted for 2023. If we had used comparable useful life assumptions in both periods, projected EBITDA growth in 2024 at the midpoint would have been $8.7 million which we believe to be more representative of Iridium's projected growth in 2024. This rate of growth is still lower than we have been experiencing in recent years, owing to a few headwinds, we will experience in 2024 that we do not expect to recur in 2025.
First, as we have previously noted, we expect equipment revenue and margins to revert to pre-pandemic levels which means we're forecasting a material decline in 2024 from 2023. We think the sales level we're forecasting this year will be the baseline level we'll see going forward. Second, we recognized a $3.5 million gain on a contract settlement in 2023, which will not recur in 2024. And finally, as Matt discussed, we are ramping up R&D spending in 2024 in support of our NB IoT initiative, Project Stardust. This will add approximately $5 million to R&D compared to 2023 and we expect to maintain R&D spending on this initiative in 2025, though, should not represent a headwind in 2025. Taken together, we estimate that these discrete items represent a headwind of about $20 million to our 2024 EBITDA forecast.
When considering Iridium's prospects for EBITDA in 2025, if we take the $8.7 million in apples-to-apples projected growth for 2024 over 2023 and then remove the estimated $20 million in nonrecurring headwinds were experienced in 2024 and we believe our prospects are good for generating close to $500 million in EBITDA in 2025. Other operational assumptions supporting our 2024 outlook, which I've not yet touched upon are as follows: we expect lower growth in our commercial voice and data business than in 2023 as 2023 benefited from a price increase. We expect our IoT business will remain strong. Demand for personal satellite communications as well as commercial applications remain the drivers here. This gives us confidence that 2024 will be another year of double-digit revenue and subscriber growth. We should also begin to see wider adoption of Iridium mid-band and benefit from the introduction of our new IoT transceiver late in the year.
I would be remiss not to touch on engineering and support, which continues to benefit from contract work with the Space Development Agency. Government engineering and support revenue will grow again in 2024 as we continue to build out the ground infrastructure and operation centers and start to man their operations for Space Force's new constellation. Though this contract revenue can fluctuate from quarter-to-quarter. Our 2024 outlook also incorporates a number of expense assumptions that may be helpful when updating your models. First, we expect SG&A to remain relatively stable this year, even as we continue to add head count to support the FDA contract.
As Matt noted, we also anticipate higher research and development costs this year, up as much as 30% as we begin our work on standards-based solutions and continue to invest in product development initiatives. As it relates to depreciation and amortization, we anticipate a decrease of over $100 million in 2024 due to the longer estimated useful life of our satellites. You will note that 2023 was impacted by this change for 1 quarter, equivalent to about $25 million in benefit compared to 2022.
But you'll also recall that in the second quarter of 2023, we completed successful on-orbit testing of 5 of our 6 ground spare satellites as we believe the construction and progress associated with the remaining ground spare satellite would no longer be used, we wrote off the $37.5 million remaining in construction in progress for that satellite by recording accelerated depreciation expense which more than offset the accounting change in the fourth quarter. As I mentioned earlier, the prospective reduction in depreciation expense is entirely to an accounting update and will have a positive impact on Iridium's GAAP earnings, pushing both net income and earnings per share, firmly into positive territory.
Lastly, Iridium now expects minimal cash taxes of less than $10 million per year from 2024 through 2026. This is new guidance for '25 and '26 and incorporates additional R&D and other attributes we expect to realize. Pro forma free cash flow is expected to rise to about $309 million in 2024 at the midpoint of our EBITDA guidance, reflecting our continued growth and recurring revenue model. We believe that pro forma free cash flow is a good measure of our business strength and investors should continue to track closely.
Moving on to our balance sheet. As of December 31, 2023, Iridium had a cash and cash equivalent of approximately $72 million. Our cash balance is ample to fund our operations and continues to anticipate the payment of quarterly dividends and expectations of share repurchases. In the fourth quarter of 2023, Iridium retired approximately 1.3 million shares of common stock at an average price of $38.71. For the full year 2023, iridium purchased 4.8 million shares at an average price of $51.40 for a total of $244.6 million. This left us with an outstanding balance of $334 million at year-end under our Board-approved authorization through December 31, '25. We will continue to execute on our buyback program, balancing our objective for deleveraging with the desire to maximize return on investment.
In 2023, Iridium initiated a quarterly dividend and paid a total of $64.8 million to shareholders for the full year. As Matt noted, Iridium's Board expects to increase the dividend beginning in the second quarter of 2024. The approximate 6% annual increase to Iridium's dividend in 2024 reflects our confidence in the company's business opportunities and prospects for continued strong free cash flow generation.
Turning to CapEx. CapEx holiday period between 2020 and 2030 is an average of -- between $50 million and $60 million per year. This remains our best estimate, but spending will not be uniform over the holiday period. Our CapEx from 2020 through 2023 averaged just under $50 million, exclusive of the launch, which is not maintenance CapEx, but rather a part of construction costs of Iridium NEXT. We expect CapEx to average closer to $60 million for the balance of the holiday period between 2024 and 2030. We expect CapEx to be over $60 million in the next couple of years as we invest in new product development initiatives like project Stardust and network efficiency programs. We expect CapEx to trend below $60 million in the latter part of the decade as we decreased maintenance spending in anticipation of our third-generation constellations.
We closed 2023 with net leverage of 3.1x of EBITDA. This was down from 3.2x a year earlier and includes the impact of our buyback activity and dividend. We think Iridium naturally delevers over time and expect to exit 2030 below 2x leverage, even after giving effect to our dividend program and all share buybacks authorized by our Board. I also want to highlight Iridium's term loan, which we will now mature in 2030. You'll recall that we extended the term by about 4 years and in September and lowered the interest rate on the facility. Iridium also enjoys a favorable position on its interest rate cap, which hedges about 2/3 of the term loan. We believe these instruments allow us to weather the current interest rate environment and would look to extend our hedge as market opportunities present themselves.
Turning to our pro forma free cash flow. If we use the midpoint of our 2024 EBITDA guidance and back off $76 million in net interest, pro forma for our current debt structure, approximately $69 million in CapEx for this year, $5 million in cash taxes and $6 million in working capital, inclusive of the appropriate hosted payload adjustment, we're projecting pro forma free cash flow of approximately $309 million for 2024. These metrics would represent a conversion rate of EBITDA to free cash flow of 66% in 2024 and a yield of approximately 7.3%. A more detailed description of each element of these calculations, along with the reconciliation to GAAP measures, is available in a supplemental presentation under Events on our Investor Relations website.
In closing, Iridium continues to execute well and deliver strong free cash flow growth. We feel good about our competitive position. And as Matt noted, we'll make some investments in 2024 to augment our opportunity set and drive new revenue growth through 2030. With that, I'd like to turn the call over to the operator for the Q&A.

Question and Answer Session

Operator

(Operator Instructions) Our first question comes from Rick Prentiss with Raymond James.

Richard Hamilton Prentiss

Well, that's a lot of moving pieces and stuff to digest, but I appreciate all the details. First question I've got is Obviously, the extension of the useful life was a $2.3 million noncash hit within 4Q. I thought I also heard there was a $2 million nonrecurring benefit from a contract in fourth quarter. Was that a cash payment? Or was that also a noncash item?

Thomas J. Fitzpatrick

It was the recognition of deferred revenue just due to a change in circumstances, just think of it as cash.

Richard Hamilton Prentiss

All right. the dividend, appreciate the increase the dividend expected and planned by the Board for 2Q. Is that one we should think of kind of the annual review of the dividend policy will be on like a 2Q basis? And why do it in 2Q versus like off of year-end results?

Thomas J. Fitzpatrick

We were focused on the full year increase. So we like the 6% increase. And so the easiest way to accomplish that was just to hit the second quarter with $0.01.

Richard Hamilton Prentiss

Got you. Got you. Okay. Makes sense. And then on the Project Stardust, I guess David (inaudible) coming in here. The -- what is the kind of expense and CapEx efforts that you're doing that will kind of lead to that investment. What efforts do you try? Are you trying to pull off with that? And is there a need to select partners? Or is the standard-based mean you just move that way. So just trying to understand what the Stardust project entails OpEx and CapEx over the next couple of years?

Matthew J. Desch

Yes. I mean it's an incremental investment on our existing network. It requires updating satellite software, some new ground infrastructure in our gateway, gateways. And the standard space activity is really in the chipsets, and we've already are pursuing that activity as well in discussions with chip manufacturers to include our band and into the standard, which is a real big lift given that it's pretty close to others as well. So all that activity, I think we've said at the announcement of Stardust a couple of weeks ago, we mentioned it was tens of millions of dollars kind of level of effort over a couple of years. That's in incremental R&D and probably more so in incremental capital, which has all been sort of described today in our guidance.
Tom went through kind of most of that in terms of what we expect that to mean between now and 2030. But in the big scheme of things, not a real big increase in what we're doing on our -- because we have such a flexible network today. And we also think it's well covered by sort of the market potential growth that it leads us, I would say, particularly in narrowband IoT and all the use cases that it could expand in that sort of general area. And as well, obviously, it puts us in a real good positioned to provide D2D kind of services as well. And there will be other -- there are -- we're deep into partnership negotiations with a number.
We've gotten a lot of things in support because we're one of the -- only real LEOs doing this right now. And so a lot of people really want to encourage and support us in that regard. So we're talking to everything from MNOs to chipset manufacturers, the smartphone manufacturers and obviously, our partner base is quite interested in as well as we evolve to providing not just proprietary IoT technologies, but standard-based IoT technologies long term. So a couple of years, but it's well underway.

Richard Hamilton Prentiss

Okay. And I appreciate the thought of $500 million in EBITDA in '25 given puts and takes in what might be a more normal one. But the $1 billion service revenue target by 2030, how much does that assume and when would it assume kind of this Project Stardust spending moving into being actually producing revenues?

Matthew J. Desch

We're not providing specific piece of that. It will not really affect that amount much until later in this decade. We're really not expecting it to ramp up until latter 2 or 3 years. So it won't be that significant a portion of it, but it is part of it.

Operator

Our next question comes from Simon Flannery with Morgan Stanley.

Simon William Flannery

Matt, just to come back to the extension of the useful life of the constellation, is this purely an accounting exercise? Or have you reassessed from an operational standpoint? I think a couple of times you referenced the CapEx holiday through 2030. But is there any thought of potentially extending that CapEx holiday as a result of this? Or is it just an accounting exercise?

Matthew J. Desch

We're not extending it today. I mean our guidance is really the guidance through 2030, and that's what I still expect. Obviously, if our satellite constellation continues to perform well, and we don't need to build a new network. It could be extended again beyond that. As I said, we got over 20 years out of our last constellation. So 17.5%, I hope is still at the lower end of what we'll ultimately end up getting. But no, right now, we're not providing guidance to extend the CapEx holiday beyond that. It certainly gives confidence that we'll achieve that. And -- but it's definitely possible that we could extend it again.

Simon William Flannery

Okay. And I guess you have to make that decision, I don't know, 2 years out, 3 years out. Is that right?

Matthew J. Desch

Well, I mean, we really assess our Constellation life every year. This year, was the first year as we ran through it and said it's been 5 years, if you will, since we started launching it, our satellites are performing much better than expected. We added 5 additional satellites in space to add even more in orbit spares than we were expecting. And really, we couldn't deny that the simulations we were running weren't pushing us out to at least 17.5 years, even beyond that. So it sort of forced an update in the fourth quarter and all the implications that it had I mean, we'll continue to do that every year, probably around the fourth quarter is when we typically assess that, and it could very well happen in the coming years as the network continues to perform this well, we'll have to update that again. But for now, I think that would probably be the expectations for at least the next year or 2.

Simon William Flannery

Great. And then on the commercial broadband, the comments about some exposure from switching from a primary service to companion service. Can you sort of size how much revenues are coming from us as a primary service that sort of usage exposure? And is that really your main exposure within that commercial broadband to some of these new LEOs or is there other exposure in other segments?

Matthew J. Desch

That's the most exposure we really see at this point. And we said we're not -- we're switching -- I mean, we were always a -- we're were primarily being used on -- I have to be careful to use that word too many times as a companion service. The majority of our installations out there are as a companion to VSAT service, VSAT services, including Starlink, don't work when it storms. They have to be turned off and ports for regulatory purposes, often or other reasons, and certainly, coverage.
And we were the majority of our installations have been used as of companion. To the extent we are used as a primary, we're seeing very low cost, Starlink coming in, being offered, particularly for, say, crew welfare and that sort of thing is sort of taking over that usage that those primary usage, and they might, in many cases, switch over to a companion type application. But it's in the single digits.

Thomas J. Fitzpatrick

Yes. No. So I can size it or we see it, Simon. We see ARPU, if you look at full year ARPU in 2024, versus '23, we see 5% to 7% down in '24 from '23. But since it's finite and kind of a small number of vessels, think it runs its course in '24 and then we get back to growth sometime in 2025.

Simon William Flannery

Growth in ARPU.

Thomas J. Fitzpatrick

Not necessarily growth in ARPU, but growth rates accelerated in '25 because you're just going to be -- you're going to have the subscriber gains and not the ARPU dilution, which we'll see in '24.

Operator

Our next question comes from Walter Piecyk with LightShed.

Walter Paul Piecyk

Can you hear me now?

Thomas J. Fitzpatrick

There you go.

Walter Paul Piecyk

The $9 million from the accounting impact, if I just use simple math on the service revenue, $585 million or so for '23, that's a 150 basis point impact to growth, right? So your midpoint 5%. So even if you added that back, the guide is effectively 6.5% now. So...

Thomas J. Fitzpatrick

That's right. You got that right.

Walter Paul Piecyk

Yes, so that feels a bit slower on services that I think what a lot of us thought you could -- would grow in this business. So is that just this ARPU thing that you referred to in the prepared comments? Are there other things you can comment there on that element.

Thomas J. Fitzpatrick

Certainly, the ARPU in broadband is a headwind that's influencing the service revenue. But if you think about our long-term guide that we issued in 2021 for average high single digits, '23 through '25 you're sitting on a 9.75% in '23. And so getting to high single digits seems pretty easy, and we'll hit that long-term guide. So the answer to your question is, yes, there is a bit of a headwind from the broadband, that's kind of limited and it's going to abate, as I said. But we feel very quite.

Walter Paul Piecyk

Can you just -- I'm sorry if I missed that. Can you repeat why it's going to abate in '24 and somehow on?

Thomas J. Fitzpatrick

So there's a finite number of vessels where we're primary, and we think that's good they will move -- they'll convert to a companion quickly because it's in their interest to because it's a cheaper solution. And so it's a, it's a small number of vessels is in their interest. And we think that the ARPU runs its course through '24 and settles in '25. And then what you'll be left within '25 is just subscriber gains and flat ARPU.

Walter Paul Piecyk

So the subscriber equipment revenue slowdown that occurred understandable, obviously, given the backlog, yada yada, all the things that we talked about in the last couple of quarters. seem to be a pretty good predictor of this lower growth rate. So I guess I'm curious on your subscriber equipment outlook for 2024. I guess you said more normalized levels. It was $16 million in the fourth quarter. So is this the run rate, $15 million to $20 million that we're expecting for equipment revenue? And then doesn't that -- is that an indicator of future service revenue growth?

Thomas J. Fitzpatrick

Yes, we don't think so. So we think growth in -- so the most pronounced decrease in equipment revenues was in IoT. The channel clearly ordered to have safety stock, and we met that challenge and enjoy those revenues. But if you look at the time series data, look at the 5 years prior to 2022, equipment revenues averaged under $90 million. They spiked $135 million in 2023 as this channel's kind of stuffing occurred. We think our prospects in IoT, notwithstanding the material decline in equipment revenues, in '24 from '23 and '23 from '22 are as strong as they've ever been. So our IoT prospects are really good. And handset similarly, I mean, we were very clear in as we were enjoying record handset sales in '23.
We were clear that -- excuse me, in '22, we were clear that our competitor was stocked out, and we were getting the benefit of that. Well, they now have phones, and we're back to a normal kind of competitive dynamic and normal levels of sales. So we think this equipment issue is going to run its course here this year. It's going to be a headwind in '24 that will abate in '25, and we'll get back to consistent with more longer-term trends but we do not think that it is leading indicator for softening in our service revenue.

Walter Paul Piecyk

It sounds like it's going to take a couple of quarters at least, in order for some of the stuff to kick back in, in terms of a reacceleration right? It sounds like 1 or 2 quarters, we're looking at '25 now in terms of a reacceleration of the growth.

Matthew J. Desch

I don't -- I think you're taking 2 things that happened and saying they're correlated just because our comments -- well, I'm not sure I'm following what you're doing. There's not an acceleration, really. The growth rate, as we said, is continuing to happen in '24,in the same way it was in '23. So I don't know what this acceleration you're talking about really is.

Thomas J. Fitzpatrick

And to be clear, Walt, we were very clear in the Investor Day in September that we were going to model and for purposes of arriving at our $3 billion in capacity for shareholder returns through 2030, we weren't hanging our hat on elevated equipment revenues. We were very I thought clear that we saw equipment revenues coming down, and that doesn't do a thing to our $3 billion in projected capacity in shareholder returns.

Walter Paul Piecyk

Understood. One last question on margins. Is that -- was that -- I haven't done the exact math on this, but we do some math, I'm trying to understand incremental gross margin in the fourth quarter on services specifically, and it looked like it dipped. Is that because of that $2 million onetimer was basically a profit less or no gross profit revenue?

Thomas J. Fitzpatrick

No, it was -- that was normal service revenue side I don't think you'd make -- I wouldn't make much of that. I mean our operating leverage model remains very much intact.

Operator

Our next question comes from Edison Yu with Deutsche Bank.

Xin Yu

Just one follow-up on the D2D side. I know you're not expecting material revenue contribution until later on. But do we have a sense on the operational time line as in where we could get more partners announced when the first maybe test phones will be out there. Just trying to get a sense on the operational ramp rather than just the revenue contribution.

Matthew J. Desch

Well, we discussed during the announcement of Stardust that we would -- the earliest that we'd be doing some testing would be '25 and the earliest we would see implementation would be in '26. And I said that we're still early, not really changing any views of that right now, but that's kind of haven't changed anything in the last couple of weeks since that time. So '26 would be the absolute earliest you would see that. In terms of announcements and everything, I mean, we'll announce when it makes sense to announce it, we don't usually just announce things for marketing purposes, all the interest that we have will be more around what our partners want to say and that sort of thing to us, it's more about building the capability and getting into operation.

Operator

Our next question comes from Chris Quilty with Quilty Space.

Christopher David Quilty

Tom, I hate to harp on the equipment issue, but just looking at the inventories, it looks like they were up about $20 million sequentially and more than 2x year-over-year. Was that a reflection of how quickly it turned down? And what have you done in terms of new order rate for product? And do you expect perhaps have to do any discounting on a go-forward basis on that? Or do you expect all the inventory to just clear and the natural force.

Thomas J. Fitzpatrick

We expected to clear in the natural course, we don't expect any discounting. That's a conscious decision on our part to prevent a stock out. If you look at our ads in our voice and data business in 2022, those investments paid off handsomely for us, and we expect the inventory to move down frankly, for -- in '24 and then down again in '25.

Christopher David Quilty

Great. Shifting back to the Aviation Certus -- certification for safety services, I think, Matt, you said you expect that certification to come through this year? And I guess is that front half of the year, back half of the year? And do you expect it to have impact on the growth in that product line? And a secondary question to that is, I don't know, have you ever done OEM installations or are these all aftermarket.

Matthew J. Desch

So the safety program, which largely developed is entering into a certification phase right now where it starts flying on airlines as a whole trial program -- that sort of is being scheduled right now with airlines agreeing to sort of fly it and demonstrate it through most of this year. We expect to achieve the certification later in the year, and our partners manage really the installations.
We don't do that. It's our -- the ones who have channels into the airlines, in particular, for these kinds of services like controller pilot data link and that sort of thing, who have channels to install these things on airliners aftermarket as they are being built, et cetera. And that's a model that we've been part of for the last 15 to 20 years, really, in our narrowband product. This will really give us a whole new capability that they're excited about with higher-speed services, and we expect to see good transition over to that new Certus technology in the coming years.

Christopher David Quilty

So does it make sense to either working with the OEM or a partner shooting for OEM installation?

Matthew J. Desch

I'm not sure what you're referring to. I mean...

Christopher David Quilty

Factory installation rather than after market.

Matthew J. Desch

Yes. I mean that's under -- that's expected. The people who make Boeing and Airbus really want Certus in the cockpit. And I'm expecting you'll hear more about line-fit applications of those as they adopt it. They both want Iridium because it's more global, it's less expensive. The antenna is smaller. The cost is better overall. It is a preferred solution for the cockpit. And manufacturers are extremely supportive of this whole effort we're underway. This is not something we're doing and pushing as much as there's a lot of pull in the market for it.

Christopher David Quilty

Got you. And one final question on D2D. You had some peer group companies, competitors form a new mobile satellite service Association looks like it's primarily ViaSat -- Inmarsat driven, though, I guess, they did pull in Yahsat and omnispace on that also. What was your rationale for not participating in that effort?

Matthew J. Desch

Lack of knowledge of it. We've -- I heard about it when you heard about it. Mark actually reached out to me this week, and we have a call tomorrow to talk about it. he asked if we would want to join, but I don't know much about even what it's necessarily seeking to accomplish yet, and I look forward to hearing more about it. I sort of assumed when I saw it as you did, that it was the geo operators, many of them who don't have spectrum, I mean, who don't have satellite sort of capabilities or more spectrum oriented kind of band together to harmonize and market together to those people who may be able to use that spectrum since most of the people who announced that really don't have the money to build, say, a stand-alone network for that way.
So I'm interested in hearing more about it. But obviously, our strategy is pretty self-contained right now. We have our own system today. We have spectrum. We have the ability to implement as Stardust our plans, but I'll certainly look forward to hearing more about it and seeing if our interest align in some ways.

Christopher David Quilty

Got you. And one more final question, which is back to IoT. Have you seen any movement? You've talked about the UAV market in the past several quarters. Should we expect to see some announcements around this year? And how would that contribution look relative to the ARPUs or number of units compared to the existing base?

Matthew J. Desch

There's still an awful lot of activity in that space. More VARs and service providers who are developing solutions. We're testing things out in everything from industrial applications as well as delivery and other things who want a lightweight, lower cost, low power kind of command and control solution. Just had a discussion yesterday, there's a number of HAPS providers, which is interesting.
It's kind of a new growth space as well who have larger vehicles who again want to use Iridium for our global command and control ability for their solutions as well. But I don't see a giant ramp-up in necessarily, at least coming through here quite yet. I mean I certainly see a lot of interest in market activities, market development activities, solution development. But I don't have visibility to kind of when that will ramp. So I don't have a I don't think it's really going to be this year necessarily. I think it's still to come.

Operator

Our next question comes from Louie DiPalma with William Blair.

Michael Louie D DiPalma

Related to the previous question and discussion on spectrum, you have discussed how you are interested in acquiring more spectrum if the opportunity arises. And as it relates to Project Stardust, are you interested in forming regional spectrum partnerships to add capacity to your network at all similar to what AST Space Mobile and Starlink are doing with the cellular carriers.

Matthew J. Desch

Well, first of all, what they are doing is sort of utilizing terrestrial spectrum, expensive terrestrial spectrum in places where it's not approved and hasn't been allocated, et cetera. And specifically, that's something that hasn't been approved by regulators and for which there will be a lot of, I think, lawsuits and other things to try to block it from people who don't have the opportunity using that spectrum. And I think there's a lot to play out over the coming years. As a result of that, those are going to be highly regional kind of offerings at best.
And really, our satellite system isn't equipped or planned to be able to kind of reuse terrestrial spectrum without -- well even -- it wouldn't even be software. It would require hardware development and new satellites launched, and we have no interest in doing that. As I said, I'm a little skeptical about the business case for that anyway for those people who have to do that and maintain the satellite systems that they have to do it.
So I'm more interested in using MSS spectrum. It's globally approved. It's allocated. It's -- can be employed without regulatory requirements. It keeps our value proposition that's always been one of the reasons we've been successful is that we -- that any customer who uses a product on ours doesn't have to worry about any partner about where their products are deployed in the world. It's an easier sell to the end consumer that they can use it anywhere as opposed to only in certain locations for certain things. So on that basis, I'm not really not very interested in sort of the terrestrial reuse model at all.
I think the L and S-band MSS model is really kind of the longer-term view as to whether we need more spectrum for the future for our next generation, we have plans right now that reuse our existing spectrum in a next-generation system that we think will add lots of capacity as we launch a next-generation system out in the 2030s and 2040s that we'll do our -- so we don't necessarily need new spectrum if it became available to us and everybody who has it today believes it's extremely valuable somehow.
So we'll see how that really plays out. they got it for free and they're hoping to make money on sort of it. And I don't see that, that's really right now that there are anyone that's sort of offering to partner or sell or lease spectrum at kind of any kind of what would make any kind of sense. So right now, we're not necessarily planning to do that, but we'll keep our eyes open for something where there might be a more reasonable approach available to us.

Michael Louie D DiPalma

That makes sense. And one other, the iPhone 14, I think, debuted in September 2022 with satellite, and it's been in the market for 1.5 years now, and the iPhone 15 came out a year later with satellite, has there been any impact from the iPhone Globalstar partnership on your consumer IoT business or your commercial voice business that you've been able to discern.

Matthew J. Desch

Absolutely none, to be honest with you. It seems to have sort of been absorbed. I think -- obviously, I think it had an effect that why a proprietary D2D solution that we had with Qualcomm sort of didn't take off more quickly because even the smartphone manufacturers really weren't seeing market share movement as a result of the capability. So yes, it's also said that it looks like -- I think I also deduce that it means it's going to take a little longer for sort of an effect to these solutions to really maybe impact the market at all right now because we're not really seeing that much of an impact right now on consumer products of any sort, probably saw bullet had challenges. They put a product in the market and didn't succeed. So it's really not making much move in terms of anything that we can discern.

Operator

Our final question today comes from Hamed Khorsand with the BWS Financial.

Hamed Khorsand

Just first off, could you talk about why you're spending more on R&D. I think that was a topic you had about (inaudible) it seems like you're doing it again this year. Is it competitive? Are you fearing that the market is moving away from you? Because it feels like your capital structure is changing as well your operating structure?

Matthew J. Desch

Well, I think as I mentioned in our results, Project Stardust is an incremental investment that we weren't expecting to make last year, assuming that our proprietary solution with Qualcomm was adopted in the market, but people decided they didn't want to buy Qualcomm's new chips and so -- and prefer a standardized solution, and that isn't available. We would like to continue developing all these new products and services that we mentioned. I mentioned we're quite excited about -- we didn't talk a whole lot about it, but building a Maritime Certus terminal that has all the regulatory capabilities into it even further solidify our position as a companion product.
It will have GMDSS, LRIT, SSAS, Citadel services all in a single terminal, which would make it by far the most popular companion product in the marketplace for any kind of VSAT terminal. That's not something I want to slow down in any way so that we can just absorb with the same amount of R&D dollars, a new investment in Project Stardust nor do we want to slow down the development on our new IoT transceiver that we've been sort of building for a while and for which our partners are quite excited about. So I wouldn't say it's anything to do with competition only a new opportunity to move into standards-based narrowband IoT that we decided to do kind of mid-year, late last year in 2023, and that's sort of adding, if you will, some shorter-term pressure on R&D and CapEx.

Operator

This concludes our question-and-answer session. I would like to turn the conference back over to management for any closing remarks.

Matthew J. Desch

Well, thanks. We do have a busy conference schedule, I know in the first quarter, so I'm sure we'll be seeing some of you in those places, and we look forward to visiting with many of you over the next coming weeks. So we'll see you also for the first quarter in a couple of weeks. Thanks. Take care.

Operator

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.