Ken Levy - VP, IR Matt Desch - CEO Tom Fitzpatrick - CFO & Chief Administrative Officer.

Andrew Degasperi - Macquarie Capital Greg Burns - Sidoti & Company Ric Prentiss - Raymond James James Breen - William Blair Chris Quilty - Quilty Analytics.

Welcome to the Iridium Third Quarter 2016 Earnings Conference Call. [Operator Instructions]. I would now like to introduce your host for today's conference, Vice President of Investor Relations, Ken Levy. Please begin..

Thanks, David. Good morning and welcome to Iridium's third quarter 2016 earnings call. Joining me on today's call are our CEO, Matt Desch; and our CFO, Tom Fitzpatrick. Today's call will begin with a discussion of our third quarter results followed by Q&A.

I trust you've had an 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 this morning may contain forward-looking statements with the meaning of the Private Securities Litigation Reform Act of 1995.

Forward-looking statements are statements that are not historical fact but include statements about 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 disclosed in our filings with the SEC. 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 our forward-looking statements at some point in the future, we specifically disclaim any obligation to do so even if our views or our expectations change. During the call, we'll also be referring to certain non-GAAP financial measures.

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 a reconciliation of these non-GAAP financial measures to the most directly comparable GAAP measures. With that, let me turn things over to Matt..

Thank you, Ken and good morning, everyone. As I'm sure you read in our press release, Iridium enjoyed another quarter of strong subscriber growth and we raised the bottom-end of our full-year guidance to reflect the underlying strength of our core business. I want to begin today with the topic on everyone's mind.

The first launch of our Iridium NEXT satellites. We've been on hold since SpaceX's launch pad incident at Cape Canaveral eight weeks ago and have been following the investigation closely to determine when SpaceX will be able to return to flight.

From our participation on the accident investigation team, we believe that SpaceX is conducting a very thorough process, tracking down all possible leads, evaluating every piece of data and exhausting every option. I will leave it to SpaceX to provide the official updates on the investigation.

Gwynne Shotwell has said a number of times recently that SpaceX still expected to have a couple launches in 2016 which presumably includes ours. Until they formally clear the rocket for launch and provide us with a specific date, I don't want to speculate on timing. However, I remain hopeful that they'll return to launching this year.

Also, I don't know if Iridium NEXT will be SpaceX's first launch once they return to flight or whether they might schedule a launch from Florida ahead of us either way, we're comfortable with SpaceX's investigation and the progress they're making and I assure you that we won't proceed to launch if we aren't confident in SpaceX and their investigation outcome.

Many have asked, how this most recent delay might impact our constellation, our customers and our launch program. So let me address each of these. First, our constellation is operating well, even as it enters its 19th year in space. We have not had any additional satellite failures since losing two earlier this year.

We continue to have one in-orbit spare which we could use if a satellite failure was in that same plane. I've seen some speculation wondering whether battery life which can limit satellite life, may have contributed to the June quarter's satellite losses.

And I can report that this is not the case and we're seeing no signs of a battery failure or even a signature of even a battery failure from any of our satellites.

Second, the previously reported loss of two satellites in the second quarter has resulted in only a very short outage twice a day at certain latitudes which could create a small impact on some users, though our customer satellite availability statistics remain high.

The resiliency of our network is a function of Iridium's cross-linked satellite architecture and the low-earth, polar orbit of our constellation. Remember each satellite circles the globe every 100 minutes. In the event that a user is directly under a satellite hole, another satellite comes into view within minutes to make the connection.

The self-healing and constantly moving nature of our network is our secret sauce and is what differentiates us from competing technologies. Importantly, our first two launches target these two holes and will bring our network back up to full satellite availability quickly.

So while we're clearly eager to get the next generation of satellites to space, health of the current constellation is not an immediate concern. What's most important is that the Iridium NEXT mission will enable a new era of growth for our Company and unlock at least two new significant revenue streams.

First, Iridium NEXT carries the hosted payload from our partners, Aireon and Harris and we want to get those payloads into orbit as soon as possible so that our customers can start to commercialize their respective businesses. Secondly, we're also excited about our Iridium Certus broadband service which is enabled by Iridium NEXT.

This service will change the landscape of satellite L-Band broadband. Not only will Iridium Certus expand Iridium's addressable market, but it will also provide a long awaited superior choice for customers who seek reliable, faster connectivity than the competition at a fair price.

As it relates to our launch mission, I want to provide an update on insurance and satellite production. We completed the placement of launch and in-orbit insurance for each of the seven planned SpaceX launches about the time of the September 1 pad explosion.

I'm pleased to report that the total cost of all policies placed for the Iridium NEXT mission was in line with our expectations. SpaceX's recent fueling incident did not change or otherwise affect our policies. So, we're good to go on insurance for Iridium NEXT.

I should also note that and this was always our plan, our satellites will not be mounted to the Falcon 9 rocket during the static fire, fueling and testing process. So that's not a risk compared to the September event. Regarding satellite production, manufacturing continues to go well.

As of today, 20 satellites have been completed and we're on track to complete 35 by year-end. This gives us confidence that satellite availability won't be an issue for the mission's launch cadence in 2017. We have 10 satellites ready and waiting in the SpaceX processing facility at Vandenberg.

Transport of these satellites from Orbital's facility in Arizona, as well as the satellite to dispenser mating process, all went very smoothly. The fairing which covers the satellite during launch is also at Vandenberg, so we're just waiting for SpaceX's rocket to arrive in a few weeks, as well as the final clearance to launch.

Until we have a successful first launch, it's hard to say exactly when the Iridium NEXT constellation will be completed. In light of the most recent delay, it looks like full satellite deployment will move into 2018. We'll know more after we get the first launch off and get the first 10 satellites into our network.

In the meantime, we know that 2018 won't be a full year of operation anymore for a new network. So Tom will cover the adjustments we've made to our long term financial guidance to reflect this. Turning to our financial performance, our business continued to perform very well, fueling a 9% rise in operational EBITDA this quarter.

We remain optimistic about the new opportunities that continue to come our way through new customer acquisition, contract wins and new products available through our partner network. Iridium added 15,000 net new subscribers this quarter, with close to 90% coming from our commercial business.

Continued growth in M2M, along with our strong relationship with the U.S. government, provide a clear path for business expansion moving forward. 2016 has been a particularly good year for Iridium in adding some sizable and strategic M2M partners.

Our progress is owed to changes we implemented last year in a disciplined manner in which our team has targeted and executed on new opportunities. Historically, we've added about a dozen new M2M partners each year. Though only a handful of these contributed significant new subscriber growth which also typically takes a few years to realize.

This year's been different, while we're still on pace to add about a dozen new M2M partners, almost half of these are of the size and capability to add multiple thousands of new subscribers to our network each year. Our team has already converted on most of these target accounts, many of which are household names and industry leaders.

This change in strategy and our strong execution bodes well for continued growth in our IoT business in 2017 and 2018. We're also working more closely with our M2M partners to help them get their products on to the Iridium network faster, so that subscriber growth can be realized in matters of months rather than years.

That's why we unveiled Iridium Edge, the new low power tracking device during the quarter. This rugged service-ready device extends the range of cellular-based solutions for fleet management and other remote monitoring applications.

Off-the-shelf availability allows Iridium Edge to be used as quick-to-integrate, low-cost solution and as one of the ways we can help our new partners activate devices more quickly. Iridium Edge is forecast to be available to partners by the end of the first quarter 2017.

Speaking to new partners, I can announce today that, Komatsu, the world's second-largest heavy equipment OEM manufacturer, is one of our newest M2M relationships. This partner which I alluded to last quarter, extends Iridium's reach with global OEMs and continues our ongoing success in the heavy equipment sector.

Today, Iridium has long term contracts with seven of the 20 largest construction OEMs which account for more than 40% of the industry's annual sales. This is pretty remarkable, given that we only entered the space a few years ago. As you can hear, I'm very excited about the new solutions and partners we're adding in the IoT space.

Iridium is uniquely positioned for long term success in satellite M2M and we anticipate that our next generation of satellites will accentuate our position by attracting more developers through our platform of services.

Specifically, next year's launch of Iridium Certus will introduce the new IP-based technology and higher data speeds for the same small antenna form factors which will be very attractive for many of our current and future IoT partners.

In the most recent quarter, our joint venture, Aireon, made a number of important announcements that elevate its standing in global aircraft tracking and surveillance.

To start, it announced a new service called GlobalBeacon with FlightAware that provides tracking information for airlines to help them meet the ICAO requirements for Global Aeronautical Distress Safety System or GADSS that resulted from the MH370 tragedy.

GADSS adoption is recommended by 2021 and recently to Qatar Airways announced its plans to adopt the GlobalBeacon solution for its fleet of 190 aircraft. Among ANSPs, Aireon added Myanmar to its list of countries evaluating satellite ADS-B services. It also recently signed its ninth contract for long term data services.

This new contract is with Seychelles Civil Aviation Authority which manages the vital airspace block between Asia and Africa. Increasingly clear that Aireon is becoming the partner of choice to ANSPs looking for a global solution for aircraft surveillance and tracking.

With the imminent launch Iridium NEXT satellites, Aireon will be able to conduct real-time in-orbit demonstrations of this groundbreaking technology to customers.

So, we believe the Iridium's next-generation constellation will redefine the technological capabilities available in low-earth polar orbit and spur Iridium's entry into many new user segments. We have tremendous confidence in the quality of our satellite design and engineering and look forward to launching the SpaceX in the very near future.

We'll let you know the specific date as soon as we can. So with that, I'll turn it over to Tom for a more detailed financial review.

Tom?.

Thanks, Matt and good morning, everyone. Let me start by summarizing our key financial metrics for the third quarter and then I'll review our balance sheet and liquidity position. Iridium generated third quarter total revenue of $112.8 million which was up 6% from last year's comparable quarter.

Revenue growth was fueled primarily by strength in government services on the back of the step-up in the EMSS contract with DISA. Operational EBITDA was $69.7 million, an increase of 9% from the prior-year period. Our operational EBITDA margin remained strong this quarter, reaching 62%.

This reflects seasonally strong revenue in the quarter and the benefit from one-time item. Specifically, the expiration of a time-based contract and a non-recurring foreign tax true-up. Accordingly, we expect full year OEBITDA margins to be below 60%.

From an operating viewpoint, we reported commercial service revenue of $65.3 million in the third quarter which was 3% higher than the prior-year period. As Matt noted, we had another strong quarter of subscriber additions, driven by our commercial M2M business.

In our commercial voice and data business, ARPU improved to approximately $44, while billable subscribers declined by 1% year-over-year. The rise in ARPU is attributable to a benefit I just mentioned associated with the expiration of a time-based contract.

Absent this one-time item, commercial voice and data ARPU would have been steady from the year-ago period. We continue to see declines in traditional telephony airtime usage, though the negative effect of this on ARPU is being muted by growth in higher ARPU products like Iridium OpenPort and push-to-talk.

In our commercial M2M business, we picked up 14,000 net subscribers in the third quarter while ARPU had held steady at approximately $15. Fueling this gain was another strong quarter of activations of the DeLorme inReach personal locator product.

This is particularly encouraging, given that the inReach device has not yet harnessed the power of Garmin's wider international distribution network. We also saw a steady stream of activations from Caterpillar which is now widely selling Iridium-based telemetry solutions.

Similar to last quarter, the headwind associated with oil and gas related deactivations has diminished. Our government service business reported revenue of $22 million this quarter, representing a 22% increase from the prior-year period. Government subscribers again grow at a rapid clip, rising 19% year-over-year. Total U.S.

government customers reached a record 82,000 this quarter, driven by 30% growth in M2M from the prior-year quarter. Continued collaboration between Iridium and the U.S. government on mission-critical application should support subscriber growth through the balance of 2016 and in 2017.

In the third quarter, subscriber equipment revenue was $19.9 million. This was below the prior year's level and reflected a decline in the sale of high margin handsets during the quarter. We continue to expect margins in the low-to-mid 40% range in the coming quarters, though this can be impacted by product mix.

Moving to our balance sheet and liquidity position, as of the end of third quarter we had drawn approximately $1.7 billion from the COFACE Facility and had a cash and marketable securities balance of $426.4 million. I should point out that Iridium's liquidity position does not change materially as a result of the SpaceX launch delay.

The majority of our outstanding vendor contracts for Iridium NEXT are performance-based. And milestone payments to vendors like Thales Alenia and SpaceX are principally tied to the launch and in-orbit deployment of Iridium NEXT satellites. So as the timeline for Iridium launches has pushed out, so too has the timeline for payments for these vendors.

As a follow-up to my comments last quarter, we have continued to work to obtain additional flexibility around our capital plan to guard against the delay in Aireon's hosting payment in 2017. There are multiple constituents involved in this process; each of the nine lenders, the French Treasury and Thales. Our goal is to finalize the deal by year-end.

Iridium does not expect these modifications to include a capital raise. This morning, we updated our full-year 2016 and long-range guidance.

Based upon the strength we've continued to enjoy in M2M, including robust subscriber growth, we've raised the bottom end of this year's growth outlook and now expect total service revenue growth between 5% and 6%, up from 4% to 6%; and expect operational EBITDA in a range of $250 million to $255 million for the full year, up from the prior range of $245 million to $255 million.

The recent launch delay caused by SpaceX incident does impact our outlook for capital expenditures in 2016. We expect approximately $100 million of milestone payments to vendors, SpaceX and Thales Alenia which were previously assumed to occur in 2016, will now occur in 2017.

As a result, we expect capital expenditures between $400 million and $450 million in 2016, down from the previous forecast of $550 million. We're taking this opportunity to also just our long-range outlook to better reflect our anticipated timeline to deploy the Iridium NEXT constellation.

When we introduced long term guidance in 2010, 2018 was assumed to be the first full year of operation of the Iridium NEXT constellation and the first year of materially lower capital expenditures given the completion of Iridium NEXT. Given the most recent SpaceX delay, that will not be the case.

Accordingly, we believe that fiscal 2019 now offers a more useful measure of the earnings power of Iridium than does fiscal 2018.

With constellation deployment expected to be complete in 2018, fiscal 2019 should be the first full year of operation for Iridium NEXT and encompass a full year of contractual revenue from hosted payloads, data fees and revenue trends for Iridium Certus, following its commercial launch.

Similarly, 2019 capital expenditures will be reflective of the post NEXT construction period run rate.

Our updated long-range outlook is now, total service revenue between $440 million and $465 million for the full-year 2019; operational EBITDA margin of 60% in 2019; negligible cash taxes through at least 2020; peak net leverage of 6.0 times to 6.5 times OEBITDA in 2017; and net leverage below 4 times OEBITDA in 2019.

With that, I'll turn things back to the operator for the Q&A portion of this morning's call..

[Operator Instructions]. Our first question comes from Andrew Degasperi with Macquarie Capital. Your line is now open. .

First on your long term guidance, does this assume a 12-month delay for SpaceX and then should we see it as a conservative point of view on that?.

Absolutely not. the problem, I think, as Tom mentioned, is that we don't know whether the constellation will go active in February, March, April, June, whatever and the amount of revenues that come from hosting fees, et cetera, changes relatively dramatically over a course of a couple of months' difference.

So, we're still assuming that we will be completing the network sometime probably in the first half of 2018, but it's just hard to constantly adjust long term guidance around that kind of moving target.

That's why, I think, just locking the 2019 number for everyone so that you can sort of see the transformation of the business between now and then makes a lot more sense..

And I would say too Andrew, that there's $47 million worth of hosting and data fees from -- on Iridium NEXT. And so, to the extent that's pro-rated in 2018, every time we would cite the 2018 guidance, we have to give you the pro forma number to get back to $47 million.

So it just seems a lot cleaner to say 2019 is the clean year when we have the fully operational constellation, you've got the CapEx that's reflective of the run rate and that's how you should think about the enterprise.

Like I said, this long term guidance is introduced in 2010 and we were painting a picture of think about the enterprise post the construction period and it's really that logic that makes us think that 2019 is the appropriate year, not a assumption of a 12-month delay in the SpaceX..

And secondly, on the satellite build, I assume you've built 20 already, you're obviously building out faster than your launches. I was just wondering, are you going to start seeing free cash flow benefits once that's complete in the second half of 2017? Thanks..

Well, we're not building satellites faster than we're expecting, we're right on track to what our plans were. So our cash flow really is tracking as it relates to there.

I think Tom mentioned that we do have payments to Thales as it relates to milestones and that sort of thing and that's kind of caused our cash flow to move, our cash payment out to move from 2016 into 2017 and I expect some more movement from 2017 into 2018.

No more cash, no less cash, just sort of a profile has changed or moved out a couple of months..

Yes, that's exactly right. It's a fixed price contract but it gets paid when milestones are completed. So, the launch delay is a milestone, so the payments move in lock-step with the launch delay..

And last question from me, the Komatsu deal, is this exclusive or essentially the multiple providers for the telematics service? Thanks..

No, it is not exclusive and we're going to try to make it effectively exclusive by performing extremely well.

Komatsu has already been doing a lot of development around getting ready to deploy our service and we think we could command a majority of new installations going forward based upon just our network performance and everything related to what we offer them.

But, typically, a lot of these construction OEM companies aren't necessarily exclusive, but what we'd like to do is get as much of their business going forward and maybe leave some of their existing business probably won't churn immediately, it will take some time..

And our next question comes from Greg Burns with Sidoti & Company. Your line is now open. .

So I just wanted to kind of, I guess, dig in just a little bit more color on the change in pushing out that the full year guidance to 2019.

So, for 2018, I mean, should we be thinking of it more now as -- instead of a kind of demarcation in 2017 where we go to this significantly higher cash flow in 2018, it's now kind of a step function into the 2019 now.

I mean, there should be a build in free cash flow depending on the kind of -- if the full satellite network goes up in first quarter, second quarter of 2018, we should still see an increase in cash flow, it's just not the full benefit, so it's going to be more the step change in 2019 rather than just a immediate benefit in 2018.

Is that the right way to think about it?.

That's right, Greg. The easiest way to think about it is the $47 million that I just referred to on the previous question. It's going to come in pro rata. You won't have a full-year of operations, so it comes in on the number of satellites that are operational. And it's just tough for us to call that right at this moment.

And like I said, every time we would quote 2018 we'd have to give you the pro forma number for a full year's benefit because that's what's relevant here. And similarly, we're going to have CapEx in 2018.

So, it just seemed to us -- like I said, this is the picture that we first painted in 2010 of what the enterprise looked like kind of at the end of the construction period and it's just now clear that 2019 is going to be the end of construction period, not 2018..

Okay.

And when we think about Certus, how many satellites do you have to get up before that becomes operational, you start realizing revenue from that service?.

It's a good question. We always expected it would be about half the constellation, because the terminals that are going to be introduced by our first partners near the end of this year, early part of next, as they start testing those out, are dual mode.

So they operate on an existing constellation and the new satellites, so they kind of go back and forth between current speeds and the higher speeds.

So that's a pretty good user experience, but it certainly doesn't seem to make sense to -- I think, for most people to sell them until they're getting at least half of the time they're getting the higher service. Remember, new satellites are going into operation daily or weekly through 2017.

So, we had originally expected around middle of the year would possibly be there, but it really kind of depends, it could be a little later than that, could be earlier. So, I might put some satellites up there, we're just going to have to declare the date sort of sometime in 2017..

Okay. And then lastly with the Aireon and GlobalBeacon, I'm just trying to understand how that is different from the service they have been talking about providing. Anyway, I thought they were providing space-based ADS-B to airlines.

How is GlobalBeacon different?.

So what GlobalBeacon is a very specific service that uses that FlightAware which currently has great relationships with all the airlines, would allow them through their channels to deliver information about location and everything around an airplane to an airline directly outside of really the ANSP.

So, the basic service that Aireon provides is real-time aircraft surveillance from the air traffic control entities. It's a regulated service. It's really what has to be provided in that airspace. It's a governmental function, typically or a governmental private function and it's sort of internal and not available to anybody else.

What GlobalBeacon does is, extend that really to the commercial environment to allow easy access for airlines to see their airplanes themselves as opposed to calling up an air traffic controller and saying, do you got any idea where my airplane is..

Okay.

I mean, what would be like aiding solution to that, like who you're competing against? And do you have a pipeline of airlines kind of looking at that service right now?.

So this is a service that FlightAware which is a partner of Aireon, will be providing to its customers, that's the relationship. I think the significance again is this ICAO recommendation to all airlines that they implement the solution to track their aircraft by 2021, they call it GADSS or Global Aviation Distress Safety System.

It came out of the MH370 and a lot -- and there have been many solutions that have been suggested as being available to offer it, whether they'd be satellite-based solutions that Iridium would provide, that would offer the ability to be able to see every airplane certainly every 15 minutes.

But if there was a distress, it would be very quickly, they could see the real-time position. Because ADS-B will be on every aircraft around the world, it will be delivered into Aireon's database. Delivering that kind of GADSS service really doesn't cost anything additional in terms of equipage, et cetera.

So it will allow the airlines to meet this GADSS recommendation effectively. And I'm sure there are other people that will offer satellite based solutions. Those cost money.

We will offer those as well, if they would like to use an Iridium satellite tracking system or a cockpit operational system, performing ADS-C or FAN-type solution, that could be helpful, but not every airplane in the world will necessarily be deployed with satellite connectivity.

So this really allows the airlines to meet the mandate without adding additional equipment..

And our next question comes from Ric Prentiss with Raymond James. Your line is now open..

I'm going to ask a question, I think you talked about CapEx and how you're expecting in 2016, it will drop. I think you said, previously you were thinking $550 million, now going to be $400 million to $450 million, that as launches might be delayed, the cash payments you make to SpaceX and Thales.

I just want to confirm is that what I heard correctly?.

It is. Our prior guide was $550 million and now we're saying $400 million to $450 million. That's illustrative of the fact that the contracts are milestone based. And because the launch is being delayed, the CapEx is delayed accordingly..

And then obviously with giving of the 2019 full year guidance, you mentioned some will be into 2018, so given the guidance that you've given, what should we assume based on your current assumptions than CapEx in 2018 would be versus than in 2019?.

I'll answer the 2019 question. We've said CapEx average is $35 million in the post construction period, so that would be 2019 expectation..

Okay, and then anything for 2018 as far as [indiscernible]?.

We'll guide on 2018 at the appropriate time, Ric, once we know exactly how things are going shake out..

Sure. People are watching the cash burn, just trying to understand that as things slip out, the CapEx can slip out. So I'm just trying to keep the cash in mind. Also, obviously, you guys are doing a lot of work with the financial flexibility. I think you mentioned that you were hopeful then to get the stuff done by year-end.

What's the process with nine lenders, the French, Thales.

When should we be looking for as far as milestones and understanding have you gotten the financial flexibility in place by year-end, how should we be monitoring that?.

So the goal is -- our collective goal of the constituents I enumerated in my prepared remarks is to get it done by year-end. It is a complex -- there's a lot of constituents, all are involved and we're working it, we're making progress and I think we will have a favorable outcome..

Okay.

So really just from the external side, I was kind of keep listening, do you guys have got a lot of sausage making going on though?.

I think that's a fair characterization..

All right. And then final question from me, [indiscernible] for all the extra detail on the two satellites that went down last quarter and how you're working on it and why you're not too concerned about right now. We have done some questions also about the de-orbiting timeline.

If it was -- battery life is not an issue, is there any issues as far as de-orbiting the satellites or could there be slipping of when you could deorbit them, just there has been some concern about what that might be with the delay in the new constellation?.

We've never given really guidance out on exactly when we would de-orbit the satellites. We've always felt we had a fair amount of flexibility on that, as long as we ensure that when we do de-orbit them, we de-orbit them quickly. That has always been our commitment to the U.S.

government and frankly it's good housekeeping for space to utilize all the fuel and the satellites and to try to get them out of orbit when we start the process within months, certainly within a year for the majority of the satellites and we still plan to do that.

When we start launching the new Iridium NEXT satellites, they replace Iridium Block 1 satellites, satellite by satellite and go into operation. We moved the Block 1 satellite out of the mission orbit into a lower storage orbit temporarily. It can stay there for a while. It obviously could be moved back for any reason, as sort of a hot spare.

So we will clearly have a lot of extra satellites here pretty quickly as the network starts transitioning over. We've already been talking internally about the process and think that, relatively quickly, for some of the satellites, maybe as early as 2017, but certainly into 2018, we would start the de-boosting and de-orbiting process.

We would do it for blocks of satellites at a time, ideally. And we would certainly expect to kind of complete the process, probably by the end of 2019, I would expect, but could be faster. But we're trying to play that by ear based upon our belief of whether we need the satellites as backups.

Obviously, we don't think we should, but just in case it's nice to have them up there and we'll remove them expeditiously when we're done. But I don't think this delay process really just moves that whole thinking out about three months or four months, I guess..

And our next question comes from James Breen with William Blair. Your line is now open. .

Just around the guidance as well, what are you assuming in terms of growth in the government contract for that 2019 year, as soon as your contract comes up at the end of 2018? And then, Tom, just as I think about the CapEx, it seems pretty systematic based on when launches happened.

So, is it sort of fair to assume that the CapEx from the back half of 2016 or 2017, that total amount just shifts depending on when the first launch starts and when the last launch finishes? Thanks..

Yes. So I think that's right, Jim. It moves in lockstep with the delays in launch..

Yes. And then as far as the contract goes, we haven't really engaged -- well we have started, if you will, discussions with the government because they like to start the process too, but it's still sometime between now and the completion of the current contract. So, we have some current expectations about how that evolves going forward.

The biggest difference is really that there are more services to provide the government with Iridium NEXT constellation, specifically Iridium Certus and of course STL is one that they -- there's a lot of enthusiasm about the satellite time and location.

So, we haven't given any guidance nor it is probably appropriate for us to give expectations in anticipation of that negotiation process. We certainly think we have a lot more to offer the government with the new network, but I think I'll leave it at that for now..

Except to say, Matt, that we're quite sure the contract will be renewed. The goes without saying..

Yes. I would believe that every discussion we've ever had is more about enthusiasm for the future, as opposed to anything else. So it's more a matter of -- just say, we'll -- keeping all our options open with the customer..

So, Jim, just one point of clarification. The contract ends, I think, it's October 20, 2018. But the government has the right to extend the contract for six months under the same terms and that has been their prior practice. So, I'd be thinking about a kind of April 2019, would be when you would see the impact of that new contract..

And our next question comes from Chris Quilty with Quilty Analytics. Your line is now open. .

Tom, can you swing back on that voice and data ARPU issue? I kind of missed the explanation there around the impact of contract timing..

Right. So what it was with a large prepaid contract that was usable, if you will, to a point in time and expired with a lot of minutes unused, so we took into revenue. It was about a million bucks of a benefit. So that boosted the voice and data ARPU. If you looked at it year-over-year, it was up about $1.

You'll know Chris, as you've been following us for a long time that for a period of time we saw kind of continuing reductions in that voice and data ARPU, in the 2013, 2014 period. What we've seen of late is a stabilizing in that ARPU.

And there was, like, three or two factors kind of competing, right? In the traditional telephony, we have reductions in airtime usage which is dilutive to ARPU, but we're having a good year with our OpenPort product which is much higher ARPU and both of those items are accounted in voice and data ARPU.

And so what we've seen of late is a pretty stable ARPU in voice and data year-over-year with those two completing factors kind of just offsetting..

And how about the commercial push-to-talk, that should be in the category.

Also, is that having any kind of a positive impact yet?.

It's having a positive impact, though not material so..

Okay.

And when did some of those large government military type contracts actually begin to make a difference?.

Yes. I mean, I'm certainly hoping in 2017.

There's a long, long list of customers who are trialing, demonstrating it, starting activations, but it has been maybe a lesson for us as we've realized these are really solution sales and so the selection process typically of governmental agencies are long and drawn out and while there seems to be a still lot of enthusiasm for the product, as evidenced by the pipeline of deals all around the world.

I think I mentioned in previous quarter call how many countries are really involved with selecting the opportunity. Activations have been slower than certainly we'd like and I just think it's the normal sales process, it takes a long time to come to fruition. But we still have a lot of enthusiasm for the product going forward..

You touched on this a bit, but with the timing delay on the constellation coming online and hardware shipments, obviously there is always the risk that customers who are waiting for Certus or other new services may delay buying hardware, can you give us sort of the nitty-gritty of where your partners and vendors are in terms of their rollout and maybe not product-by-product, but in terms of general classes, when it will be available and do you anticipate any problems of customer delays based upon the current launch delays?.

Yes. So we're talking Iridium Certus. And, in fact, we just had a partner -- meeting partner workshop in California a couple of weeks ago. So this is all very fresh in my mind. There was clearly a lot of enthusiasm for Iridium Certus from the hundreds of people that were there.

They see the advantages over it in terms of cost, coverage, price, et cetera and so are really looking to get forward -- get onto it. The VAMs, we have four VAMs as you know that we have license to create Iridium Certus products, showed actually their initial products in the maritime, aviation and terrestrial segments.

So, that also added to the excitement, because people are now seeing name brand suppliers, showing real hardware that is quite attractive in size and capabilities, et cetera, to anything that's really out there. So, another reason for enthusiasm.

We're right in the process right now, this year, of signing up, selecting, approving the service providers that are allowed to sell the product. We've had initial discussions with just about everybody around the world and are going to be in negotiations and selections of those over the coming few months.

And, I think, once those selections happen and those people start announcing, you'll start hearing a lot more details from them about the product and what they're taking to market and who they are taking the market to. The overall product itself is in the testing phase. We continue to talk about the milestones.

The core hardware platform for the terminal is pretty much done and it's in the hands of those partners, that's how they're building their terminals. The satellite software, et cetera, is being tested thoroughly and pretty -- and all that should go to completion through demonstrations and datas to around the middle of next year.

We've been consistent with that. The only thing that really has moved here has been the first launch date and the number of satellites deployed for the previous question about how many satellites would it take to turn on and generate revenue is the only thing that's really changed.

I believe that if anybody was on track to deploy Iridium Certus for a fleet of ships, for example, would not be thinking that differently. I don't think, even with the SpaceX launch delay, about the opportunity in the market in 2017 and 2018 for using that product on their fleet.

So, not a lot has changed really from that perspective; the enthusiasm hasn't changed nor the technical platform hasn't really -- is still progressing well. There are three market segments. So we have maritime, aviation and terrestrial. I would say that maritime is probably first out of the box.

A few products that would go into that market will be the first ones to hit. Aviation, there is a ton of enthusiasm about because of all the advantages Iridium Certus brings in the aviation sector, but it just takes time to certify solutions and get STCs and all that sort of thing.

So I don't think we're going to see a lot of aviation hitting the market in 2017. I think that will be more in 2018 and 2019, as the network is complete. And then terrestrial, there's also -- we aren't even in the terrestrial broadband space today, whatsoever. And I've seen some really cool products demonstrated or shown.

And I think, for the first time, you could see some products coming out next year that could be attractive and would really potentially drive some nice sales particular in 2018. So, all hands on deck and Iridium Certus is going well and I don't think really anything has changed much there..

Okay. A final question, you mentioned some positive traction with the DeLorme inReach which has been a long time good performer for you. I haven't been to a retail store in years ever since Amazon came into existence. And at least on Amazon, it's still branded DeLorme.

So I don't know in-store at Cabela's or anybody else whether they're yet branding and expanding that distribution to get the full reach of Garmin.

When does that happen? Do you have a sense of when they'll begin rebranding and expanding that distribution possibly in time for Christmas?.

Yes, honestly, I think you really should talk to the team at DeLorme and Garmin about that. I think, as Tom mentioned in his remarks, we haven't seen it either and yet we're getting great performance from that customer, that partner. Really it's been a great year for them and for us as a result of that.

And I think we're excited, they have shown us a future technical road map and I'm pretty excited about that too and what the potential for it is.

I do believe when they do start, whether it's putting the name Garmin or certainly running it through the Garmin distribution channels, particularly in other parts of the world, because I think DeLorme has done great job in North America, particularly and a couple other countries as well.

But Garmin government has a much wider distribution and can, I think, accelerate the global roll-out of that product once it really starts hitting it. I don't know if it's going to be by Christmas, you'll have to ask them. I certainly would expect it to happen more in 2017, regardless..

[Operator Instructions]..

So, I think, it sounds like we covered the ground for everyone. I'd say it was a good quarter and look forward to seeing you all again on our fourth quarter year end call and also to give you further information on our Iridium NEXT plans as soon as we have them. So, thanks everyone for joining the call..

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