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

Good day, everyone, and welcome to the Lumentum Holdings Third Quarter Fiscal Year 2022 Earnings Call. [Operator Instructions] Please also note today's event is being recorded for replay purposes. At this time, I would like to turn the conference call over to Kathy Ta, Vice President of Investor Relations. Ms. Ta, please.

Thank you, operator. Welcome to Lumentum's Fiscal Third Quarter 202 Earnings Call. This is Kathy Ta, Lumentum's Vice President of Investor Relations. Joining me today are Alan Lowe, President and Chief Executive Officer; Wajid Ali, Chief Financial Officer; and Chris Coldren, Senior Vice President and Chief Strategy and Corporate Development Officer. Today's call will include forward-looking statements, including statements regarding our expectations regarding the pending acquisition of NeoPhotonics, including market opportunity, expected synergies, financial and operating results and expectations regarding accretion, time to close, strategies of the combined company and benefits to customers and the markets in which we operate. As well as the impact of COVID-19 on our business and continuing uncertainty in this regard, trends and expectations for our products and technology, our markets, market opportunity and customers. And our expected financial performance, including our guidance as well as statements regarding our future revenues, our financial model and our margin target. These statements are subject to risks and uncertainties that could cause actual results to differ materially from our current expectations, particularly the risk factors described in our SEC filings. We encourage you to review our most recent filings with the SEC, particularly the risk factors described in the quarterly report on Form 10-Q for the quarter ended January 1, 2022. And those in the 10-Q for the quarter ended April 2, 2022, to be filed by Lumentum with the SEC today. The forward-looking statements provided during this call are based on Lumentum's reasonable beliefs and expectations as of today. Lumentum undertakes no obligation to update these statements, except as required by applicable law. Please also note, unless otherwise stated, all results and projections discussed in this call are non-GAAP. Non-GAAP financials are not to be considered as a substitute for or superior to financials prepared in accordance with GAAP. Lumentum's press release with the fiscal third quarter 2022 results and accompanying supplemental slides are available on our website at www.lumentum.com under the Investors section. This includes additional details about our non-GAAP financial measures and a reconciliation between our historical GAAP and non-GAAP results. With that, I'll turn the call over to Alan.

Thank you, Kathy, and good morning, everyone. Strong demand and solid execution by our team around the world resulted in all third quarter financial metrics being at the high end of our guidance range. We are in an excellent position to grow revenue in Q4 and in fiscal '23, and we believe we will see continued revenue growth beyond 2023. Long-term tailwinds are driving our markets and demand for our differentiated products is already strong and growing. Unrelenting growth in data generation and consumption is driving the cloud and networking markets we address. Our customers are just beginning multiyear infrastructure upgrades that require our leading-edge photonics. Use cases for our high-performance lasers for 3D sensing and LiDAR are expanding beyond mobile handsets. And in Q3, we announced a new reference design in building automation. New automotive customers are turning to our lasers for LiDAR, and we are engaged with more customers on extended reality applications. Demand for our commercial lasers also continues to grow as industrial and microelectronic factories and semiconductor fabs expand and upgrade their capabilities and increasingly utilize the leading-edge lasers we supply. Near term, telecom customer demand is outpacing the supply of third-party components, most notably semiconductors that we need to build many of our products. Our supply chain team is making excellent progress to alleviate component shortages, which we expect will drive strong sequential growth in telecom revenues in the fourth quarter. We expect this telecom growth combined with the increased output from our recently commissioned datacom capacity will more than offset normal 3D sensing seasonality in Q4. We expect Q4 revenue will increase sequentially and year-on-year. Our Q4 revenue outlook would result in a new company high for a fourth quarter. Though component supply is increasing, demand is growing even faster. We expect more than a $100 million revenue impact as a result of the gap between demand and supply in Q4. This is up significantly when compared with an approximate $65 million gap we saw in the third quarter. While we expect these supply shortages to continue to improve with the diligent work of our team and our suppliers, given the accelerating demand environment, we will likely see customer demand outpacing third-party material supply into calendar 2023. The NeoPhotonics acquisition remains on track for the previously announced time line of closing in the second half of calendar 2022. We are working diligently with antitrust authorities in China with their approval being the final key closing condition for the transaction. Integration planning with the NeoPhotonics team gives me strong conviction that the combination will create value for our customers, our suppliers and our shareholders through a more comprehensive portfolio of differentiated products for our cloud, networking and automotive customers as well as meaningful cost synergies. Now let me provide some detail on our third quarter results. As expected, telecom and datacom revenue was down quarter-on-quarter due to supply constraints despite very strong demand. New applications for our 10G tunable transmission products are accelerating. These include metro access and fiber deep applications for cable MSO and networking customers as well as wireless fronthaul for mobile networking customers. The wireless fronthaul application is just starting to be deployed and is expected to deliver meaningful revenue by the end of the calendar year. We are increasing our manufacturing capacity for our 10G and soon-to-be released 25G tunable products in our wafer fab and our back-end assembly and test factories to address the rapid adoption of this differentiated and enabling technology. Pump laser sales are robust and grew more than 50% from the same quarter last year. As we have mentioned previously, elevated pump shipments frequently have been a leading indicator of future telecom demand. In addition, submarine cable suppliers are deploying subsea cables at record levels, driven by the robust demand from hyperscale data center operators. This is another leading indicator of future telecom demand. We expect these infrastructure investments will help propel Lumentum into double-digit growth starting in fiscal '23 for multiple years. EMLs serving high-speed cloud data center applications reached a new record in revenue. New EML manufacturing capacity will allow us to ramp our datacom shipments even more to help us better fulfill strong customer demand for our differentiated products. Accordingly, we expect fourth quarter EMO revenue to increase significantly from the third quarter. Looking ahead to the fourth quarter, we expect telecom and datacom revenue to be up strongly quarter-on-quarter due to the improvements in IC component supply, but still significantly below the level of customer demand. We continue to work diligently with our suppliers and on alternative sources of supply to alleviate shortages. Turning to industrial and consumer. Third quarter revenue was down from last quarter as expected due to 3D sensing seasonality. We are expanding our 3D sensing and LiDAR platforms into new applications in the industrial market. In the third quarter, we announced a reference design with Ambarella for building automation and occupancy sensor systems. The design uses Lumentum's flood illuminator module for high accuracy time-of-flight 3D sensing together with Ambarella's AI system on a chip, enabling the application of small sensors with local processing for occupancy monitoring, intelligence space management and smart retail. In automotive, we have expanded our development activities with new LiDAR customers, and we are very pleased to have entered into a customer-supported development agreement for a long-range LiDAR with a market leader in the ADAS space. In addition, we have begun our production ramp of our multi-junction VCSEL arrays for Hesai. The customer pipeline for our products serving in-cabin driver monitoring systems is also growing. In addition, we have early product traction with multiple customers who are developing extended reality solutions, which we expect will come to market in 2023. We expect fourth quarter industrial and consumer revenue to be down sequentially with typical consumer product seasonality. Our commercial laser revenue was up again quarter-on-quarter as expected, achieving near-record levels, primarily driven by fiber lasers serving automotive and industrial applications. Ultrafast lasers for manufacturing of semiconductors and consumer electronics also grew sequentially. Looking ahead to the fourth quarter, we expect laser revenue to grow again quarter-on-quarter, driven by new products and the overall market. We expect laser quarterly revenue to surpass our previous record as this business grows over the coming quarters. Before turning it over to Wajid to run through the numbers, I'd like to acknowledge our employees' commitment to implement sustainable practices. To meet our company-wide goal of net zero carbon emissions by 2030, we have transitioned more sites to renewable energy. Since January, our site in Ottawa holds a renewable energy certificate, and we installed solar panels in our site in Slovenia. Our sites in the United Kingdom started procuring 100% renewable electricity in May, and our San Jose headquarter has achieved the Lead Silver Certification, another step forward in our goal of net zero emissions. We are also very proud to have achieved the EcoVadis Gold Rating for Lumentum's advanced performance and sustainability. In addition to our progress in sustainability initiatives, I would like to thank our employees around the world for all of their hard work and resilience during such challenging times. With that, I'll turn it over to Wajid.

Thank you, Alan. Net revenue for the third quarter was $395.4 million, which exceeded the midpoint of our guidance range. Net revenue was down 11.5% sequentially and down 5.7% year-on-year. GAAP gross margin for the third quarter was 42.3%. GAAP operating margin was 11.8%, and GAAP diluted net income per share was $0.35. Third quarter non-GAAP gross margin was 49.5%, which was down sequentially and year-on-year, primarily driven by lower revenue and higher supplier costs. Third quarter non-GAAP operating margin was 26.5%, which decreased sequentially and year-on-year due to lower revenue. However, non-GAAP operating margin was above the high end of our guidance range. Third quarter non-GAAP operating income was $104.9 million and adjusted EBITDA was $125.1 million. Third quarter non-GAAP operating expenses totaled $90.7 million or 22.9% of revenue. SG&A expense was $40 million. R&D expense was $50.7 million. Other income and expense was a net expense of $0.9 million on a non-GAAP basis. Third quarter non-GAAP net income was $88.9 million and non-GAAP diluted net income per share was $1.19 and was at the top of our guidance range provided on our last call. Our fully diluted share count for the third quarter was 74.5 million. Our non-GAAP tax rate remains at 14.5%. On the balance sheet, cash and short-term investments increased $542 million sequentially to $2.6 billion, primarily driven by our convertible note offering. During the third quarter, we generated $76.6 million in cash from operations and purchased 3.3 million shares or $324 million, which includes 2 million shares repurchased concurrent with the issuance of our 2028 convertible notes. As of the end of Q3, we have purchased a total of 7.8 million shares, of which 5.8 million shares were purchased for $487 million under our $1 billion share buyback program. In Q3, we also funded a $30 million loan to NeoPhotonics to support their revenue growth. This loan is consistent with the terms of our merger agreement. Turning to segment details. Third quarter optical communications segment revenue at $344.2 million decreased 13% sequentially due to the expected seasonality in industrial and consumer and continued material and component shortages in our telecom business. Optical communications segment gross margin at 49% decreased sequentially and year-on-year primarily due to lower revenue and product mix. Our third quarter laser segment revenue at $51.2 million increased 4% sequentially and 62% year-on-year. Third quarter laser's gross margin at 52.9% was approximately flat sequentially but up year-on-year due to higher volumes. Now on to our guidance for the fourth quarter of fiscal 2022, which is on a non-GAAP basis and is based on our assumptions as of today. We expect net revenue for the fourth quarter of fiscal '22 to be in the range of $405 million to $430 million. Our telecom and datacom revenue is expected to grow by approximately $50 million sequentially in Q4. As Alan indicated earlier, due to the accelerating demand, this Q4 guidance also reflects over $100 million of impact to revenue, driven by shortages of third-party components. However, we believe this demand is durable due to customers being at the initial stages of their network upgrades. Based on this, we project fourth quarter operating margin to be in the range of 26.5% to 28% and diluted net income per share to be in the range of $1.25 to $1.40. The midpoints of these guidance ranges reflect our expectation of record revenue, operating margin and diluted net income per share for a fourth quarter. Our non-GAAP EPS guidance for the fourth quarter is based on a non-GAAP annual effective tax rate of 14.5%. These projections assume an approximate share count of 72.5 million and interest and other income and expense that is a net expense of approximately $1 million. With that, I'll turn the call back to Kathy to start the Q&A session. Kathy?

Thank you, Wajid. [Operator Instructions] Operator, let's begin the Q&A session.

[Operator Instructions] And our first question comes from Simon Leopold from Raymond James.

First, just a very quick clarification. On the value you've highlighted the effect of the shortages, $65 million in the reported quarter and $100 million in the coming quarter. Are those cumulative numbers or specific to the quarter? And then my question is, you had previously moved operations out of China and reduced your exposure to China. I'm sure you've got some sourcing from China. And so I'm looking for a little bit of insight in terms of whether you have and if you can quantify your risk from the China lockdowns. And on this topic, have you gained share from competitors because you have less direct China exposure?

Thanks, Simon. So specific to the quarter, the $65 million and $100 million, it's basically, as we enter the quarter, what do we see as a shortfall based on customer demand and what we're able to supply. So I guess you'd say it's specific to the quarter, meaning while we're growing telecom and datacom revenue by $50 million from Q3 to Q4, the gap between our ability to supply and the customer demand has moved from $65 million at the beginning of Q3 to $100 million in Q4. So I hope that answers your question. Meaning as we look at today, we're not able to satisfy $100 million of customer demand, some of that which rolled over from last quarter, of course. Does that answer your question, Simon?

Yes. So it's basically $65 million plus an additional $35 million. So it's building because you hadn't met the demand in March quarter, and you can't meet an incremental $35 million?

Yes. I mean you could look at that -- yes, at the same time, we're eating away at some of it given that we are growing our telecom and datacom revenue by approximately $50 million from Q3 to Q4. So what we're trying to say is while we're able to grow that business by 20% quarter-on-quarter, the demand is growing even faster than that from Q3 to Q4. As for your China sourcing impact in -- on our business, we do have a factory in China still, and it was shut down during the lockdowns in the end of March for 13 days. So that did impact slightly our Q3 revenue and more importantly impacted our ability to supply into our other factories for Q4. And that's factored into our guidance in Q4. On a component standpoint, we have been working diligently over the last couple of years to sole -- eliminate sole sourcing and get the ability to have dual sources and assurance of supply in situations like we're having today. So as we look forward, minimal impact to -- from the lockdowns on suppliers impacted by shutdowns. And as far as share gains are concerned, it's really hard to say because I think if we were able to satisfy the demand, yes, we'd be gaining a lot of share. It's hard for me to tell without looking at what our competitors are announcing to know if we're gaining share. But I'd say we probably are.

Our next question comes from Samik Chatterjee from JPMorgan.

I guess for my first one, I see you decided to exclude the $5.8 million of expense related to buying broker parts in the quarter from your non-GAAP numbers, if I'm reading it correctly, which does suggest that you think it's a bit more temporary and you talked about sort of improving IT supply. So I just wanted to see -- I mean, if I'm interpreting it right, are you thinking it's a bit more of a temporary sort of even for you where you're going -- buying from brokers at higher prices? And what's maybe embedded in your fourth quarter guide relative to broker purchases? Or is that really moderating very quickly. That's why you're treating it as temporary? And then I have a follow-up.

Yes, I'll start off with that one. Yes. So on some very specific products, we are taking a look at some exception buys for components in order to meet our customer demand that we're seeing out there. We're seeing normal price increases on components for kind of what I would call regular standard products. We are including that in our non-GAAP numbers because we're seeing that not being temporary, to use your words. But yes, where we are seeing some temporary spot buy opportunities that help us meet customer demand for specific products. We are pulling that out to show, hey, this is what we think is temporarily impacting us. We do think that will continue to impact us in Q4. I don't have a specific number because we are continuing some purchases in the month of April and the month of May. But yes, we're continuing that in our fiscal Q4. And like Alan mentioned earlier, we're expecting this -- some of these supply shortages to continue to impact us throughout the calendar year and where we see specific opportunities to go and meet customer demand. And the opportunities there from a spot buy standpoint, we will continue to do that.

For my follow-up, you guided to a $50 million increase quarter-over-quarter in telecom and datacom revenue. And Alan, you talked about the EML capacity increase. Just wanted to see if you can split that out a bit further of how much of the $50 million increase is coming from datacom. And is this sort of step 1 in terms of EML capacity and use -- maybe we get the full sort of benefit of this in the next quarter. Just wanted to think if this is the new run rate in terms of revenue for EML in 4Q? Or is there another sort of leg up in terms of capacity?

Yes. Well, of the $50 million, maybe perhaps I should say, on a percentage growth basis, the EMLs are growing faster than the telecom business. And so if you look at it as a whole, the $50 million represents about 20% of that business in Q3 going into Q4. So EMLs are growing faster than 20% and telecom growing slower than 20% to get to that $50 million. We do have additional EML capacity coming online. This is pretty much the implementation of what we've been talking about over the last 1.5 years, coming online really in full course in Q4. We do have more coming online that will impact really calendar '23. And so we're -- we'll get incremental improvements of capacity through efficiencies and productivity, but this is pretty much the incremental capacity that we've added that we've been talking about for the last 18 months.

Our next question comes from Alex Henderson from Needham.

Great. Just on that last question, you said incremental capacity in Q4. Do you mean fiscal Q4 or calendar Q4, just to clarify.

For the EMLs, Alex?

Yes. You were talking about EML capacity coming on additionally in Q4, but it wasn't clear whether you meant yearly or calendar.

I meant fiscal. So we saw some of the improvement coming in Q3 as a result of that capacity coming online. And in Q4, this quarter, we have a full quarter of that capacity, and that's why we're seeing a very large step-up in our ability to ramp EMLs in the June quarter.

Is there additional capacity coming on in the back half of the calendar year?

I would say that capacity is being installed in the back half of the calendar year that will impact our ability to grow EML capacity in calendar '23. But between now and really calendar '23, it will all come through yield improvements, productivity improvements and things like that. It will be more incremental as opposed to the step function that we saw in our fiscal Q3 and now in our fiscal Q4.

I see. My question that I wanted to address as opposed to tailing on to the last one was you talked about double-digit growth in '23 and beyond -- fiscal year '23 and beyond. Can you give us some -- a little bit more granularity on the assumptions embedded in that, is you talked about the 3D sensing business as kind of flat -- up 5%, down 5%, kind of flattish in previous conversations. Is that the assumption in the double-digit growth in '23 and beyond embedded in those -- in that guidance?

Alex, this is Chris. Let me maybe take a crack at that. I would first reference folks listening, too. We detailed this in our presentation at OFC, where we highlighted our assumptions over the next 3 years. And as you highlighted, the industrial and consumer minus 5% to plus 5%. And -- but telecom and datacom being up into the double digits and same with lasers. I would say, nearer term in fiscal '23, the reason for the minus 5% to plus 5% on the industrial and consumer is we do expect that there is the possibility of share normalization. We've had a very outsized share in 3D sensing. That share normalization being offset by new applications in the automotive, industrial, extended reality markets, if you will. But nearer term, that's an evolving market. So we'll probably see more than the minus 5% in fiscal '23 by industrial and consumer. But reciprocally, the telecom and datacom lasers will probably be at the high end or above the estimated CAGRs that we highlighted in that OFC presentation. So current situation in 3D sensing or industrial and consumer is well incorporated in our assumption of fiscal '23 revenue growth.

As a follow-up question, there was no mention of ROADMs so far in the call, at least not that I heard. Could you talk a little bit about the situation in the ROADM market relative to how much it's being impacted by supply constraints, what the underlying demand feels like and your ability to sell and install those products in customer systems.

Yes. Good question, Alex. And we're not forgetting ROADMs, ROADMs are critically important to us and to our customers and to their customers. I'd say that ROADMs and ROADM line cards have been impacted significantly by the IC shortages and where we've seen extremely strong demand. So if you go back and look at our pump shipments year-on-year up 50-some percent, ROADMs are the other direction. And that's really because of constraints, not because of demand. So demand is extremely strong. And as the IC shortages improve and we're seeing it in Q4, ROADMs will increase as a result of that. So a lot of pent-up demand for ROADMs. That's why in Wajid's prepared remarks, we think it's durable because in a lot of cases, we're sole-sourced in these advanced ROADMs and customers critically rely on our technology and products for these next generation of networks. So ROADMs are very, very strong. I wish I had more ICs because I'd be able to ship a whole lot more in Q4 and beyond.

So if the ROADM accelerates at some future period, does that then result in a reduction in the growth in pumps because you're sucking the pump volume into your amps into the ROADMs?

I think that kind of goes back to history, yes. We internally supply into our ROADMs. So there could be some reduction in pump revenue, but we've added a lot of capacity over the past several years in pumps. So I'm not sure that will be that significant of an impact to revenue.

And we're continuing to add pump capacity as we speak. So as we defer more of those pumps to our internal consumption, it probably ties pretty well with our added capacity that we're adding over the next few quarters.

Our next question comes from George Notter from Jefferies.

Maybe just extending the conversation on ROADMs. I think you guys were working to qualify alternate suppliers on componentry into your ROADM products. Can you tell us where you are in that qualification process and when that might help alleviate some of your supply constraints?

Yes. Well, I mean, across the board, we've been looking at alternative sources that are plug compatible. And those certainly have gone a lot smoother than ones that require a Board respin. And so those Board respins have been challenging, as you can imagine, but we're confident that they're going to continue to progress and alleviate some of these concerns on some of some of our component shortages. But I'd say that for more of the common parts, we've made more progress than on some of the complex parts that require both a PCB spend as well as firmware changes, and that's just taking more time. But we expect that to really come to closure over the next few months so that we'll see some ability to get these complex ICs in early calendar -- early fiscal '23.

Got it. And then the other one I had was on DMLs. I think if I go back over the last few quarters, you guys were working off some inventory in China. I'd love to get an update there.

Yes. Thanks, George. Good question. We have seen some of the DMLs that we -- if you recall, over a year ago or so, we reversed some revenue, deferred some revenue. That has been slowly starting to be recognized as those DMLs are being consumed by our customers and their customers. Still some inventory to go, but I'd say it's encouraging that there is movement. So I guess you can translate that into 5G base stations being deployed again in China, not nearly to the pace that they were 2 years ago. But certainly a signal that they are being deployed, which I think is a good sign for the future.

Our next question comes from Meta Marshall from Morgan Stanley.

A couple of questions for me. Just one, you kind of noted progress on the wireless fronthaul opportunity. I just wanted to get a sense for how large you think that opportunity could be as it ramps kind of next year? And then second, just as you look to size kind of the $100 million difference between kind of supply and demand, just how are you guys thinking about how much is forward ordering versus kind of how much is equipment that they would have actually wanted in that quarter?

Well, I'll answer the second question first and then ask Chris to tackle the fronthaul and the 10G and 25G tunable stuff. I'd say that we have long lead time purchase orders from customers for requests outside of the existing quarter. We don't count that into $100 million. This is for orders that customers have said they want in either this quarter or prior quarters. And so from that perspective, I think the demand is real. Now if we got all the ICs that we could ever imagine tomorrow when our customers want them all immediately, probably not because they're reliant on other supplies and other suppliers to satisfy those network build-outs. So I think it's real. I think it's durable, and I think it will bleed off over time as the worldwide semiconductor situation gets better, but we don't see that being alleviated until calendar 2023. Chris, do you want to take the wireless fronthaul question?

Sure. And I'd also like to tag on to the $100 million piece. I think there's something important to highlight, right, that it's not $100 million a little bit across every product line, product lines that are not constrained or up tens of percent year-over-year, whereas as Alan highlighted earlier on the question on ROADMs, ROADMs are the other direction. But as they're integral to the network and systems, we expect they should be running at comparable rates at the other product lines that are not constrained at the present time. And lo and behold, the $100 million kind of reflects that when we -- if you were to sort of hypothetically add that back to existing or our guidance, then all of a sudden -- all the product lines are up in similar tens of percent year-over-year kind of basis. Turning back to the wireless fronthaul. We participate in 2 ways, specifically in wireless fronthaul today. One is, as we've talked about in prior calls, supplying the DML lasers or 10 gig going to 25 gig lasers in our -- what we would call datacom, even though it's a telecom application. But very analogous to the kind of lasers we supply into hyperscale data centers. That business has been stressed as we talked about earlier. But it has run at several tens of millions of dollars per quarter, and we believe that over time, it can get back to that. So several tens of millions of dollars per quarter opportunity. The other way we play is, as Alan alluded to, 10 gig and 25 gig tunable lasers as WDM approaches start to penetrate wireless fronthaul as well as in the cable MSO fiber deep architectures. Collectively, those -- we expect very early stages but are starting to ramp up to be multi-hundred million dollar a year market opportunity. So a very exciting leg to our story in that over the past a number of years, our telecom transmission has been reserved, if you will, to metro, longhaul coherent applications, and we haven't participated more in access or edge applications where the volumes are quite high. And now with the adoption of fiber deep and WDM architectures in 5G fronthaul, a very large market opened up that's very unique to our kind of tunable laser technology.

Our next question comes from Tom O'Malley.

My first one is just on the current reported quarter. If you look at telecom and datacom, it's down about 9% sequentially. You gave some breadcrumbs on the pump lasers being better, ROADMs having some component constraints. But could you just give us a little more color on the split between datacom and telecom as you look from the December to March quarter and how you get to that minus 9% overall?

Well, I don't think we want to get into specifics on exact revenue levels, but I think what you can assume, as Alan highlighted, we've been in Datacom capacity constrained and got a little bit of extra capacity in the March quarter, so you can assume datacom was relatively flattish quarter-over-quarter and that any of the decline was driven by COVID-19 surge driven supply constraints somewhat unique in this case due to the surge in December and January time frame hitting the telecom almost entirely.

Helpful. And then my other one is just on the datacom side. So you're giving EML stats year-over-year now, and it looks like those are growing quite nicely. Could you just help level set how big EMLs are part of that datacom business now? You just talked about how EMLs were running tens of millions of dollars at the peak, but they're lower than that. Any kind of help with the split out there just so we have some feel for how big that business is getting.

I think the best way to say is it is in the range of 75% -- 70%, 75% of our total datacom business at present, EMLs that is.

Our next question comes from Rod Hall from Goldman Sachs.

I wanted to come back to the comment that I think you said industrial and consumer will be down more than 5% in fiscal '23. I kind of wanted to dive into that a little bit and understand at least from a qualitative point of view, do you think that is mainly due to the end market and kind of normalization of the end market? Is it share related? Is it content related? Maybe could you dig into that a little bit? And then I have a follow-up.

Yes, Rod, I think this is our expectations with regard to share normalization or equalization for the past 5.5 or over 5 years, we've had a very, very large share of our leading-edge customer. And we expect that at some point in time, share will normalize. And that's why we're setting an expectation that it could be down more than 5% from fiscal '22 to '23. Not anything to do with the market or device numbers or growth at our end customer. This is really more as we look at share specifically.

Alan, you guys on content, you're not really expecting much change?

Sorry about that. Well, I was asking if that answers your first...

I just -- Alan, I wanted to follow that up and just say you're not expecting much of a change on content. It sounds like, I mean, just normal kind of maybe declines in pricing, but otherwise not much of a content change.

I think it could in the coming years. We're working on many different lasers of different types and until they go into a product. It's hard to say how much, what models they go into and things like that. We're not counting on in that guidance any Android business, and that could be an upside. But I'd say we've been working with Android for years, and it had its day when we were shipping meaningful revenue to Huawei on their handsets. But I'd say as far as content at our leading customer, there may be some increased content, but some of that will be offset by, as you say, price reductions over time.

Okay. And then my follow-up was just on the new applications, XR and automotive. I wondered if -- I thought on automotive, your commentary suggested maybe external sensing is coming a little closer. You've got a little more visibility, but I don't know if I misinterpreted your comments there. But I wonder, could you just talk a little bit about XR when roughly timing on that. Is it early part of the calendar year? And then automotive, is that right? Or are you getting a little bit more visibility on application for external sensing?

Rod, this is Chris. I would say that what we're highlighting is a lot of design win activity, if you will, both in the automotive and the extended reality applications. So very strong customer traction, design wins racking up with multiple customers and more importantly, meaningful customers in those spaces. From a timing standpoint, I think as we've been clear, automotive is a very long-term market. So it will start small and increment upwards steadily. Extended reality has probably more opportunity in nearer term, but we still think -- I think we said this in the prepared remarks, that we expect customer products to be launched probably calendar '23 time frame would be our expectation. But they don't -- customers don't share with us their exact timing around product launches and what they're going to be per se. As you can imagine, these customers are very secretive in this space about what they're doing. But I would say as we know that certain numbers have been shipped this year and -- or sorry, in calendar '21 and expectations of growth into '22 and more meaningful growth in '23.

Our next question comes from Christopher Rolland from Susquehanna.

My question is around comms. I guess there's a couple of parts to the equation here, supply and demand for that $100 million supply shortfall, I was wondering if you could kind of break that up between ROADMs, TC and DC. And conversely, from the demand picture, which of those 3 are you seeing kind of the biggest surprise in demand versus your expectation?

Yes. I would say the shortfall is primarily telecom. So if you say it's all on telecom or 95% of it is in telecom, the split between ROADMs and transmission products is pretty close. I'd say ROADMs have been impacted most in the past. And as we -- and the March quarter was impacted significantly and where we're seeing some relief in the June quarter. So I think as we get the semiconductors and we're working diligently, and I wanted to say thank you to the supply chain team. They've just done a fantastic job. But I'd say ROADMs are the biggest hit as an individual product line, but then it goes into amplifiers, blades and transmission products as well.

Great. And then on the 3D sensing side, just to follow up there and get my head around it. If I'm doing the numbers right, we're down almost 30% year-on-year for the June quarter. I know this is a small base in June. But as we look out into next year, could we be down -- how would you put the odds at being down significantly more than 5% year-on-year? Could there be a situation in which we're down 20% or something like that? Or in your best view, are we just looking to fall kind of much more modestly here?

I would say first that we didn't give a specific 3D sensing number for the June quarter guide. I would say that I guess it was a year ago or so, we said that the overall markets due to chip size reductions would be down 20%, 25%, something like that. And what we've seen is, in fact, exactly that. And hence, that's why you're seeing the year-over-year decline in 3D sensing, for example, in the June quarter. What the market or what our -- or more specifically what our revenue is going to be in our next fiscal year, we're not providing guidance much more beyond what we've said. But certainly, the color we're providing is that minus 5% to plus 5% 3-year CAGR, you could imagine it being a little more V or U-shaped, if you were to look at it on an individual '23, '24, '25 basis as Alan alluded to, share normalization. If it happens, it's happening, at least from a model standpoint, assume next year and then new applications come in and offset that in '24 and '25 time frame. So I don't want to get into specifics of how much of a decline, but certainly more than the 5%, but on a 3-year basis being somewhere flattish to slightly up, slightly down in aggregate.

Our next question comes from Michael Genovese from Rosenblatt Securities.

A lot of good questions have been asked. So I'll just ask a quick one question. You've got the $50 million sequential increase in telecom and datacom revenues this quarter. And just looking at your operating margin and EPS guidance, it seems like that actually is coming at a pretty decent gross margin. So can you just talk about how you did that in this environment? Are you avoiding expedites on this incremental amount? And talk about how you're -- have solid gross margins even in the supply chain, please?

Yes. So I'll start off. So I think you've seen for a number of quarters, we've been talking about our overall company margin model. And for 2 years in a row on a fiscal year basis, we've been fairly consistent in achieving or exceeding our model both from a gross margin standpoint and from an operating margin standpoint. And actually, if you take a look at the midpoint of our fiscal Q4, we will actually improve gross margins year-over-year. And one of the biggest drivers for us is really operating leverage. And so when our lasers business is growing as it has year-over-year and sequentially, that's covering a lot of manufacturing overhead for us, and that's helping us drop more to the bottom line. In addition to that, where we're seeing a lot of growth, both in our transmission and our telecom business, the margins for those products are quite good relative to historically what we've seen. Our 10G tunable products are doing quite well in transmission, and we're seeing the benefit of that fall to the bottom line. Our operating expenses have stayed fairly consistent as we've continued to invest in R&D. And so that's been helping us, too. Yes, we're expecting some expedite fees to happen. When we have one component left and the only way to get the product to the end customer is to do a spot buy, we'll continue to do that. And we think it will help our overall customer satisfaction. And our main goal is to meet or exceed our customer expectations, and that's how we're running the company, and that's how we're driving our financial model as well. So yes, I mean it's really just the leverage we're seeing on our telecom and our lasers business. As well as our datacom business improving quarter-over-quarter. We posted some EML numbers, as Chris said earlier, and that's really helping us as well. So it's really those things.

Our next question comes from Ananda Baruah from Loop Capital.

Congrats on the strong results. Just 2 quick ones, if I could. I guess the first is with demand continues to accelerate, how -- any context you can share with us about how this impacts second half of calendar year seasonality? And then just a clarification. Do you guys -- is it accelerating at a faster pace today than you thought it might have been 90 days ago? 90 days ago, you guys were quite positive as well. And then I have a quick follow-up.

Yes. Thanks for the question. I would say it's accelerating as we expected, given we've been signaling the transition to 400, 600, 800G coming and the new network architectures focused on our leading edge ROADM technology. So I'd say the demand is coming into play as expected. The supply, on the other hand, is tougher. And so I'd say we're not lacking demand. And I don't think we're going to be lacking demand for several quarters. It's a matter of what does the semiconductor situation look like in the second half of the calendar year, and that will really dictate what the telecom revenue looks like in our fiscal Q1 and fiscal Q2. So we're working diligently to continue to try to improve that. But the situation is not totally solved yet. And so that's a challenge for Q1 and Q2. And we're only going to guide one quarter at a time because the visibility is a little bit top beyond the short-term horizon and what's going to happen with semiconductors.

That's super helpful. And then Wajid, just a quick follow-up on the previous operating margin question. Is that to say that we should expect good leverage going forward? You mentioned you're putting up good leverage in the June quarter, you have been in recent quarters. So should that be something we should expect as well going forward?

Yes. No, certainly, we're expecting to have continued leverage as we move into fiscal year '23. I think the one point of caution is, is we are seeing increased component costs that are impacting us. And as that continues to happen in fiscal year '22 with the supply shortages that Alan talked about, we're going to continue with our target model, but we're not going to communicate any changes to it. In addition to that, just as a reminder, our expectation on NeoPhotonics is that, that closes in the second half of our calendar year. And as that comes on board, we're going to have a lot of work to do from some of the synergies commitments that we've made to all of you. And that's going to have some transitional impact on our overall gross margin model as well. So I just want to make sure that we all keep that in front of us.

Thanks, Ananda. Victoria, I'd like to squeeze in one more question, if we could.

Our final question comes from Dave Kang from B. Riley.

I have 2 questions. First one is regarding China. What's the situation with Shenzhen factory? How much revenue does it generate? And is it running 100% or partially? And my follow-up is regarding the supply chain impact, it was $65 million now. This quarter, it will be $100 million. When do you think it will inflect?

Okay. So China, Shenzhen is back running full force, and it's hard to say how much revenue it generates because it makes components to go into other factories. And it primarily produces subassemblies and components for our telecom transmission business. So a large percentage of our telecom transmission business goes through that factory. And as I said earlier, we were shut down for 13 days in March. That impacted our revenue in the Q3 numbers as well as the components that flowed into other factories in Q4, and that's factored into our guidance. The -- when does the supply chain -- the question on the -- your second question on supply chain, when does it inflect. Is that the...

Yes. So it's been going up. When do you think it will start to come down?

It's hard to say. If you could tell me when the semiconductor shortages could end, I can tell you the answer to that. But I'd say, we believe that the demand for our telecom products and our leading edge ROADMs and high-speed transmission products is extremely robust, and it's going to continue to be robust. And that's why it gives us confidence in the double-digit growth in fiscal '23 over '22 and beyond. So I'd say demand is going to continue to be strong. We're going to have component challenges into calendar '23. And when does that go away is a little bit of a hard question to answer. But I'd say that demand is, again, not our problem in the short term, and I don't think it's our problem to calendar through fiscal '23.

Now I'd like to just pass the call back over to Alan for some closing remarks.

Great. Thank you, Kathy. I want to thank our customers and suppliers for their partnership in these challenging times. I would also like to leave you with a few thoughts as we wrap up this call. I am very excited about the accelerating customer demand in telecom, datacom and lasers and the work our team continues to do to improve our supply of third-party ICs and to increase our manufacturing capacity to support this ongoing demand strength. Additionally, the opportunities we have in automotive, extended reality and industrial applications, which increasingly leverage 3D sensing and LiDAR capabilities at Lumentum are emerging and will drive diversification and growth. Our market-leading products and technologies positions us well for these opportunities ahead. With that, I would like to thank everyone for attending, and we look forward to talking with you again during upcoming investor events, which you will find posted on our website. Thank you for attending.