Lumentum Holdings Inc. (LITE) Earnings Call Transcript & Summary

All right. Welcome, everyone. Thank you for joining us at the OC event, and this is Lumentum analyst and investor briefing. So thank you for joining us. We are also webcasting this event. So there's [ 150 ] people on our webcast, right now. I'll take questions after we're done with our prepared presentation from the audience here. So it will take us about 25 minutes or so to get through our slides, the prepared presentation, and then you can feel free to step up to the mic. I'll take questions from the room first and then, if we have time, I'll take questions from the virtual audience. So I'm Kathryn Ta, by the way, I have introduced myself, but I'm the new Head of Investor Relations here at Lumentum. And I'm very pleased to have Alan Lowe, President and Chief Executive Officer here and also Chris Coldren, Senior Vice President and Chief Strategy and Corporate Development Officer, will also be on the stage later in the presentation. So we will be making some forward-looking statements. And so I encourage you to read the safe harbor. The forward-looking statements are based on our expectations that are reasonable, given the data that we have today. And they do reflect some risks and uncertainties, and see if you look at the total -- complete risk statements that are in our filings with the SEC. We have no obligation to update such statements, but we will be within the forward-looking statements. All right. And with that, I would like to give the stage to the Alan.

Great. Thank you, Kathy, and welcome to the team. It almost feels normal, again, and it's great to have people in the audience and back at OFC and walking through the convention center and seeing customers face-to-face, it really does feel good. So welcome. Thank you for coming today, and thank you for joining online. There's three things that you can leave this meeting with, they're right here on this chart. I think everybody believes that bandwidth requirements continue to be unrelenting and growing at a 30% compound annual growth rate and continue to do so, fueled by all the things that we do today, virtually as well as all of the deployments of 5G that are making the edge of the network need more and more bandwidth. And so that drives bandwidth through the network and through the data centers. And I have a few charts to talk about the growth in these various areas, but we are, truly, believers. And I was on a panel, yesterday, talking about supply chain. So we'll talk about that a little bit. And I jumped a little bit, but I do think this time is different, and demand is very strong, fueled by fundamental drivers of bandwidth requirements. And so I do believe that we're investing, in anticipation of this continuing, and we're really making sure that all the factories are lined up for satisfying our customer demand, as we get through some of the supply chain challenges that we have. As we entered the pandemic, we were in the early stages of about to just ramp new technology, whether that be the new ROADM architecture for our customers or the higher speed 400, 600 and 800 gig speeds of wavelengths. That got delayed, and it is now really in the early stages that should have been probably 1.5 years ago. And so we are ramping up our capacity in anticipation of very strong demand in new ROADM architectures as well as higher-speed transmission products at 400 and 600 and 800 gig. So additionally, in the data center, and I was sitting in the panel with one of the network architectures from Meta yesterday, they're increasing their CapEx, year-on-year, by 50%. So it is -- the drivers for demand inside the data center are phenomenal as well as machine-to-machine interaction, as AI becomes more real. So we're investing also to increase our capacity in our datacom [ EML ] chips, in response to this, what seems very, very rapid growth of datacom. And the fundamental economics for hyperscalers to transition to these higher speeds is phenomenal. When you can get 400 gig in a transceiver that was doing 100 gig and have that 4x bandwidth go through one fiber, the economics are very, very compelling. So that shift is happening, and that's why we're adding capacity. And our EML capacity has grown rapidly year-on-year, and we'll continue to do so. Chris is going to talk about the new opportunities for our 3D sensing in automotive, industrial and consumer markets. And we believe that the fundamental technology that we have today for our consumer, lead customer is -- has given us a great learning and a great ability to show to other industries the high-capacity, high-reliability, high-quality aspects of our products and technology. And as we advance that technology and invest in other types of VCSEL arrays and other kinds of lasers, for example, for LiDAR, I get pretty excited about the opportunities ahead. So I think we're in the early stages of that. And automotive is going to take some time, but we are investing and have invested probably for the last 3 or 4 years, and we're starting to get design wins that we announced last quarter. So to take a look at what we see as the bandwidth requirements and growth in CapEx, 20%, and I think that's probably understated, as you look at the different bars here on the chart. I already talked about internet traffic, but bottom line is, growing 20% year-on-year is fundamentally driving our requirements to add and invest in our infrastructure and our ability to grow our output for our datacom chips, our telecom transmission chips, and that means investments in fabs that we started a couple of years ago that are now coming online. What this means is, really, that we're seeing accelerating demand and the need for these higher-speed transmission products and new ROADM architectures that really drive the capability of satisfying that end-user demand that is really driving the fundamental need for adding this capacity. If you look at the telecom side, as I said earlier, that the new network architectures are fundamentally more compelling to our customers. And as I said before, in 2020, those network architectures should have been -- started to be deployed, and we started that really in late 2021. And if we had more semiconductor chips, we would be on our way to really, really ramping our WSS units. If you look at the 400-gig-and-above core count, 75% compound annual growth rate. Now, that doesn't mean that revenue is going to grow 75% because as we go up the learning curve for some of these products, whether that be tunable lasers, modulators or receivers or the fundamental module, whether that be CFP2 or a ZR or ZR+, those core counts are real. And I fundamentally believe it, and that's why we're investing in our ability to meet this demand, and that's why we're so excited about our pending acquisition with NeoPhotonics as well. On the WSS front, our customers are betting their next-generation architecture on Lumentum. In most of the cases on these very high core count or CDC type of applications, we're sole sourced with our customers. And that's why we've invested ahead of time to make sure that as they ramp their production and as they ramp their network deployments, we'll be there for them because we want to make sure they continue to work with us in R&D, and we invest in their future for them. And so it's a very, very exciting time for the telecom network deployments and our investments to satisfy their needs. And I talked already about this a little bit, on the hyperscalers. I mean, the economics are just super compelling. We are very constrained on our ability to ramp. And I think, year-on-year, we were up 45% in our EML output in the December quarter and growing each quarter this year and adding additional capacity that will come on late this calendar year to enable us to continue to ramp, to meet this kind of what seems to be insatiable demand within the data center and the hyperscale space. We continue to invest in R&D, across the board, in all of these different technologies to be ready for the next generation. I was meeting with one of the engineering architects of one of the very large hyperscalers, and she'd gotten some early test results for one of our more advanced datacom chips, and she was super excited. But now, they're changing their architecture for 1.6 and 3.2 terabits, using our datacom transmission technology. So exciting time across the board. Datacom, hyperscale add to the network, and we're investing to satisfy that. So with that, I'll turn it over to Chris.

Great. Thanks, Alan. So turning to 3D sensing, industrial, automotive, consumer opportunities for us. These are exciting as well, provide a lot of avenues to growth. For those that are following along via webcast or have the slides on our website, we're on Page 8, I believe. What we show here is a range of opportunities that we're pursuing in the consumer, industrial, automotive space with our 3D sensing lasers. In general, we play at the laser chip level today. That's because we provide a lot of differentiation at that level, and the industry structure is quite favorable. While there's a lot of folks pursuing the applications that are listed the top here, very few of them have wafer fabs, if you will, producing in millions or hundreds of millions of chips per year. That gives us a considerable competitive advantage in the market. And I draw analogy to how we're playing this in reverse chronological order, but we have played in the datacom transceiver market and move down to the -- transceiver to the laser chip level, as Alan just talked about, given the favorable dynamics at that level of the ability to differentiate products. We're starting that way in the case of the markets that are shown here. That said -- well, and when you're playing the laser chip level, obviously, the -- as you see here across the selling prices of the type of products we provide, are in the dollar-to-dollar, each, and then you get multiple lasers per customer units generating several dollars or even tens of dollars in the case of LiDAR per customer unit. That generates, today, about $1 billion market that we're addressing. And over the next 5 years or so, we expect the market to more than double, with a pretty heavy contribution -- more than half of that growth coming from things like augmented and virtual reality and LiDAR. And in fact, in the case of LiDAR, perhaps the market even accelerates after that 5-year time point, given the time scales of LiDAR are such that probably the adoption rate is accelerating, moving forward. But looking ahead, we have the opportunity to increase both our dollar content. I guess, you would call it at the chip level. We're looking at opportunities to integrate more optics and even in some cases, electronics chip-scale integration, if you will. But we also have the opportunity to maybe even move up to modules and subsystems, depending on the opportunity and the industry dynamics. Probably the most obvious one that's shown here, in fact, we have listed as a module-level solution, is providing coherent LiDAR modules, if you will, into industrial, whether they be metrology-type applications or other 3D imaging, if you will, in industrial opportunities, where precision and performance is really paramount. It's taking the page from what we do both in 3D sensing but also what we do in our coherent Telecom Transmission business. But we continue to look at where we play in the ecosystem and whether there's logic to moving or changing over time and/or when we would intersect. But [ plant ] the chip level today is what we do, and it's a very healthy. I think you can see that there's several consumer-based opportunities here, really the inflection here, we expect, over the next two years. And the augmented and virtual reality brings a pretty exciting, new opportunity here while similar type of laser to what's used in the smartphone and tablet space but perhaps more lasers per customer unit, given that you need eye tracking, gesture control or gesture tracking, world-facing. So there's multiple applications within a glasses or headset environment, beyond what is needed in a handset, if you will. And then, the other major driver of growth, as we mentioned, automobile, as Alan highlighted, and we talked about on our last call, we've got a range of customer design-ins and partnerships that are underway, that range in dollar content in the tens of dollars, could be even higher, in particular, one of the things we're most bullish on state LiDAR. We think it, one, solves a lot of reliability concerns but also manufacture ability to drive cost, and in order to achieve that all-solid-state functionality, for example, the partnership we highlighted with leverages our multi-junction addressable or the elements of the VCSEL array or addressable, that's how lean steering is accomplished, very, very high VCSEL content, if you will, in that solution, but the system-level solution, very cost effective, very reliable. Often doesn't get a lot of attention, but we've got a great [ LiDARs ] business. We've got industry-leading profitability in this business that's driven despite being a smaller player relative to some of the bigger guys that are out there. That said, we derive that great profitability via leveraging the scale of the rest of our business and particularly in the telecom side of things but also, the experience in learnings and developing products that are going into higher-volume applications, ones that you don't go in and service a telecom or module, if you will, in the field, the way many industrial lasers or customers are used to. So the fact that we come out with very robust products that are very reliable, less service needed, it creates a lot of customer value proposition. There's a lot of new markets that are emerging for us, whether that be in next-generation semiconductor device fabrication, solar cell fabrication, display manufacturing. And when you look at electric vehicles across the board, virtually -- the body, the motors, the batteries, the power electronics are all heavily reliant from laser-based manufacturing. Given all those opportunities, we're confident that we can grow faster than the market, introducing new products and picking up market share. We hit on a lot of these topics, but I think it's important to kind of step back to the 60,000-foot level. There's multiyear secular trends across consumer, industrial communications, where we play today. There's also new opportunities, whether they're in like sciences, aerospace and defense, even -- that we don't even touch, really, today more peripherally but are emerging to be very reliant on photonics, provides not only long-term tailwinds for the markets we or play into, but also new markets to enter to drive additional growth. Turning to Page 11 for those following along. On our last earnings call, we highlighted in the press release and in statements but maybe a little less quantitatively, so we're breaking that out in more detail here that we believe, at least for the next several years, we can grow at a nice double-digit CAGR, and we're providing that breakdown here to get to that math. And what you can see here is in the telecom and datacom portion of our business, we feel that the market is set up for very strong growth. On top of that, we've done a lot of work to clean up, if you will, our product portfolio, exiting product lines that are based on technology that would be cannibalized by this upturn in 400 gig and above speed. So that was a negative hitch to growth over the past couple of years but then, should pay off, in terms of allowing us to be over indexed, if you will, to the high-speed growing portions of the market. Next, turning to the industrial and consumer portion, as shown here. To be clear, this is really -- as we report out our numbers, so this is 3D sensing and some late industrial [ diode ] laser semiconductors, does not include our commercial lasers business, which is -- we'll get into, next, and that industrial consumer, both our market and revenue today, heavily, 3D sensing in the mix. What we're showing here is minus 5% to plus 5% growth rate. The reason for that is not because there's not great opportunities, but we're trying to be very conservative in our outlook, given we have a very strong market position, and certainly, we have the potential for share to normalize over time. We have nothing to say that, that's going to happen, hasn't happened over the past 5 years, but we want the street to think about this as if you -- if things don't revert to some kind of equitable market share distribution between us and our competitors, and that's all upside, but if it does, "Hey, we still have a very great growth outlook for the combined portfolio of products we provide." Again, nothing to suggest there's a change from the past in 3D sensing. But for modeling purposes, I just want to take a conservative point of view and hence, why we said greater than 10% didn't put a bound on the upper side of it. Then in the laser space, we believe we can grow quite a bit above market growth rates. So in this case, 10% to 15% CAGR, and that's really driven by our increasing exposure to new applications. Over the past number of years, our team has been very focused on building a great cost structure and manufacturing platform for our lasers to enable us to, at the same time, take the brand-new products we've been developing, be very cost effective to be able to enter the markets that we haven't played in, historically, or new markets that are just emerging and be extremely competitive. So the upshot of greater than 10% CAGR, if you kind of revenue weight all of those growth rates across in our portfolio. Then -- there we go. I want to reiterate our financial target, financial model, 50% growth rate, 30% or sort of 50% gross margin. 30% operating margin with the greater than 10% growth rate, highlight though the LTM comp that's shown on this page, obviously, down from fiscal '21. But we highlight that in the past two quarters, we've had more than $90 million of negative revenue hit due to supply shortages. That revenue is not perishable. It will be in future quarters. And there's more. Obviously, everything I've talked about to date has been organic growth rate. We've got the exciting NeoPhotonics transaction that we announced in early November. This gives us even more exposure to the 400-gig-and-above tailwinds that are in the market as well. Obviously, there's a lot of other benefits in combining the -- efficiencies in combining customer relevance, et cetera. And we've got a playbook. We think we did okay, did a pretty good job on the Oclaro transaction, and this has a lot of similarities, at least from the operational side of things. So we've got that playbook to pull thumb and really are confident that this will create shareholder value in the coming years, in addition to organic growth and target financial model that we showed earlier. So based on that confidence, a couple of things that happened literally last week, our board authorized increasing our buyback program, share buyback program that is to $1 billion, added a little extra time out to May of 2024. We've been pretty aggressive about buying back shares to date. So that's why we felt that increasing it and extending it was warranted, particularly given the confidence in our growth outlook. It's great to take out some shares today and reward those who stick with us on the growth that is expected to come. At the same time, we did a convertible notes offering last week. And concurrently with that, purchased $200 million worth of stock. That's not included in the $1 billion, so that's in addition to the $1 billion that we expanded our program to. And then finally, before wrapping up here, I just want to highlight -- and I want to make sure I got my notes here to make sure I hit all the exciting products, it's a long list. Here at OC, a lot of things we're demonstrating, and you can go see, showcased in our booth, really recommend stopping by and talking to our team, but we've got a high-performance 400-gig DCO modules in a demo as well as in our booth, live traffic. These are high-performance products, 800-gig coherent modulators for next-gen telecom transmission. Also, didn't highlight it as much in our presentation, but another exciting opportunity is DWDM and moving out to the edge. And so we've got a rapidly growing business, providing 10-gig and 25-gig tunable transceivers, and these are high-performance telecom transceivers in the same form factor as datacom modules plugging into cable MSO, remote opportunities, 5G backhaul or sort of fronthaul applications and really growing nicely. We also see next-generation ROADMs, much higher [ core ] count to enable increased scalability. We've got a great new product just starting with we call ROADM node on a [ blade ]. So over the years, we've taken the opportunity to integrate selective switches, optical suppliers, channel monitoring, integrate all of that in a compact blade. Now, we're integrating multiple of those, if you will, on a blade, so that customers can reduce the number of swaps taken in their systems by ROADMs and amplification, increased the number of 400-, 600- and 800-gigs [ ports ] that can plug into their chassis, reduces space for their system, increases value for their customers. On the Datacom side of things, as Alan highlighted, some exciting stuff. 200 gig-per-lane EMLs that enable 800 gig, 1.6 terabit and beyond datacom transceivers, 100-gig DML. So this enables a next generation of 400-gig transceivers that will be coming down the pipe, that are lower cost for customers. And then two other, we have a suite of high-power CW lasers, so providing light or photonic transceivers for shorter distance, lower-performance applications, so that we also tap into that market, in addition to the EML, DML portion that we provided. And then a suite of VCSELs, right? We've got a great VCSEL platform, have shipped billings of them into the consumer electronics space, leveraging that same platform to be able to provide VCSELs into the datacom market and have exciting 25-gig, 50-gig VCSELs for shorter-reach applications. So a complete portfolio of products for data center -- next-gen data center applications as well as next-gen datacom or telecom applications. So I think we hit on a lot of these points. It's a really exciting time in the photonics industry. We're seeing tailwinds across all of our markets today. New markets are emerging for our photonics that -- electric vehicles and virtual reality, autonomous vehicles and the like, and we're armed with differentiated products, customer design wins with market-leading customers. So with that, I think, with that, I managed the time well there, we'll open up the Q&A session.

[Operator Instructions] Let's take the first question.

Simon Leopold with Raymond James. Two, if I might. One, sort of, mandatory that we have to get a check on the supply chain constraints. So I believe, last earnings call, you talked about primarily affecting ROADMs, but also affecting your datacom transceivers and lasers. Could you give us an update on what's constrained in your outlook for recovery? And the second thing I wanted to ask is, if you could give us an update on essentially your relationship with Huawei as well as your opportunities, as operators maybe try to move away from Huawei to other OEMs that might customers?

Okay. Yes. Thanks, Simon. I'll answer the second one first. We still have a very good relationship with Huawei. Over the last couple of years, they've gone from a very large customer to a smaller customer because they've moved everything they could away from U.S.-based suppliers, and they buy what they can only get from the U.S.-based customers. We're sole-sourcing a lot of products there. We have products where the alternative is a U.S. manufacturer. And so from that perspective, we're going to continue to sell products that we can sell to them that they can't get elsewhere. So that's number one. I think, number two is, to your point, as network operators around the world make a shift from a Huawei-based network to a Western-based network, whether that be Ciena or Nokia or Infinera, that's actually a tailwind for us because we have a much larger share of wallet of those other network equipment manufacturers. And so as we lose a dollar at Huawei, if you will, we probably gain $2 at Western network equipment manufacturer as those networks shift. And they're starting to shift, a lot of activity around it, but I'd say that's really going to come in a more meaningful way late this year and into calendar '23. As far as the shortages are concerned, we are impacted mostly by semiconductors that go into our ROADMs, but we're also impacted by semiconductors that go into our transmission products. And Chris mentioned the Remote and cable and [ missile ] operators and driving 10-gig and 25-gig transceivers. We're impacted there but to a less extent. We're able to get some of those things resolved. There's a lot fewer semiconductors to work on, on a small transceiver like that as opposed to a ROADM or ROADM line card, where there's many, many semiconductors, and the surprises that we get on a weekly basis of, "Oh, you're not going to get the stuff before you're going to get," I would say, are lessening. And so we're more predictable, with respect to what we can load our lines with. But at the same time, we're also increasing capacity for our ability to manufacture ROADMs because if the supply chain is disrupted and we lose manufacturing capacity, and then all of a sudden, we get a whole bunch of chips. In the past, we haven't been able to run those through. So we're adding [ run ] capacity, so that when we are able to get those semiconductors, we can flex up in our capacity to meet their demand. I'd say, our datacom is really constrained by our ability to grow our wafer fab capacity in Japan. And we did, year-on-year, 45% in the December quarter on our EMLs, that's going to continue to grow in the March, June and throughout the calendar year. But that's really controlled by us. We don't -- we're not really reliant on others, other than equipment coming online, and those lead times have actually gotten longer as well.

Thanks for the question, Simon. I just wanted to mention that the slides that we just showed are available on our website that also will help our virtual audience to follow along with the Q&A. So next question.

Thomas O'Malley, Barclays. You showed an interesting slide, showing of opportunities in 3 across the board consumer applications. Then you showed a slide about industrial consumer, your 3-year a flat guide. Could you talk about how much in that flat guide embedded for non-smartphone 3D sensing applications? What are you assuming to fix that flattish from ?

Yes, I'm reluctant to get into much more detail, but I would say that we're assuming that within that range, from minus 5% to plus 5%, it's probably back into a scale of what is new, if you will, driving in the mix, if you will. So we will have more customer and application diversity in 5 years from now than today.

And then, just a second one. Just may ] a little less than some other people at the show. You're hearing more and more about co-package optics, and the timing of that kind of in the air,right now. One, can you just comment on when you actually see that transition taking place or ? And two, how would it impact the ?

Yes. We view co-packaged optics, in general, as a significant opportunity, what -- where there's really two places co-packaged optics could go. The first is, at some level, replacing datacom transceivers as they are used today. But at the same time, you continue to need semiconductor lasers to through silicon photonic devices. They tend to be at the lower-end or lower-performance spectrum. So that's why, as I highlighted the products we talked about we're showing here, high-power, continuous-wave lasers that are used for those co-packaged optics opportunities. But the really exciting piece of co-packaged optics, at least in my point of view, is that today, there's still the market for very short-reach interconnects on board, connecting CPUs or GPUs. The memory, for example, is through copper interconnects, and we see copper's demise at some point and that being replaced by short-reach optical interconnects. The volumes there are astronomical in every one of those [ links needs ], laser light going through it, and that's an area where we're developing very high-performance multi-wavelength modules to be able to serve that market and create an upside to any of the numbers that we've shown here today.

We'll take the next question.

George Notter from Jefferies. I guess I wanted to ask about the supply chain environment and as you walk on the trade show floor and talk to folks, it sort of seems like there's certainly appetite to build inventory at customers. And I think that's true, if you're a carrier or you're a system supplier, it seems like there's incremental inventory that people are attempting to build. So like -- what insights do you have into your customers building inventory? Are the orders that they placed upon you non-cancelable, like how described the environment and the risk, the momentum?

Yes. This was a good question, yesterday, for the panel because we had one of our customers and one of our suppliers on the panel as well. And from my perspective, telling customers they can't cancel orders or push orders out, is not really a partner. And so we really work with our customers to understand transparently, are these products being deployed, where are they being deployed? And is this real demand? And I think from our perspective, we're so far short of satisfying real demand or stockpiling of demand that at least in the next 6 months, that's going to be the case for sure. I'd say that the earnings release yesterday was pretty telling. I mean, their order rate was phenomenal. And whether that gets deployed over the next two quarters of the next year, is really a function of our ability to get the semiconductors that they need to be able to satisfy their customers. So certainly, in the short term, there is no stockpiling. I think there's probably a desire. I would love to have a week of semiconductor chips, so that I can deal with an airplane not taking off on time. But today, there is no flux in the supply chain. So if a plane gets delayed, I don't get parts to the line, and I lose capacity. So -- and I think that's going to be the case. I mean, Chris talked about the $90 million that we've missed in the last two quarters. I think, that's going to actually grow in the short term because the order rate, to your point, is actually extremely strong across the board. And as we add capacity, we'll hopefully take care of that by getting to the semiconductors we need for our customers.

Thanks, George.

I think -- yes, [ non-degree ] with cap. So two quick ones. First quick one, the growth rates that you gave, does that include NeoPhotonics?

It does not contribute NewPhotonics but that would be to the numbers.

Got it. Got it. Any perspective on how the numbers might change, if ] NeoPhotonics?

Well, I think, certainly, we're very bullish on the opportunity that they have to grow, if you will. And so when that business becomes part of us, it's certainly going to grow at the stand-alone opportunity and probably even faster, given our ability to strengthen their balance sheet, if you will, by being part of us and giving customers confidence that we'll continue to expand manufacturing capacity. So I think, by over-indexing to 400 gig and above, and what's beyond that is going to continue to drive strong growth rate. I think, as you saw in the chart the highest growth rate amongst the segments with telecom and datacom. And amongst telecom and datacom, 400 gig and above, telecom is the highest growing portion of the business. So I'm not going to give numbers at this point, but stay tuned for that, when the transaction closes, but it certainly should accelerate growth rates on an even bigger base.

Yes. Maybe if I could just add. I mean, one of the deals -- one of the parts of the deal was that we offered NeoPhotonics a $50 million loan, so that they could invest to meet the kind of demand we talked about. And so that's also a signal to their customers that we are behind them, whether it closes next month or several months from now, and gives them the ability to ramp up their capacity to levels that the hyperscalers of the world need today.

[ 11 billion ] by the way, really was -- seriously. And second question is, Chris, when you were talking about 3D sensing and the upside to downside, I thought I heard a comment about if we return to equitable market share. Did I hear that accurately? And could you give us the context around that?

Yes. And I think the key point here is, we, for a number of years, have now maintained an outsized market share position with very large customers, who are trying to -- not that they want to take us down to non-leadership position in the market, but at least they have seen value in having multiple suppliers. And so our -- for modeling purposes, again, our assumption is, okay, that share will normalize to some more, as you said, more equitable position. And if the last 5 years predict the future, that's probably a conservative assumption. But certainly, as we talk to investors and analysts, everybody here are very nervous about that situation. And so we feel, and we're certainly not seeing the value of it in our trading performance, if you will. So I think if folks can look at that and think of that as -- assume the share normalization, and if we overexecute and overachieve again, then that's only upside. But we don't need that to drive the kind of growth rates we're talking about.

And can you just remind us quickly, when do those contracts get set? Because I think there's volume agreements in the contract today. And so what's the cadence of those contracts getting?

Yes. They're typically multiyear. And then, as new products get launched, that's typically where we get more share because we're typically the R&D partner of choice and the one that has shown that we can go up the ramp better than anybody else. And so typically, then, as you add a new product, you have that negotiation discussion. But it's typically, multiple years.

Thank you, another -- take the next question over here.

It's Rod Hall, Goldman Sachs. I just want to come back to the industrial consumer growth rate. Obviously, it will be difficult to do this the full year, getting the call on. So I'm wondering, how we should be thinking about the pacing of that growth, as we have seen flat over the period that you guys ]. That's, kind of, first question or about a little bit...?

Yes. I mean, it's a little bit difficult to say what the next year, the next year, the next year is going to be. That's what we kind of took the 3-year CAGR approach to say, at some point, that's long enough in the future to say that this assumption of share normalization materializes. That said, as we look to the next year, great products, we performed very, very well this year. There's new dollar content opportunities coming. So we continue to see upside with the modeling purposes. We want to keep people thinking about that upside as upside.

And then on to the next question to the opportunity as you look out in the next fiscal year, we know, not this year but next year. Is it the material or is it a year?

Maybe, say, differently, I think, as you look out in 2025, it's a multi-hundred million dollar market. Now, as you look nearer term, it's going to be, obviously, determined by which customer launches which product , and there's some pretty heavy hitters. So it's not a 1,000 customers that are launching products in the next 1 to 2 years. So the timing is really dictated by those customers ramping. But I think, as we look at least the next two years over that 5-year time horizon, we expect that 1 or 2 major customers will launch reasonably significant programs to public consumer, if you will. So probably in the next 12 months, it's relatively modest, just as we sit here today. But if you ask me, 12 to 24 months, I think we start getting meaningful revenue from that.

Okay. And just one more with , which is, it may be a non-acquirer. We started to think about the end exposure for a lot got a long history in the supply, back when was . And so I'm just wondering, is there opportunity there? And I know a lot of the parts . I'm curious what you think there...

Yes. I mean, Chris talked about some new opportunities for optics in new markets. I mean, we've, probably for the last 9 months, have a dedicated team looking at what other markets that we don't participate in, could utilize our technology and EVs and batteries, things like that, but also, defense is also something that we're looking at. We may do that completely organically, but we also may look at inorganic small acquisitions to be able to accelerate our entrance into those kinds of markets.

.

It -- certainly, sensing is one of them, but there's a lot of focus on things like directed energy, right? And as somebody that makes a lot of semiconductor laser diodes in the United States, that turns out to be an interesting opportunity as well. But there's a range of things in that market.

Next question.

Fahad Najam from Cowen. I want to go back to your long-term outlook for the telecom and datacom market, how much of that outlook is based on exposure to China? And are you seeing similar trends out of China as you see for the . Any kind of color you can add to the geographic outlook?

Well, I can talk about China in the short term. I'm having more meetings with our friends in China over the last 3 months than I've had over the prior 2 years. And so the demand for our products is very strong today. I'll let Chris comment about how does that factor into our overall double-digit outlook for the next couple of years. But we have not seen a slowdown other than the 5G, and we talked about that about a year ago that 5G deployments have slowed down our DML shipments into China. That inventory was built up. It's starting to get depleted, and we're starting to see that in the second half of this calendar year that we should be shipping again meaningful DMLs as 5G really gets going again. But I'd say that, from my perspective, Huawei is not the only game in China. ZTE and FiberHome are becoming more meaningful. And they're also trying to get outside of China, but that's still probably early days. But I think we're partnering with all of those customers to get the technology that they need for the networks that they need. And today, the demand is very strong there.

Yes, I just would add that the percentage of China in the mix is probably going to decline over time in that -- and so we're sort of under-indexed, if you will. The reason for that is both, we highlighted the point of gaining $2, when we sell to Western customer, if we lose $1 from Huawei as an example. But that also, in a similar vein, that the kinds of products we supply are fairly narrow range of products, meaning, one, that they can't get from non-U.S. suppliers, and some of them are lower ASPs. Some of them are higher ASPs, so the average dollar content is lower from those customers. Therefore, what's driving a lot of the growth is the Western world, if you will, but as Alan's highlighting, China is ramping and growing in next-gen technologies, just like the West. So the same networking trends and bandwidth challenges, apart in China, they maybe didn't come down just hard over the past couple of years with both COVID impacts to deployments and then maybe start ramping a little earlier some of the next-gen technologies. But in terms of the growth rates that we're projecting there, there's not a high dependence on China doing anything unusual and delevering a bit from China as the Western customers become a much more significant portion of our revenue.

Can you remind what China is percentage of ?

Yes. That's actually kind of a tricky answer because what we put in our filings is ship-to-type. And obviously, a lot of Western customers have us ship to China, where they manufacture. So I think you should think of China maybe being more, when you take a step back and say what percentage of the global growth in GDP RNA, that's kind of in the range of where our China exposure should land in the 30% of aggregate revenue, if you will.

And I apologize, one more follow-up on the comment about equitable market share in the consumer 3D sensing market. Are you assuming still a 2-player market in the 3D sensing consumer market or is it a 3-player market?

I'm not sure that necessarily matters a huge amount. We're just assuming is that we can maintain a very good market position, very good market share, if it wasn't for the recent past looking backwards, having to move down from that to still a very strong market share position. So this is an industry where, particularly at the chip level, large barrier to entry, in terms of being able to manufacture with that quality, reliability and performance, at the volumes that we're talking about. And so we've seen a very worthy competitor take years to be in a position to ship. And so we're not anticipating this being a 4, 5, 6 horse race. So if there's a third that emerges and picks up a little bit of business, I don't think it really changes, ultimately, what our long-term market share is going to be in the space.

We have a number of questions that have come in from our virtual audience. The first of which is Alex Henderson, Needham. And I think this is for you, Alan. So can you comment that, indeed, the industry is fundamentally different from prior conditions. And another way that you didn't mention is that it is much more consolidated. So how has industry consolidation changed customer and competitor behavior? What does it mean for us?

Yes. I mean -- thanks for the question, Alex. Yes, the industry has consolidated. And I think, if you look back 5 years ago, every 3 or 6 or 12 months, there was an auction, and we would have to bid on products, and then we get our share, and you'd live with it. And now, it's a very, very different situation, in that when a customer launches a new architecture or technology, we sit down with them and talk about what costs they need and what cost we have and how we come together on the specifications that allow us to meet their cost targets, but at the same time, achieve our profitability. And so then we sign up for multiyear agreements, where we know they're going to buy a bunch of product from us in a very high share, and they can count on our price reductions over time. And then we -- so it's very, very different, from that perspective. And I think that's why I believe that, whether all of the orders that we get from customers are real orders? It almost doesn't matter because a lot of the products that we provide today, we are the supplier for it. And we have contractual obligations to satisfy what we can to our customers at prices that are predetermined. So I do think, that's a very good point, Alex, and one that the dynamic of the industry is extremely different than it was several years ago.

I just want to add to that. I think, having been somebody that spent a lot of time talking with customers in a number of years ago, in this reverse auction environment. It was always very frustrating that -- it was difficult to get everybody on the page that, probably better for them and for us to be in a situation, we're much more focused on what can we do together to create more innovative and disruptive solutions, if you will, as opposed to both chasing a price spiral downwards. Now, the reality is prices do have to go down over time. That's just the nature of technology products, the ability to deliver bandwidth more cost effectively over time. But now, the ability to continue to make technological leaps to be able to solve bandwidth problems and in that process of customers, gain more value. We gain more value with the overall solution. So it's a much healthier for us and for our customer environment.

Great. So I have another question from . He is asking what is the thinking behind the note offering been $750 million but only using $200 million to . Why the need for additional cash on the balance sheet, and what is the purpose for additional cash?

I think, it's general corporate purposes, kidding. Well, we did also increase the buyback by $300 million. I think our perspective is, we have been aggressive at buying the stock. Our outlook for the future is very bullish, and continuing to buy back our stock at these levels seem to make a lot of good sense to us. And so we're also, as I talked about before, investing in capital for the needed expansion in our fabs, which are not cheap, as well as the back-end equipment and facilities that we are investing in today. So I think, let us get through the NeoPhotonics acquisition, but I think we will start that integration, and there may be something that comes after that. Nothing to announce today, but certainly having the ability to act fast on any kind of future acquisitions was something we were considering, but I'll stick with the general corporate purposes.

And I think, you can add like we look at our capital structure and how the company is financed continuously. And when you don't want to be reactive, if you will, in that kind of thinking, you want to be very proactive to be looked out over several years, expect a lot of growth in the company, have some uses of cash between buyback, between an acquisition that's pending. We have some notes that mature in 2024 as well. And when we looked at the overall economic environment, rising interest rates and all of that, should we wait a year and do something or go now and made the decision that's probably better now than later or at least the risk is higher, if you look into the future? And so that's why we took the opportunity, when we kind of saw the [ cloud ] part in the market, and we were able to get great terms and do a good deal.

Great. Another question from Tim Savageaux, Northland Capital Markets. He said, given the supply issues that you discussed, is it fair to assume growth in telecom and datacom, at or above the high end of the range in the near term, into the next year? Or is that recovery to ?

I mean, certainly for datacom because it's under our control, we will be growing faster than the market as we gain share. And as the shift goes from 100 gig to 400 gig and above in the data center. So I think, that's the datacom story. On the telecom story, as we said in the last earnings call, we believe that the June quarter will be better than the March quarter, and a lot of that is coming from continued growth in telecom. As we expect that the semiconductor chips will free up. If you remember, one of the things we said was that in the March quarter, we were really impacted by COVID and the impact it had in December. So we didn't get the chips we needed in December and January to shift through the March quarter. We haven't had that kind of major event disruption this quarter, so far. And so that leads us to think that the June quarter will continue to improve in telecom as well.

Great. We have one last question from the online audience. It's probably for Chris. Any update on NeoPhotonics, in terms of China?

Nothing to add, relative to prior commentary. We're going through the process, constructive dialogue with the officials in SAMR, and we highlighted, when we announced the transaction, we would expected it to close in the second half of calendar '22. And that continues to be our expectation. I think, as we all know that process always takes longer than the U.S. antitrust, in general. And everything is going as expected.

Fantastic. Okay. So we're at the top of the hour, and I'm sure we'll be around in the room, if people in the room wanted to stay back and ask a few more questions. But with that, I'll end the broadcast. And thank you for attending our interesting session.

Thank you, everybody.