Ladies and gentlemen, thank you for standing by, and welcome to the KLA Corporation September 2019 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. [Operator Instructions] I would now like to hand the call over to Mr. Kevin Kessel, Vice President of Investor Relations for KLA Corporation. Thank you. Sir, please go ahead.

Thank you, Christian, and welcome to today's KLA earnings conference call to discuss the results of the September 2019 quarter and outlook for the December 2019 quarter. I recently joined KLA, and today marks my first KLA earnings call. I'm glad to be here and look forward to meeting and talking with all of you over the quarters and years ahead. Joining me on the call are Rick Wallace, our President and Chief Executive Officer; and Bren Higgins, our Executive Vice President and Chief Financial Officer. During today's conference call, we will discuss quarterly results for the period ending September 30, 2019. We released these results this afternoon after the market close, and they are also posted on the Investor Relations section of our website at ir.kla.com. Today's discussion of our financial results and outlook is presented on a non-GAAP financial basis unless otherwise specified. A detailed reconciliation of GAAP to non-GAAP results is in today's earnings press release and the earnings slide presentation posted on the KLA Investor Relations website. Our IR website also contains a calendar of future investor events, presentations including those from our recent Investor Day and corporate governance information as well as links to KLA's SEC filings, including our annual report on Form 10-K for the year ended June 30, 2019. Our comments today are subject to risks and uncertainties reflected in the Risk Factors disclosure in our SEC filings. Any forward-looking statements, including those we make on the call today, are also subject to those risks, and KLA cannot guarantee those forward-looking statements will come true. Our actual results may differ significantly from those projected in our forward-looking statements. With that, I'd like to now turn the call over to our Chief Executive Officer, Rick Wallace. Rick?

Thank you, Kevin, and welcome to KLA. Good afternoon, everyone, and thank you for joining us on today's call. I'll start with a brief strategic overview before I cover the business highlights from the quarter. Please turn to Slide 4. KLA continues to see strong momentum in our business from the secular trends we articulated at our recent September 17 Investor Day. Our performance this quarter clearly demonstrates how KLA is benefiting from our strategies for growth, technology, leadership and operational excellence. We delivered another solid quarter, with revenue in both GAAP and non-GAAP EPS finishing above the range of guidance, a result of strong customer pull for KLA's solutions and focused execution, despite a backdrop that still includes some industry headwinds in key segments. As a global leader in process control and supplier of process-enabling solutions for the data era, KLA remains at the forefront of the most important industry trends and technology inflections in the electronics industry. Our deep collaborative customer relationships, broad IP portfolio and differentiated solutions that address our customers' most complex challenges is the recipe that sustains our market leadership. Our business also continues to benefit from more complexity within semiconductor devices as well as multiple mega trends driving demand across multiple product generations in numerous key industries. Please turn to Slide 5. Underpinning our success and consistent outperformance is the KLA operating model, which codifies our corporate values and management principles. We have been running the company this way for a long time, but we discussed it much more extensively at our recent Investor Day to better illustrate its power and impact on the KLA business and explain how it represents critical core competencies that we believe can enhance the long-term performance and profitability of acquired businesses. The KLA operating model is essential to…

Thank you, Rick, and good afternoon, everyone. Please turn to Slide 10 for a review of the September quarter financial highlights. This was a very strong quarter for KLA, with revenue and EPS each coming in above the high end of our guidance ranges. Our free cash flow results also marked a new record for the company. Total revenue for the September 2019 quarter was $1.413 billion, which was above the range of guidance of $1.31 billion to $1.39 billion. Gross margin for the quarter was 60.8% in the upper end of the guided range for the quarter of 60% to 61%, driven by incremental revenue growth and a stronger-than-expected semi process control product mix. GAAP EPS was $2.16 and non-GAAP EPS was $2.48, both of which were also above the range of guidance of $1.75 to $2.05 and $2.04 to $2.34, respectively. Our cash flow execution was exceptional this quarter, as both cash from operations and free cash flow came in at record levels of $496 million and $464 million. We are proud of our financial results this quarter, and we remain focused on executing across all markets as we move forward, with a focus on our integration and synergy plans for the Orbotech acquisition. A key element of our investment thesis is KLA's commitment to returning cash to shareholders. On September 17, we announced a 13% increase in our quarterly dividend level to $0.85 per share. This marks the tenth consecutive annual increase in our quarterly dividend level, reflecting our confidence in our business strategy, the strength of our free cash flow generation and our ability to grow it over time as well as our commitment to returning value to shareholders. In terms of returning capital to shareholders during the quarter, we were consistent and effective in our…

Thank you, Bren. [Operator Instructions] With that, Christian, we are ready for the first question.

[Operator Instructions] Your first question is from John Pitzer from Crédit Suisse.

Congratulations on the very solid results. Rick, I want to go back to one of your comments you made in your prepared comments about the Gen5 optical inspection platform. You mentioned that you expect shipments to double in 2009 (sic) [ 2019 ] and for that to still be a good growth driver in 2020. So I guess, I'm trying to get a sense from you, where in sort of the potential of that product cycle over time do you think we are? Maybe using a baseball analogy might be the right way to look at it. And as you think about logic/foundry being strong next year, to what extent is that just a call on laundry -- on logic/foundry CapEx versus perhaps a Gen5 product cycle that's still in the sweet spot and accelerating?

Sure, John. Thanks for the question. Yes, I think if I thought about Gen5 right now and you want to use baseball, it's the third inning. It's relatively early in the life. We're now on the second iteration of that product in terms of a new platform, and there are several iterations to follow. Engineering is being dedicated to that, but we're now starting to see broad adoption and really the beginning of adoption in production, where what we really had been dealing with for the most part, in the last couple of years, was during the development cycle. So we're still early on, and we think it will become a primary tool. We always thought it would become the HVM tool once we got to EUV in ramping in production, and it definitely feels like it's doing that. The upside to that is, we've seen more memory adoption in the Gen5 than we originally anticipated. So that's probably upside to what we had originally envisioned.

John, it's Bren. The only other thing I'd add to that is as we talked about at our Investor Day, there is an application for reticle qualification in the fabs for EUV reticles. So we're -- as we start to progress that use case, it's how the tool is used for full wafer coverage to be able to qualify reticles as they're being used in the wafer fab. So an incremental opportunity there that is slowly evolving, but we think it adds another dimension of growth for Gen5 as it gets adopted.

That's helpful. And Bren, maybe just for my follow-up, I want to go back to the operating model you shared at the Analyst Day around op margin, sort of, bridge between now and 2023. You're already, kind of, ahead of the revenue -- op margin targets you gave in that model, both in the September quarter and in the December guide, and it sounds like December OpEx is unusually high per your comments. And in general, that op margin target you gave only assumed incremental op margins at the low end of your historical guidance. I'd hate to obsolete the model less than a quarter out, but can you talk a little bit about whether or not you think there's upside or conservatism to that model you gave at the Analyst Day?

Yes, John, it's a good question. I mean, I think one of the challenges in putting together the model is it reflects a different mix of business that we have today but also expected moving forward. So clearly, in the September quarter and even in December what's driving our business in terms of incremental growth is the process control part of the business and that is behaving consistent with the historical model that we've had out there. So anytime the mix is shifting in that direction, you're going to see outperformance against the base model. So what we tried to reflect was based on our expectations for growth over time, what the mix of business would do and how that would impact the model. Obviously, there's synergy and other actions that are embedded in that. But anytime we have an inflection driven by the process control part of the business, we're going to see a period of outperformance. But in terms of your long-term view, I think it's the right way to think about it.

The next question is from C.J. Muse from Evercore.

I guess, first question, on the foundry/logic side of things, can you speak to whether you're seeing increasing breadth of spending in that category and really focused on the leading edge? And as part of that, if you could speak to also reuse? That clearly was a headwind, basically from 20-nanometer down, but it sounds like 7 non-EUV progressing to 5 EUV, but that's a seismic shift here that is driving a real uptick in process control intensity. Would love to hear your thoughts on both of those things.

Yes, CJ, I'll take the question, and then Bren will add some color to it as well. It's not very broad right now. I think that the foundry/logic is -- it's not particularly broad. We expect it to broaden out next year in '20 as it expands. In terms of reuse, there are a couple of dynamics that are quite different from the prior cycles. One is the just very large number of starts -- design starts. So there's a lot of demand being driven by the fact there are so many different devices that are being introduced at these advanced nodes. So we don't suspect that there will be that much. The other thing is, we have new product cycle, which are really adding lot of capabilities. So obviously, customers are always trying to optimize their capital, and we'll see them be as efficient as possible. But we don't expect a repeat in the way we've seen it in some of the prior cycles. And that's based on our early returns and also some of the additional challenges, the advanced nodes are pushing, even without broad EUV. And once you broaden out EUV, you get even more of that.

Yes, I think the big difference between this node, and let's say, 20-nanometer, even down to 14, 16, is we're seeing much broader end market adoption that's driving these design starts through the advanced foundry and the leader there. And so all of that is sort of preventing any reuse as that capacity is deployed to support all that activity, and a high-mix environment puts more pressure on yield and delivering to tight product windows. So it's a good story, and we expect this node to have some legs. And certainly, the investment for N5 is starting to pick up and strong expectations for that.

Very helpful. If I could sneak in a quick second question. You talked about a return to $370 million, $375 million OpEx in March. And I guess, curious, how should we think about potential synergies related to Orbotech as we move through 2020?

Yes, CJ, I outlined it at Investor Day. There's a number of activities that are happening, and we'll see those play through as we move through the year. One of the dynamics that I wanted to put a little bit of extra time into in the prepared remarks was some of the investments that we're making to drive longer-term structural cost reduction, namely, relocating some operations from subscale or higher-cost regions into lower-cost parts of our global footprint. So as you're ramping up one team and ultimately ramp down another team, there's some incremental investment, but those have returns over time. So as I think about that range, I think we're in that range as we move in, based on how we're sizing the business right now and thinking about top line, we'll be operating in that range as we move through the year with the usual program development expenses that could cause you to move one direction or the other. I would think it would probably be lower end or slightly below the range as we move towards the end of the year. At the beginning of the year, it's probably at the higher end.

The next question is from Krish Sankar from Cowen and Co.

I just had a quick one, Rick, you kind of spoke about how Gen5 is going to double in 2020. Is that primarily coming from foundry's 5 nanometer? Or is there a significant chunk coming from DRAM adoption, too?

So Krish, that was -- the comments were that we saw a doubling in '19 from '18. And although, to Rick's earlier point about Gen5, I mean, we're really encouraged by starting to see that deployed into production use cases and away from just defect discovery and R&D applications. So -- and then the print check, as that rolls in, in terms of how it supports EUV reticle qualification. So we would expect to see further growth as we move into next year in Gen5. But we haven't quantified how much more. I mean, we are shipping both generations to support activity, both for 7-nanometer and for 5, but you'll see more adoption of Gen5 in production as N5.

Yes. And just to add to Bren's point, what happened in '19 also was penetration across almost every customer, major customer, in terms of the early evaluation and use so that sets up the case for the longer run. But that was really a proliferation year for '19, where we got it out and established in all the leading jobs.

Yes, I mean, certainly, Krish, just one other thing, as I look at the order profile, it would imply we're going to have growth in the number of units as we move into next year. I don't know if it will double, but we'll certainly see growth in adoption next year.

The next question is from Harlan Sur from JPMorgan.

Great job on the quarterly execution and strong free cash flow. On EUV litho adoption and your mask inspection business from Analyst Day, I think we could gather that your reticle inspection franchise is going to grow pretty strongly this year, up about 45%. But if I look at the shipment profile for ASML, let's say, over the next couple of years, it's still a runway for very strong growth, around 25%, 30% shipping CAGR. With that in mind, how should we think about the trajectory of your mask inspection business looking into next year?

Yes. I mean, I think, Harlan, you're right. We've had a very good year and a down WFE year in reticle inspection and that's been adoption for the Teron platform, both for optical and EUV inspection support. Some of the drivers moving forward is, yes, the number of scanners but also design starts tends to drive reticle inspection demand. So we would expect to see some incremental demand on the 7-nanometer side as we see incremental capacity there. And then you'll start to see some activity as we move into next year. As we talked about at Analyst Day, we do have a new platform that's specific for EUV inspection that we'll start shipping in 2020 also. So you'll see a mix and match. But as we stand today, I think given what's out there for 7-nanometer and then for the layers that will be happening in EUV, we believe we've got the market pretty well served with the capabilities that we're offering.

Yes, makes a lot of sense. And then on the Specialty Semiconductor segment, you drove pretty good quarterly results despite given the trade headwinds. And so again, if I look at the trends heading into next year, you've got a big step up in RF content, in 5G smartphones, in infrastructure, you've got more power products in auto and industrial, and of course, all this advanced packaging and things like system and package. All of this should provide a pretty good backdrop for growth next year, but wanted to get your initial views in terms of how you're thinking about 2020 for SPTS?

Right. Well, obviously, we're newer to that business. But as we spend more time with the team and spend more time with customers, it does -- it is well positioned to grow. And as you've pointed out, we had some headwinds that happened in '19, but still managed to get some traction. '20 looks like a good year for SPTS. And our current modeling of it, based on what we see for the, as you say for RF supporting 5G, it looks like a, hopefully, a 10% to 20% range of growth for that business as we go into next year. So we feel good about it, and we're continuing to learn more about those customers.

The next question is from Timothy Arcuri from UBS.

Rick, so if I look at your WFE share this year, you're going to gain about 50 bps of WFE share, but I think a lot of people are looking at foundry CapEx and they're sort of worried that this is a near-term peak, at least here in the back half of the year. And if you look at the capital intensity numbers, like, we haven't seen these kind of quarterly numbers only really once in the past 20 years from kind of a quarterly capital intensity point of view for nonmemory. So -- also people are looking at TSMC CapEx, and it's kind of $5 billion in Q4, and that's kind of a $20 billion run rate. So how do you respond to concerns that this is sort of a near-term peak in foundry logic shipments? Is that not what you see? Do you see the quarterly shipment rates continuing to grow in foundry/logic next year?

Sure, Tim. Well, it is true that we had strength, as we mentioned, and you're -- half-on-half considerable strength, but really not in a lot of customers. There are other customers that will broaden out in next year, and so we feel pretty good about how we're positioned. We think that the actual mix between foundry and logic and memory for this year is about equivalent to what we see for next year and that's driving some of the capacity or the intensity that we're seeing. So we feel pretty good about next year. Obviously, we had a good second half to this year. But we're counting on some broadening. And again, some of what we're seeing this year is ordering that will actually have deliveries in next year. So we'll see some of the benefit of that revenue next year. Bren, do you want to add any color to that?

No, I think Rick covered it. I mean, look, if you think most of the investment we've seen this year has been from the leader. And so the broadening out across a pretty diversified end demand environment into next year, we feel comfortable given the comments we made in the prepared remarks that we see some sustainability in foundry and logic as we move into 2020.

Okay. And then, I guess, I had a question on Orbotech. If you look at systems revenue, it was pretty flat. It looked like display was still pretty weak. What's assumed for December, Bren, for Orbotech shipments? And what's the outlook for the flat-panel display and the PCB business? I guess, the real question is, why is the flat-panel business still interesting to you? Because it seems like it's kind of on its lows, and it would be arguably accretive if you sold it because it has much, much lower margins. So just -- can you just, kind of, talk through all that?

Well, so we didn't guide the individual segments. I mean, I will say that looks like we'll have an uptick sequentially in FPD next quarter in terms of overall planning. That does tend to be lumpy, but based on the order profile we saw in the June quarter and lead times there, that -- those deliveries happen in December. In that business, it's a build-to-order business, so usually when we get the orders that the schedules hold pretty well. So I would expect it to uptick. I think as we think about next year, I don't think we'll see any real recovery in the flat panel business. And I think that after 6 years of growth, we saw 2020 -- or 2019 was a difficult year. I don't think it's going to grow much in 2020 either. So we're focused on the business. I mean, I hear your points, and we're -- some of the actions we're taking that I referred to in terms of cost, structural cost, reduction actions are particular to that business. And so the first thing we're going to do is get that business in a place that we believe is an acceptable level of profitability at these trough levels and position that we can -- as the business recovers, that we can scale it and drive operating leverage through that model. But we're taking these actions first. I hear your point, and we'll see how it plays out over time.

The next question is from Vivek Arya from Bank of America Merrill Lynch.

Congratulations on the strong growth. I had 2 questions as well. For the first one, I'm curious, what is the right way to think about process control and density in EUV versus non-EUV systems? I imagine EUV was not as big of a factor this year. Do you think it's a bigger factor for you next year?

Yes. Process control intensity for EUV is higher than it is for non-EUV. In spite of the fact that there are fewer layers. I mean, obviously, the value of EUV is a reduction in layers. And so we would see that, that has some mitigating effect on the increased intensity around EUV, but you really get process control intensity in a couple of ways: one, you get it in a mask shop because you have to do more inspection to assure that you have a high-quality mask. And we're still, as an industry, trying to work out some of the challenges, pellicle, nonpellicle and getting mask to be good in the lifetime of the mask; the other thing is in print check to verify and validate that Bren mentioned in an earlier question, we see that as an opportunity and that's been driving process control intensity around Gen5. So overall, it's a plus for some of our businesses, it is a minus for some, but the net-net of it all is process control intensity ticks up in an EUV world as the industry goes forward. And on balance, it's a net positive for KLA.

And my follow-up, how large was domestic China spend for you this year? And just conceptually, how are you thinking about it going into next year?

Yes, it's a good question. Actually, as I said in the remarks, I mean, at the beginning of the year, we thought it'd be down 10% to 15% from a very solid number in '18. And it turned out with some projects that came back into the year, one major one in particular, we're going to end up about flat. And as I look at our preliminary scoping into 2020, and you do have some mix on what kind of projects are going to be invested in, but it looks relatively flat to me again. So I don't see it changing much as we move into 2020. So it's in the $650 million range plus or minus in terms of revenue level for the company. And I would expect that to stay relatively consistent based on what I see today for 2020.

So that is not part of the outperformance assumption for next year?

No. I think -- look, I don't think it's going to fall off. I mean, certainly, it improved the performance in -- with a factor in performance improvement in 2019. So as I think about next year, no, I don't -- I'm not banking on a lot of extra business in China to deliver the year in terms of how we're planning. I expect a flattish environment in native China.

The next question is from Quinn Bolton from Needham & Company.

Just wanted to follow up on the question from Vivek about EUV driving higher process intensity. I assume that that's true in the memory side of the business as well. So as we look at EUV insertion in 1Z and then 1 Alpha processes, do you see a meaningfully higher process control intensity in those steps as we see EUV coming into high-volume manufacturing? And then I've got a quick follow-up.

Yes. I mean, it's early for memory. As you know, they're still trying to figure out what role EUV will play and how many layers, but that is definitely a driver for us in a positive sense, again, for the same reasons. You have the mask shop and you have on the wafer, and you have a reduction in layers counterbalancing it. I don't think it's a huge change in the intensity. It's a nudge up, but it's not a big change in the overall intensity, maybe on the order of, if you're going to be at an average of, say, 10% process control intensity, you might get 0.5 point to 1 point of increase in terms of overall process control intensity on that. So it depends, but it will depend how much it's deployed. And ultimately, it will also depend on what the strategy is relative to what people are going to do with pellicles or not do -- have pellicles. It will change a little bit in how it plays out for us.

Okay, great. And then a follow-up question was just, it sounds like you're getting a little bit more optimistic in the memory side of the business, and that might be somewhat driven by the indigenous Chinese guys. But just wondering if you could comment based on, sort of, orders for the Surfscan business. Are you seeing -- starting to see better activity in the NAND business? Or is it fairly isolated to certain customers at this point? You wouldn't call it, kind of, the beginning of a broader-based recovery?

No, I wouldn't call it that. I mean, certainly, that business inflected pretty strongly through calendar '18, and so it's been down this year. And I would expect to see a little bit of recovery in the wafer part of that, which, ultimately, supports memory in 2020. But right now, I think it's too early to see that impact. So I wouldn't say I see that as a leading indicator of new business. But certainly, given the market position of that product and how memory drives wafers, if you do see a pickup in memory, we will see that business impacted.

The next question is from Joe Quatrochi from Wells Fargo.

Congrats on the results. As it relates to your service business, I was wondering if you could give us an update on some of the capacity idling that you've seen, particularly around the memory side in your installed base? And maybe, how do we think about that relative to your gross margin guide for the December quarter?

Yes. No, it's a good question. I mean, one of the things that we are seeing that drove an uptick in the revenue level for Service quarter-to-quarter was a reduction in idling, so more of that capacity being utilized. And so that drove some incremental revenue to the Service line for the semiconductor side of the business. So look, you do have a gross margin impact to growth in Service on the overall model, which is contemplated in the guidance that we give generally. And the thing to keep in mind is it with our Service business as it grows, the profitability stream, we believe, is an accretive stream to the overall total. So it's not affecting, I think, quarter-to-quarter, it has very little effect on the overall gross margin in terms of the guidance. So it's really -- when I think about guidance quarter-to-quarter, it's mostly around some of the product dynamics and how that plays through our model.

Okay, that's helpful. And then on the specialty semi side, now I think you're calling for flat this year. Can you help us understand the -- was there any incremental impact indirectly from Huawei? Or is this more related to weakness that you're seeing in just general demand?

Yes. It -- I wouldn't say it's the incremental effect that we talked about last quarter. So we continue to see that. And we talked about, if I recall, $40-ish million impact for the year related to that one in particular. There are broader trade issues. Certainly, there's the issue with Korea and Japan and how that's affecting Korean OSATs. So there are number of trade issues beyond Huawei that's affecting that business. Now in the long run, as those customers in that area -- or the end customer, the customers' customer starts to redesign around some of these issues, it creates opportunities for us. So -- but in this year, it's affecting that business. Also, automotive has been on the margin a little bit weaker, so that's affected that business as well. So it's still showing, I think, a decent result this year against the backdrop of the WFE environment we're in and would expect to see it recover next year in terms of growth, as Rick suggested earlier, but it's a business that has a really good market position. I think the long-term drivers around 5G, advanced packaging, power, and so on, are good ones. And so we're encouraged by what we see, and we like the business model of that business.

Yes. And more specifically, I think on Huawei, if you thought about the reactions of the suppliers to Huawei was there was uncertainty when the -- they had first came in and trying to figure out, and I think everybody froze. So the baseball analogy would be there was a rain delay in terms of trying to figure out what the next step was. And now there is more activity going on in the supply chain. So we think that business ultimately comes back, but it was impacted probably more and longer in the year because just uncertainty around that. Now I think people are dealing with it and that will actually cause the business to resume its growth.

The next question is from Patrick Ho from Stifel.

Rick, maybe first off, in the past, you've talked about increasing metrology intensity on the NAND flash side of things, particularly as layers increased. As the industry starts shipping to 128 layers, do you see any potential new opportunities on the inspection side given the complexity of the layers and the increased number of layers for next-generation devices?

I think the main opportunity -- it's good question, Patrick. The main opportunity for inspection is really in what's happening on the bare wafer on Surfscan and dealing with the flatness requirements as the wafers get higher, both the cleanliness and flatness. So it may be less obvious at first. I mean, we've talked in the past about being hopeful about finding defects and deep trenches, but it's probably more flatness-driven and desire to have clean wafer. So that's where really been where we see it. Metrology has some opportunities. We talked about at the Analyst Day with the work we have on CD-SAXS, in terms of great opportunity there. But specifically for inspection, it's more about Surfscan.

Great. As my follow-up question, in terms of the domestic China opportunity, obviously, we're seeing some of these capacity built begin and you're benefiting from new capacity build out. But given some of it is also trailing edge logic-type of capacity builds, is this simply because they are new capacity? Or are there opportunities within the, I guess, the foundry and logic segment of that region that you see new opportunities, both for your inspection and metrology products?

Yes. I think that the -- there are a number of players that are small, maybe subscale, in terms of what you think of it as the larger. So there's actually some inefficiency in that supply chain if you're starting smaller companies in terms of trying to support for a strategic region -- reason. So that, kind of, creates additional demand for us and we're benefiting from that. Also, I think that as they're trying to figure out some of these capabilities, then there's opportunity for us to help them and that drives both good share but also good intensity. And then lastly, I think the move for auto, as China is an important auto market, there are more Specialty Semiconductors and there's more opportunity for us there as a result.

Yes. Patrick, I mean, look, I think this year is a heavier memory year in China. Not by much, I'd say, probably of our mix of semiconductor business, it's 55-45 to memory. I look at calendar '20 and it's pretty balanced. So to your point, there's a lot of activity around IoT. Rick talked about automotive. And so the yield earnings challenges in logic are more complicated. The tools that we are shifting are configurable in terms of helping customers meet their requirements, also providing some upgrade pass as they start to progress some of their technologies. So -- and they're serving, I think, real markets ultimately. And so I think that they're designing for specific opportunities, which is good just in terms of the long-term trajectory of that investment and their viability. So we feel pretty good about all of it. And I think it's a mix across both segments there.

Your final question is from Toshiya Hari from Goldman Sachs.

Congrats on the very strong results. I just wanted to go back to the half-over-half kind of cadence for your business going into next year. You guys accurately called your core process control business being up, I guess, every quarter on a sequential basis, this year, a couple of quarters ago. So I guess the question is, based on customer projects, based on customer conversations, what are your thoughts into the first half of 2020 relative to the second half of '19? Do you think you can grow your business? Or should we be expecting more of a flattish outlook on a half-over-half basis?

Well, like I said, I think we're little early to guide second half -- or half-to-half trajectory, then obviously, I think a part of how we do in the fourth quarter will impact how the March quarter lines up. I mean, across the segments, again, as we said earlier, we feel that there's broadening and sustainability investment in logic and foundry. I mean, if you just take a step back and look at the second half, up 20-ish percent over the first half, it does, obviously, imply that WFE is higher in the first half -- or the second half versus the first half, so the run rate of this WFE is probably 10% to 15% kind of growth into next year at these levels. So it does imply that you've got a second half. You have to have that kind of growth to, sort of, sustain at this level of business. But I think that's probably as far as I can go at this point.

Yes. I guess what I would add to that is we feel pretty good about understanding our position relative to WFE. We're not in a great position to predict WFE. Does that make sense?

Yes, it does. As a quick follow-up. On China -- on native China WFE, I think, Bren, you talked about 2020, at least at this point, kind of, looking kind of flattish relative to 2019. When you think about the mix of spend on process control, could that move higher? Or what are your preliminary thoughts on that?

Yes -- no, as I said earlier, I think it's balanced across the business that we expect, and we expect a flattish level of business in 2020. So I'm not expecting -- at least for our business, not expecting growth but not expecting decline in 2020 at this point. We'll see how it plays out. But that's the least where we stand today.

Thank you. And that concludes the Q&A session. I would now like to turn the call over back to Mr. Kessel for closing remarks.

Yes. Thank you, Christian. And thank you, everyone, for your time today and interest in KLA. Christian, please conclude the call.

Ladies and gentlemen, that concludes today's conference call. Thank you for participating, and you may now disconnect. Have a good day, everyone.