Lumentum Holdings Inc (LITE) 2017 Q4 法說會逐字稿

Good day. My name is Jack, and I'll be your conference operator today. At this time, I would like to welcome everyone to Lumentum's Fiscal Fourth Quarter and Full Year Earnings Call. (Operator Instructions) Thank you.

Chris Coldren, Vice President, Strategy and Corporate Development, you may begin your conference.

Thank you, Jack. Welcome to Lumentum's Fourth Quarter and Full Year Fiscal 2017 Earnings Call. This is Chris Coldren, Vice President of Strategy and Corporate Development. Joining me on today's call are Alan Lowe, President and Chief Executive Officer; and Aaron Tachibana, Chief Financial Officer.

This call will include forward-looking statements, including statements regarding the markets in which we operate, trends and expectations for products and technology, purchasing trends, Lumentum's expected financial performance, expenses and position in our markets. These statements are subject to risks and uncertainties that could cause actual results to differ materially from our current expectations. We encourage you to review our most recent filings with the SEC, particularly the risk factors described in our 10-Q filing for our fiscal third quarter ended April 1, 2017. The forward-looking statements we provide during this call, including projections for future performance, are based on our 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 are non-GAAP. Non-GAAP financials should not be considered as a substitute for or superior to financials prepared in accordance with GAAP.

Our press release with our fourth quarter and full year fiscal 2017 results is available on our website, www.lumentum.com, under the Investors section and includes additional details about our non-GAAP financial measures and a reconciliation between our GAAP and non-GAAP results. Our website also has our latest SEC filings, which we encourage you to review, and supplementary slides relating to today's earnings release. Finally, a recording of today's call will be available by 11 a.m. Pacific Time this morning on our website.

Now I would like to turn the call over to Alan for his comments and business highlights.

Thank you, Chris. Good morning, everyone. This is an exciting time at Lumentum. We just concluded our second fiscal year as a stand-alone public company and exceeded the $1 billion revenue mark for the first time. Relative to the prior year, we grew our revenue by 11% and net earnings by more than 50%. We capitalized on our investments in 100G Datacom with FY '17 Datacom revenues being up 125% from FY '16 levels. Also, we put ourselves into the leadership position for the ramp of 3D sensing with more than $200 million of booking -- of orders from customers since April. Our strong results came from both the introduction and ramp of our differentiated new products and our focus on operating leverage. I'm very proud of the Lumentum team for these achievements and thank them for their continued hard work.

Since our last call, we've made excellent progress on our strategic goals. Our revenue from 100G QSFP28 transceivers nearly doubled sequentially in the fourth quarter. 100G products represented approximately 80% of our overall Datacom revenue. Growth in our 100G more than offset declines in lower speed transceivers. Additionally, during the fourth quarter, we shipped approximately $5 million of 3D sensing revenue. On our last call, we indicated that we had several million dollars of 3D sensing orders. Since then, orders have been very strong for multiple applications. We have a lot of work to do to translate these orders and to ship revenue where we expect 3D sensing to be a material contributor to our business in the coming quarters as well as the coming years.

Our near-term focus in 3D sensing is to ramp production to higher levels than anticipated just 3 months ago. We expect monthly volumes to increase throughout the end of the calendar year to meet our strong customer demand. Like many in the optical communication industry, we continue to have some near-term headwinds. Demand from China has yet to materially improve. Our Chinese customers continue to consume inventory already purchased. It is too early to know if this inventory correction will be resolved in the September quarter, or if it will continue into the December quarter. As inventories decline to targeted levels at our Chinese customers, demand should return to significantly higher levels than what we have seen over the past 2 quarters. The timing of this recovery is uncertain.

North America telecom demand on average has been strong and growing over the past 2 years. North American demand has also been relatively lumpy on a quarter-to-quarter basis and was down sequentially in the fourth quarter. However, we expect the overall growth trend in North America to continue primarily driven by metro deployments. We believe our North American customers have a strong business outlook and our quarter-to-quarter lumpiness is being driven by the timing of new deployments and customer inventory management.

Fourth quarter revenues were down 13% to $222.7 million. As highlighted earlier, we had strong QSFP28 demand, which was the primary driver of the 28% quarter-on-quarter Datacom increase to $50 million. 100G Datacom grew more than 50% sequentially and CFP2 sales, once again, contributed to growth. Looking to the first quarter of fiscal 2018, we expect telecom revenue to soften further as customers continue to burn off inventory. Despite, these near-term telecom headwinds it is a very exciting time at Lumentum. Our photonics technologies are increasingly critical enablers of leading-edge communications, industrial and consumer applications. Demand continues to grow for bandwidth across the world’s data centers and communication networks. New submarine cables are being lit up and built across the world. In addition to traditional communication service providers, cloud operators are increasingly driving new deployments of optical networks and submarine cables. Hyperscale data centers are transitioning to 100G.

We continue to make excellent progress with our ROADMs. In addition to capturing a large share of initial -- the initial network deployment this year in China, we have shipped our first of several new advanced ROADMs specifically designed for China's next generation networks. This positions us extremely well as the ROADM Supplier of Choice, both in the near term as well as in the longer term in China. We expect China ROADM growth will be a meaningful growth driver for our Lumentum business in the coming years.

We are working with all of our customers worldwide on newer platforms that incorporate higher performance, smaller and more cost-effective ROADM solutions. We believe our new product pipeline will further our leadership position over time. Manufacturers around the world are increasingly using advanced laser-based techniques to increase productivity and precision and enable new processes. Leaders in next-generation consumer electronics, virtual and augmented reality as well as the automotive industry, are looking to laser-based 3D sensing to enhance capabilities and enable new applications.

Our investments in new products and technology position us well for all of these future trends. At Lumentum, we are focused on accelerating the speed and scale of cloud, networking, advanced manufacturing and next-generation 3D sensing applications with our photonics technologies. Our strategy to invest in growing markets, develop the best products and technology and foster the closest relationships with market-leading customers is succeeding and makes our future bright at Lumentum.

I will now hand it over to Aaron for more details on our financial results and our guidance for the first quarter of fiscal 2018.

Thank you, Alan. We recently concluded fiscal year 2017 and achieve new record highs for Lumentum. Net revenue for the full fiscal year surpassed $1 billion and was a new record high as we grew 10.9% compared with last year. GAAP gross margin was 31.8% and GAAP operating margin was 4.8% for the year. We achieved non-GAAP gross margin of 34.7%, increasing 170 basis points from last year, as we are shifting the mixture of products to those with a richer margin profile. Non-GAAP operating margin was 12.4%, a new record high for Lumentum and increased 320 basis points over last year driven by the gross margin expansion and favorable operating leverage from higher volume.

Now for the fourth quarter result. Net revenue for the fourth quarter was $222.7 million, which decreased 12.9% sequentially and 7.9% compared with the same period last year due to softer telecom demand. GAAP gross margin for the fourth quarter was 30.2%, GAAP operating margin was 1.8% and GAAP diluted net loss per share was $0.90. The GAAP diluted net loss per share includes the impact of a $1.06 loss per share from a noncash unrealized loss on the derivative liabilities for both the Series A preferred stock and convertible note, the noncash effective interest on the convertible note as well as a noncash write-down of deferred tax asset.

Our fourth quarter non-GAAP gross margin was 32.9%, which decreased 150 basis points sequentially and 120 basis points compared with the same period last year. The sequential decrease was driven by a 200 basis point decline in Optical Communications gross margin partially offset by an 80 basis point increase in Commercial Lasers gross margin. Non-GAAP operating margin for the fourth quarter was 9.2% and declined 340 basis points sequentially. The decline in operating margin was primarily due to the sequential volume decrease causing unfavorable operating leverage quarter-to- quarter. Non-GAAP earnings per share was $0.39 based on a fully diluted share count of 63.8 million and included $900,000 of interest income and a $3.7 million tax benefit.

Now for some additional detail. Optical Communications revenue was $186.8 million, a decrease of 14% sequentially driven by a $44.7 million or 27% decrease in telecom revenue, partially offset by an $11.1 million or 28% increase in Datacom and a $4.3 million or 35% increase in industrial and consumer revenue. During the quarter, our 100G transceiver revenue increased more than 50% and our revenue from QSFP28 transceivers nearly doubled sequentially. Optical Communications gross margin at 31.1% decreased 200 basis points sequentially, mainly due to lower telecom revenue.

Commercial Lasers revenue was $35.9 million, a decrease of 10% sequentially which was driven by a decline in fiber laser revenue. Our primary customer is transitioning to our newer, higher-performance Gen 3 design, and while we have been ramping the Gen 3 fiber laser for a few months, we have been capacity constrained during the ramp due to unique equipment and materials compared with our older Gen 2 fiber laser. Fourth quarter Commercial Lasers gross margin was 42.1%, and increased 80 basis points sequentially. Operating expenses totaled $52.8 million or 23.7% of revenue compared with last quarter at $55.7 million or 21.8% of revenue. R&D expense was $32.3 million and SG&A expense $20.5 million.

Our balance sheet remained strong exiting the fourth quarter with cash and short-term investments of $555.3 million. During the fourth quarter, our net inventory increased by $28 million sequentially. There were a few factors driving the increase. First, we completed the closure of a fab in Connecticut late in June and we transferred production to our fab in San Jose, California. In order to support the move, we produced some transition inventory. Secondly, with the sequential revenue decline, we were unable to slow down the supply chain. Already having materials and capacity in place, we built some product that will be needed over the coming months. And thirdly, we accelerated our 3D sensing ramp immediately after receiving customer approval for production readiness early in the fourth quarter. Capital equipment additions were approximately $41 million in the fourth quarter and our spending remained elevated due to our investments in both facility improvement and equipment to bring up our manufacturing facility in Thailand, which was purchased back in March.

Now on to our guidance for the first quarter of fiscal 2018, noting again that all projections are on a non-GAAP basis. 3D sensing revenue will increase materially in Q1 and is only limited by our ability to ramp faster. However, telecom and Commercial Lasers revenue is expected to decline sequentially. We project net revenue for the first quarter to be in the range of $245 million to $265 million, with operating margin in the range of 13% to 15% and earnings per share to be in the range of $0.50 to $0.60.

Now I will turn the call back over to Chris to begin the Q&A session.

Thank you, Aaron. (Operator Instructions) Jack, let's begin the question-and-answer session.

(Operator Instructions) Your first question comes from the line of Simon Leopold with Raymond James.

I wanted to see if we could get a little bit more understanding on the trajectory of the 3D sensing. So you indicated up materially in September, but listening to others in the supply chain, we have the impression that there should also be a multiplicative effect in the December quarter, and then just in terms of rounding out the 3D comments, you talked about $200 million in orders, what time period does that represent, is that 1 years' worth of orders or is that over a longer period? And I have a follow-up.

Thanks, Simon. Yes, the trajectory on 3D sensing, we're focused very hard on ramping as fast as we possibly can, and that's what we put in our guidance from -- with respect to our expectations for this quarter. December quarter will be up materially from the September quarter although we're not going to give guidance. The orders that we've received to date, and these are from multiple customers, are primarily for deliveries or requested deliveries in this calendar quarter -- this calendar year, sorry.

So in calendar '17?

Yes.

Wow, okay. And just my follow-up is an easy one. The QSFP28 strength, there are many, many flavors of those kind of products, CWDM or LR4. Can you give us a little bit more detail of where the strength is coming from in terms of flavors of QSFP28?

Yes, Simon, in terms of the fourth quarter, basically half was LR4, half CWDM for QFSP28 flavor.

Your next question comes from the line of Patrick Newton with Stifel.

I guess to put a finer point on 3D sensing, given the $200 million in orders, given that you hope to be recognizing them in this calendar year, can you just level set us on how much 3D sensing revenue has been recognized through the June quarter, and what you have embedded explicitly in the September quarter guidance?

Yes, we said in the prepared remarks we had $5 million in the June quarter. And that was from multiple customers, so keep in mind we had legacy customers as well as new customers in that $5 million. Embedded in the September quarter, as we said, we expect telecom and lasers to be down slightly from the current levels and Datacom flat to up slightly. So you can do the math on our expectations with respect to the 3D sensing in the September quarter from there.

Great. And then I guess, just on the gross margin side, Aaron you gave some reasons why the gross margin declined sequentially given the factory move, inventory build and not able to slow down as fast, but can you comment at all on the pricing environment? And then as we look forward, given that 3D sensing will become a bigger portion of mix, can you help us understand the delta between optical and 3D sensing gross margin and how to think about that over the next several quarters?

Sure. Thanks, Patrick. In terms of gross margin for the fourth quarter, so we were down about 150 basis points sequentially. 2/3 of that decline, so about a point was tied to the volume decline and the rest was mix related. So the margin decline had very little to do with the pricing environment and our cost reductions have been able to keep up the price erosion. So we didn't see a lot of debt on the gross margin in the fourth quarter from a price standpoint. As we go forward, your question about OpComm margins and 3D sensing, so looking at the aggregate first, in totality, our margins are going to accrete or expand as we go forward as 3D sensing does have a higher gross margin than our corporate average today. We're not going to specifically call out what the 3D sensing margins are, but they are higher than the corporate gross margin average. So Optical Communications where the 3D sensing resides today from a segment perspective, will continue to increase. When you look out 6 to 8 quarters, there’s no reason why we can't get our gross margins up into the high 30% range, now today we're in the 34% range. There's no reason why we can't expand our margins 200 basis points to 300 basis points over this next 6 to 8 quarter.

That's corporate average, correct?

That's correct, corporate average. And then I did also mention the wafer fab closure from Bloomfield, Connecticut and moving it into San Jose. As we have been mentioning on previous calls, we expect to get a 70 to 80 basis point improvement from that reduction. We'll start to see that benefit in the next couple of quarters.

Your next question comes from the line of Rod Hall with JPMorgan.

So I wanted to start off with North America, see if you guys could maybe give us a little bit more color on what's going on, and particularly, I know you called out that inventory management and timing of deployments are the 2 reasons. Can you give us some kind of a ranking of those 2 things? Do you think it's more inventory management or is it more deployment timing that's driving the weakness this quarter and next? And just any further color on what's going on in North America and then I have a follow-up.

Yes. I think the challenge we have with the North America deployments specifically metro and that's not just 1 customer or 1 carrier, is the deployments are very lumpy and our customers need to be prepared as those carriers decide to deploy. And so I would say that the primary driver for the slower telecom revenue in North America was inventory management and is inventory management going forward. But as I said, that could change overnight and we could be expedited next week as a result of decisions being made at carriers or the MSOs. So I think it's primarily inventory related.

And Alan, could you just say on that, what's your expectation from now to the end of the year? Are you kind of thinking that the way that North American deployment goes or those deployments go is roughly the same by the end of the year, or you think it’s a little bit lower? Or how do you think we end up at the end of the calendar year?

That's a really good question and one that I think is probably better answered by our customers. I will say that we are still in the first quarter of deployments in North America for metro. So I think we've got multiple years to go in building that out. The question in my mind is, how many cities get built each quarter and that's really a function of when the carriers decide to pull the trigger on capital.

And then I wanted to -- my follow-up, I just wanted to ask you on 3D sensing capacity, what are you guys doing -- it sounds like the orders have moved up quite a bit more than maybe you expected like you said 3 months ago. What are you guys doing with regards to capacity? Are you laying more capacity on or putting in place more capacity by the end of the year, what are you doing with respect to that?

Yes, I mean we are -- even 3 months ago we were adding capacity through this calendar year to meet the demands of what we thought were at the time. And as you said the expectations and demands have actually increased since our April earnings call. We've been expediting capacity increases at our foundry partners and helping them with their suppliers of capital equipment to bring that on earlier. In addition to that though, we've also been working on -- figuring out how do we accelerate the productivity improvements of the capital that we have to get more out of the existing assets. So the lead times for some of the best equipment and processing equipment are long and so a combination of trying to pull those deliveries in as well as getting more out of the existing assets at our foundry partners is what we're focused on.

Can you -- is there any way you can quantify the capacity increase relative to what you originally thought, like are you increasing capacity by the end the year 50% over what you thought you would have to deliver 3 or 4 months ago, or any kind of quantification you can give us on that?

It’s hard to say. I think back of the envelope, it's probably not 50%, but it’s 25% to 30%, would be my guess. I think if we could do more, we would -- our customers would probably take it.

Your next question comes from Alex Henderson with Needham.

The first question I have for you is on this $200 million in orders deliverable in the 2017 time frame. I mean, do you think you can actually attain that or is that something that they would have like to have, but you're going to be forced to slide some of that into the first quarter next year? And then the second question along the same lines is, as we look out to calendar '18, have you any sense of scaling of that business into the following year? I mean, is it -- would you expect that if you are doing $200 million in a half that it would be somewhere in the $300 million to $400 million for the full year '18?

Okay. First question, are we going to be able to deliver it all, I think certainly that's our customers' expectation, and again, this is multiple customers, but our expectations are assuming that we execute to our plan, is that we do satisfy those orders in this calendar year -- at the balance of the calendar year. As far as 2018 demand is concerned, I think a lot will depend on how well we execute this year and how much business we earn from our customers with respect to how much they can count on us for. So it's all about how we execute this year and then how successful our customers' products are in the market. So it's hard for me to give you a number, but I'd say that our expectations are that we're going to execute and earn our customers' trust and earn their business.

So again, is it likely that, that magnitude of full year '18 would be a reasonable way to think about it, or am I just off mark? I mean, can -- we have no basis to forecast on, so it would be very helpful if you can give us at least some sort of general indicator on it.

I mean, my expectations are that it continues to increase, especially as we add additional customers. So I don't think your number is crazy. Again -- but it has to do with how well we execute and how we build our customers' trust. But given the pipeline of additional customers that we have, my expectation is that this is a growing business from this launch here.

Can you talk at all about the pricing environment as we go from '17 to '18? Or can you talk about what your expectations are relative to the number of cameras we'd be looking at on an individual device, or any metrics along those lines to give us some qualitative sense of how the design parameters are going to change over time?

Well, pricing environment is really a function of how well we execute and how well we do in the second half of calendar '18. I think we have agreements in place that's allowed us to make commitments to our partners to buy a lot of capital equipment. And so for the next year, we are pretty well agreed upon in contract basis for pricing. As far as the number of devices per device, I'm not going to comment on that. I think again, we're working with multiple customers on different applications that have single devices as well as multiple devices. And so I really can't comment on what our customers' expectations are -- that they do with these devices.

One last question, then I'll cede the floor. Can you -- given your ramp and expectations, have you changed either your yield expectations or your market share expectations? I think you've said your yield seemed to be coming in quite a bit better than you’d initially expected and it sounds like vast majority of the market is still the right descriptor of share, are those 2 still accurate?

I think for the most part, yes, I think we're pleased with our execution on yields. We certainly could do better, but I think we're -- our expectations are being met with respect to yields at this stage of the game, and then market share I think we're in a pretty good position and have a very good market share of the business of our customers today.

Your next question comes from the line of Joseph Wolf with Barclays.

Just a quick follow-on there, if you could give us any color in terms of competition, do you see the competition right now and the related pricing as similar to yours, meaning your competitors could also deliver as much as they can produce, and that you have a yield and capacity lead right now and that will change over the course of 2018. Or are there other things that you see as being competitive advantages right now?

Well, I think our strategy to partner with high-volume industry-leading foundry partners is paying off from our perspective, and we've been working with these foundry partners for years and so it's not a simple task to do, to do this kind of ramp. I can't really comment on where our competitors are and specifically where our competitors' pricing are because I frankly don't know their pricing. I will say that the data we have where -- we were expecting that certain number of units 3 months ago and that number has gone up, healthy, we're in pretty good position.

And then just switching gears to the margin guidance, if I look at -- it seems like similar levels to the margins that you hit in the fiscal second quarter at the same revenue level, but if we look forward to the mix and the commentary on the gross margin, can you just give us some of that color on where different line items will come out given the mix that you expect and the new opportunities? Should it be very different or is there a big R&D expense we should be expecting in 2018 as you ramp up some of these opportunities?

So Joe, thanks for the question. In terms of that, as we look into fiscal year 2018, from an overall OpEx standpoint with our base business being where it's at, today the OpEx percentage of revenue is actually a little bit higher than where we were when we were doing $260 million a quarter at this base business. So in terms of OpEx, I would anticipate us being somewhere between the 21% to 22% range as we go through FY '18 and as we continue to ramp 3D sensing, the incremental spend will not grow as fast as the top line, but you can see us get down into the 21% range. As we have mentioned before, we're going to continue to invest heavily in areas where we believe markets are large and growing and, as Alan had mentioned in his prepared remarks, 3D sensing has multiple application beyond the initial consumer model here that’s taken off. And we've been a leader in this market for some 4 years or 5 years as we had been in the gaming console business many years back, and there's no reason again, why our margins won't expand and from a targeting standpoint, like I had mentioned, the gross margins should get up into the high 30% -- so 37% to 38% looking out 6 to 8 quarters is not unreasonable. And a lot of that will flow through to operating margin.

Okay, that’s helpful. Just finally, and maybe it's early, but given the current market and the inventory comment on North America for telecom, as you exit the year, do you sense that there's going to be any pricing issue on the core telecom components for 2018 -- the calendar 2018?

Yes, I mean, I think there's always pricing concerns and cost concerns that we're focused on trying to drive our cost down to meet the expectations of our customers. Assuming this environment as we get into the annual price negotiation, one could expect that we would be on the higher end of the 10% to 15% range. But that hasn't started yet, and I'm hopeful that the inventory in China will burn off and North America will start deploying more and we’ll have a more supply demand balance. I will say that for specific products like ROADMs and ROADM line cards, those prices have been agreed upon, again into 2018 already. So that in my mind is already fixed and set for most of our high-volume customers, but it's the other coherent components and other products that may be under some pressure on the higher end of that 10% to 15% range.

Your next question comes from the line of James Kisner with Jefferies.

I guess the first thing was, you've been saying that the orders are for multiple customers. I guess can you clarify, is it fair to say that the vast majority of that is -- it's really primarily one customer, or is it more diversified than we're thinking?

It's primarily one customer, but I wanted to make sure that it was clear that we are working with other customers, both legacy customers that we’ve had in the gaming and personal computing industry as well as additional customers that are going through qualification in anticipation of getting into volume in calendar 2018.

And then separately, one of your competitors has been talking about -- it's very clear they are implying that there is a ramp in high-power version of this laser, not just the low power version. Is it fair to say you're participating in that opportunity as well, or both opportunities?

Yes.

That helps. Could you also just talk more about a less interesting topic, I guess, for some folks is the Lasers business. So I mean it seems like this business has a problem like every 1 or 2 quarters and you're having trouble getting your arms around it. And given the margins it's important for earnings, you talked about having trouble getting equipment, is that something you didn't see a quarter ago? Is there something here that's not beyond your control, like what's it going to take to get that business sort of back to this $40 million level and stay there like -- what kind of trajectory you expect in that business in the next few quarters?

Yes, it's a good question, James, and one that I've been frustrated with. So I’ll say it is 100% in our control, not out of our control. We did some late design changes to optimize the product and the quality of the product and reliability of the product that then had us changing some of our suppliers and some of the equipment needed to build the product. And so I'd say that, that is well within our control. We should have anticipated it, we didn't and we're addressing that from an execution standpoint. But we do have to fix it and we do have to be able to be predictable in our execution with our customers. And today, it has been not satisfactory in my mind. So I'd say we are on track, we now have the design, equipment, the materials coming in, but we are constrained due to these changes that we should have anticipated a while back.

So it sounds like, I think revenue wise, this could get a little better in the near term but it’s going to take a few quarters to improve, is that fair?

My expectation is -- and we guided that Lasers would be flat to down a little bit this quarter, I think that's continued. The Gen 2 was in the number, in the Q4 number. It's gone now. So it's all Gen 3, so it's really a function of how fast we can ramp Gen 3. My expectations are that Lasers increases in the fourth quarter, but we're not there yet, we're not giving guidance. I would say though that the fiber laser demand in the Gen 3 is quite strong, both at our main customers as well as additional customers. So it's really in our ballpark, in our wheelhouse to execute and ramp that up as fast as we can.

Just making one quick clarification here. You said that ROADMs was an inventory issue, is it fair to say you don't think you're losing share or there's not a second source for ROADM line cards in North America, and then I'll pass after that?

I think that -- we have -- if you look at the major carriers in North America, I don't think we've lost any share there. I think the primary issue is the inventory and the timing of the deployments. I think we are in a very, very good position in China and the rest of the world is counting on our ROADMs, current generation and future generation. So I would say that we're not losing share per se. I wouldn't say that the ROADM market was down as much as our ROADMs were last quarter, but I think that's a specific customer inventory issue and not one of losing share.

Your next question comes from the line of Doug Clark with Goldman Sachs.

My first one is on the balance sheet inventory. I know you mentioned 3 particular reasons why it increased. The one in terms of -- just on shipments to customers and kind of building inventory because of excess supply and timing, how much of that $28 million sequential increase was that particular factor versus the other 2?

So that factor was probably half of the increase Doug, in terms of -- again. So we've been adding a lot of equipment capacity over the last 6 quarters as China was turned on and North America metro as well. Having parts and equipment on hand, it made sense to go and build some of this out and it’s product that we're going to be selling over the coming few months. And so we made the decision to go ahead and build it out, so roughly half of the build was tied to that.

Okay, that makes sense. And then a follow-up question on the QSFP28 revenues up about 50% sequentially. Since pricing is kind of more regular and rolling continuously, can you talk about the pricing dynamics in that industry as you and kind of the rest of the industry ramp volumes and continue to see pretty nice increases in revenues there?

Yes, I think it's really 2 separate markets that the QSFP28 addresses. The LR4 which really the client side of telecom and that kind of is stable with fewer competitors, it's a very high reliable telecom kind of performance that is needed for those kinds of products and those products are stable and very solid from a pricing standpoint. The hyperscale QSFP28, CWDM4, PSM4, those kinds of products, it's a tough business. I mean, the margins are challenged, the pricing is challenged, but we are still getting significant orders for the products and we're being told our prices are significantly higher than some competitors. So I think it’s a question in my mind is, will some of these competitors that are providing aggressive prices actually deliver or will we continue to be able to capture share with what we think is an appropriate price for the current generation product. Now that said, we've got a second-generation product coming out this quarter and then we've got another generation coming out in early calendar 2018 with very differentiated technology and cost points that I think we can address the hyperscale market even if the pricing decline comes to the point where our customers are asking it to go.

Okay, that's helpful. And then one kind of follow-up question, just on China in particular. I'm curious if -- or if you can help quantify kind of what the decline in this June quarter was and kind of if you're expecting a further sequential decline in the September quarter or if we’re just holding at a kind of stable low levels, given the inventory reduction dynamics continuing to persist?

I think it's more stable low levels. If you look at the first half of our fiscal year compared to our second half, it was down dramatically. The 100G components were down, pump lasers weren't down, so that's again, still an indicator of future growth or tight supply and not over inventory situation, but I'd say our expectation is that it stays at that low level, plus or minus a little bit in the September quarter, and then we'll see what happens with tenders and buildouts in inventory as we look into the December quarter. But I think at this point it's too early to tell whether it goes into the December quarter or we see a rebound in the December quarter.

Your next question comes from Mark Kelleher with D.A Davidson.

Just as a follow-up to that, I'm just curious as to the visibility you have into the inventory drawdown in China. Can you see that inventory, is there -- has there been an increase or decrease in the velocity of that being taken down, or are you just getting sort of indirect commentary from your suppliers?

Yes, from our customers, I'd say we have good visibility in the raw material and what they have that they haven't built yet. We have less visibility into what their finished goods are and I think that's the challenge we have. I do think the deployments are happening. There was discussion about tenders being awarded this week, which would be a good sign and earlier than I had expected them to be awarded. So I do think the drawdown is happening. We just don't have as good visibility as the finished goods and systems with our components in it as opposed to their raw material.

And kind of the follow-up to that, can you just talk about your expectations for your ACO ramp?

Yes. So we continue to make progress. I think we are sampling more and more customers every week and every month getting feedback and getting into the qualification stage. I'd say that in reality, it's probably not material until the end of the calendar year and probably more likely into calendar 2018 because it is a challenge here when there is an incumbent there to be able to first look at the mind share of the customer and then to make our product look like their product and work well with their DSPs. So we're going through that right now, and I think we're making good progress. It's really more -- for us it's a calendar '18 then.

Your next question comes from the line of Troy Jensen with Piper Jaffray.

I got a question for Alan. Alan you made some comments in the prepared remarks about ROADM growth and the next-generation ROADM. I'm just curious to know if you think or your thoughts on the market in China, if you think it's kind of been delayed? And the reason I'm asking is, Andrew Schmitt had put a note out here recently talking about a new generation of WSS or ROADMs, and I think he's calling them an MxN. I think it's the Multicast Switch is integrated in on the line card here or something. So could you just touch a little bit about what this next-generation ROADM is, are competitors going to have it, and the time line for when we think China really does turn on for ROADMs?

Yes, I think there’s the ROADMs that are getting deployed today and those are the ROADMs we make today for North America metro and that we introduced last year or the year before. As I said in the prepared remarks, we have shipped next-generation stuff and it's really the end-by-end type of ROADM that is very custom for the China applications. We have many of those engagements going with our Chinese customers and that's why we think that we're really positioned well for the longer-term next-generation of deployments in China as well as the near-term deployments. I'd say that the near-term deployments have been kind of lumpy with respect to the China telecom deploying a few 100 ROADMs, in fact trying it out. I think we're not expecting a lot in our guidance for this quarter from a deployment standpoint, but I think we're going to ship some, but it's really more of an end of the calendar year and into 2018 when the next generation end-by-end starts coming out in a meaningful way.

Can you just explain the difference in the new, the next generation, is it the Multicast Switch integrated in?

I wouldn't characterize it as a Multicast Switch being integrated in. I would say it's an alternative architecture where…

You need a Multicast Switch.

Yes, we may not need a Multicast Switch to achieve even further functionality in the advanced ROADM architectures.

Okay. And as far as shipping these, will they just ship for samples or ship for revenues?

They are samples. As part of our development -- joint development agreements we have with some of our customers in China.

Your next question comes from the line of Richard Shannon with Craig-Hallum.

Maybe I’ll dive into the QSFP28 topic. You talked about -- a little bit about growth both from LR4 and CWDM4 in the June quarter. Curious if you can lay out for -- how you expect the growth for your business to go in the next couple of quarters? You mentioned some -- maybe some tougher pricing, although you're expecting some lower cost design in CWDM4, but wondering where you're focusing your attention?

Well, our expectations -- my expectations are that both will grow, I think unit growth and CWDM4 will far outstrip the unit growth in LR4. But we do expect growth in both. I think one of the things that we said in the prepared remarks was that we're starting to see the CFP2 contribute to our growth. And so I think there's a long tail in CFP2 in China. We have a good market share there, but I'd say that our expectations are growth in both LR4 as well as CWDM4 especially as we introduce these new generations of CWDM4.

Okay. I'll follow specifically on the CFP2 topic -- the CFP2 topic there, Alan. I think one of your competitors suggested it would be kind of a slow fade over time. It sounds like you're expecting some level of growth there. Can you differentiate that or help us understand why you think at least there's going to be some growth beyond this quarter?

Well, I think it's primarily China. I’d say most of the North America deployments are QSFP28 today or have transitioned or are in the process of transitioning. I'd say that the client side of the China deployments are going to be CFP2 LR4 for quite some time. So there’s, I think a long tail just like we saw on the long tail on CFP, they're still being deployed, although they are finally starting to go down in a meaningful way. I think there's going to be a multi-quarter if not multi-year deployment of CFP2s because there’s chassis out in the network that have slots that need CFP2s. So I think that there’s, first of all the inventory looks like it's kind of taking care of itself on CFP2 in China. So that's why we saw a pick up and we expect to see it continue to pick up as the deployments happen in China.

Your next question comes from Meta Marshall with Morgan Stanley.

I just wanted to kind of see if this is a good summary of areas where you were still kind of adding capacity and it being kind of pump lasers, ROADM, 3D sensing and QSFP28, are there any areas else that we should think that you are adding capacity?

Yes, I think that's a pretty good summary (inaudible).

Okay. And then I might have just misheard kind of in your prepared remarks, but you mentioned a write-down of a deferred tax asset, was I correct or was there more explanation of that or just kind of normal course?

That's correct, yes. There was a write-down of deferred tax assets, roughly $30 million or so where from a likelihood standpoint, we would not be using those assets going forward.

Okay. Most of the questions have been asked, so I will pass it on.

I would now like to turn the call back over to Alan Lowe, President and Chief Operating Officer for closing remarks.

Thank you, operator. I want to thank our customers for their business and again thank our employees for their hard work in growing our business and putting us into an excellent position in the market. We are in the middle of a worldwide expansion of bandwidth. Photonics are increasingly becoming critical to manufacturing and to the next generation high-volume 3D sensing applications making the future bright at Lumentum. We regularly discuss our business at Investor Relations events. These events are listed on our website in the Investor Relations section and are regularly updated.

This concludes our call for today. We would like to thank everyone for attending and we look forward to talking with you again in another 3 months. Thank you.

This concludes today's conference call. All participants may now disconnect.