科磊 (KLAC) 2018 Q1 法說會逐字稿

Good afternoon.

Welcome to KLA-Tencor's First Quarter Fiscal Year 2018 Earnings Conference Call.

(Operator Instructions)

Ed Lockwood with KLA-Tencor Investor Relations, you may begin your conference.

Thank you, Christine.

Good afternoon, everyone, and welcome to our conference call.

Joining me on our call today are Rick Wallace, our President and Chief Executive Officer; and Bren Higgins, our Chief Financial Officer.

We're here to discuss quarterly results for the period ended September 30, 2017.

We released these results this afternoon at 1:15 Pacific Time.

If you haven't seen the release, you can find it on our website at www.kla-tencor.com.

A simulcast of this call will be available on demand following its completion on the Investor Relations section of our website.

Today's discussion of our financial results will be presented in a non-GAAP financial basis unless otherwise specified.

A detailed reconciliation of GAAP to non-GAAP results can be found in today's earnings press release and in the Investor presentation on our website.

There, you'll also find the calendar of future investor events, presentations and conferences as well as links to KLA-Tencor's SEC filings, including our annual report on Form 10-K for the year ended June 30, 2017.

In those filings, you'll find descriptions of risk factors that could impact our future results.

As you know, our future results are subject to risks.

Any forward-looking statements, including those we make on this call today, are subject to those risks, and KLA-Tencor 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'll turn the call over to Rick.

Thanks, Ed.

I'm pleased to report that KLA-Tencor delivered record shipments, revenue and non-GAAP earnings per diluted share in the September quarter, the results of continued successful execution of our long-term strategies and demonstrating once again the critical nature of process control in enabling growth and innovation in the semiconductor industry.

Q3 results featured strong demand from memory customers and from the Japan and China regions.

New orders in September were a record for the third calendar quarter, and new orders have now totaled approximately $3 billion on a year-to-date basis in calendar 2017.

Coupled with our market leadership and supported by strong backlog, these results point to continued momentum of growth for KLA-Tencor and position the company to deliver new highs for revenue and earnings and strong stockholder returns in 2017 and 2018.

This is a very exciting time for the company given our market and technology leadership.

The road ahead features more opportunity for the industry, our company and for our stockholders than ever before.

Before I turn the call over to Bren, I'd like to take a few minutes to highlight how our strategies for customer success, innovation, productivity and talent are helping to drive KLA-Tencor's record-setting performance today and are building a solid foundation for continued growth in the future.

The first and most enduring objective for KLA-Tencor is customer success enabled by the use of our process control solutions.

To achieve this goal, we focus on partnering with our customers to help empower their growth.

We drive engagement to enable us to develop differentiated solutions for our customers, most complex yield challenges in the current node and the anticipated yield challenges in future technologies.

The rapid growth of our China business is an example of how successful execution of collaboration and customer success strategies are fueling growth.

China's investment and their domestic semiconductor industry is driving an inflection for the overall WFE market, and KLA-Tencor's market position in China is strong both in terms of process control adoption and market share.

Business levels in this region are expected to grow significantly in calendar year 2017, and this momentum should continue through 2018.

The focus on innovation at KLA-Tencor is tightly integrated with our customer success efforts.

We are investing in a very high level to drive innovation in process control and deliver a steady stream of advanced solutions and services that will enable our customers to succeed with their most critical technologies.

Today, we are developing 5 major technology platforms, 2 of which are targeted specifically at memory, and we're modeling higher process control adoption and record revenue from memory customers in 2017.

For example, growth from memory customers is a key factor in the record business levels we're experiencing in our bare wafer inspection and metrology products.

Wafer manufacturers have been adding new capacity to ramp wafer demand, and IC customers are moving to higher-layer accounts and increasingly more complex architectures.

In this environment, our new bare wafer products are needed to support more stringent wafer flatness and process tool cleanliness specifications in advanced technologies.

I'm also excited to report that our Gen 5 broadband plasma wafer inspection platform continues to gain momentum in the marketplace.

We're seeing increasing utilization of installed tools with our advanced logic and foundry customers where the unique value of Gen 5's illumination wavelength range is being demonstrated on 7-nanometer discovery and EUV print check applications.

Based on these results and customer's feedback, we expect Gen 5 to continue to support customers in development of 7-nanometer technology, and we see it playing an even larger role on the development and ramp of the 5-nanometer node.

A third component of our strategic objectives is productivity and the commitment to operational excellence.

Through continuous execution of these strategies, our growth and market leadership results in industry-leading margins and cash flow and provides the resources for future investment and ongoing strong returns to stockholders.

And finally, our focus on talent.

KLA-Tencor's talent initiatives are centered on attracting, developing and inspiring our global workforce.

We recruit top talent worldwide from experienced candidates as well as high-caliber graduates from leading universities.

Ultimately, the success of our execution on talent strategies lie at the core of our long-term success.

So to summarize.

Clearly, this is an exciting time for KLA-Tencor with the company reporting another quarter of record results in Q3 and demonstrating the strength of our strategic objectives in an environment of very strong overall industry demand.

Looking ahead to 2018, given our anticipated backlog entering the year and the momentum we are experiencing today, the stage is set for another outstanding record-setting year for KLA-Tencor.

And with that, I'll now turn the call over to Bren for his comments.

Bren?

Thanks, Rick, and good afternoon, everyone.

As Rick highlighted in his opening remarks, the September quarter featured another record results for the company driven by memory demand and solid operational execution for KLA-Tencor.

Shipments were a record $977 million, revenue was at the top end of the range and GAAP to non-GAAP diluted earnings per share each came in above the range of guidance for the quarter.

This result was driven by continued strong demand across our product portfolio with particular strength from memory, wafer and captive mask shop customers.

Our manufacturing and service organizations continue to deliver strong operational leverage on our revenue growth with incremental gross margins in excess of 80% over the last 2 quarters.

Revenue and earnings per share were also records in Q3.

Revenue was $970 million, and GAAP diluted earnings per share was $1.78.

Non-GAAP diluted earnings per share was $1.80.

In our press release, you'll find a reconciliation of GAAP to non-GAAP diluted earnings per share.

With the exception of when I explicitly refer to GAAP results, my commentary will be focused on the non-GAAP results, which exclude the adjustments covered in the press release.

Now turning to highlights for the September quarter demand environment in terms of shipments.

Foundry was 40% of shipments in September driven by an anticipated broadening of the customer base for investment in 10-nanometer production and 7-nanometer development and by continued investment in legacy technology nodes.

We are currently modeling foundry shipments to be approximately 20% of the total in the December quarter.

Memory was 46% of shipments with deliveries evenly split between DRAM and NAND.

We are currently modeling shipments to memory customers to be about 67% of the total in the December quarter with NAND representing about 55% of the memory mix.

Logic was 14% of shipments in September and is currently forecasted to be approximately flat in the December quarter.

Regarding the approximate distribution of shipments by product group.

Wafer inspection was 46% of shipments; patterning was 30%, patterning includes shipments from our reticle inspection business; non-semi, which includes our back-end component inspection business, was approximately 3%; and service was 21% of shipments.

Looking forward, we are modeling December quarter shipments to be in the range of $945 million to $1.025 billion.

At this point, we expect shipments for the first half of calendar year 2018 to grow in the low single digits compared with the second half of 2017 and be roughly equally balanced half-over-half over the course of the year.

This outlook is strengthened since our last earnings call in July.

Turning now to the income statement.

Revenue was $970 million in September, finishing at the top of the range of guidance, and we expect revenue to be in the range of $930 million to $990 million in the December quarter.

Non-GAAP gross margin was 63.7%, above the range of expectations for the quarter.

The stronger-than-expected gross margins in the quarter were due to a number of favorable factors in the systems business, including the favorable product mix and better-than-modeled parts expense and labor utilization in the service business.

Looking forward to the December quarter, we expect non-GAAP gross margin to be in the range of 63% to 64% as we expect a similar product and service mix quarter-to-quarter with commensurate cost dynamics in our manufacturing service operations.

Total non-GAAP operating expenses were $252 million in September, and non-GAAP operating margin was 37.8%.

This operating margin result is in line with our updated target model for annual revenue levels in the $3.6 billion to $3.9 billion range that we outlined earlier in the year.

We are modeling operating expense levels to be approximately $260 million in the December quarter due to the expensing of significant prototype materials and other expenses related to technology development programs targeting reticle process control for EUV high-volume manufacturing applications.

Given company revenue expectations for CY '18, we are currently sizing operating expense levels next year to be approximately $250 million per quarter, consistent with our public operating model.

Our non-GAAP effective tax rate was 17% in Q3 and below the guided rate of 20% for the quarter, reflecting the higher mix of revenue from products developed or manufactured offshore and certain other discrete items impacting the tax rate.

For the December quarter, we expect the tax rate of 18% due to similar factors.

For calendar '18, we are adjusting our long-term planning tax rate down to 20% largely due to the anticipated revenue mix in the year.

We intend to update our tax rate guidance periodically based on the revenue expectations.

Finally, non-GAAP net income for the September quarter was $284 million, and we ended the quarter with 158 million diluted shares outstanding.

Turning now to the balance sheet and our cash flow statement.

Cash and investments ended the quarter at $3.1 billion, an increase of $40 million compared with the June quarter.

Cash from operations was $374 million in September, and free cash flow was $358 million.

We paid $156 million towards our outstanding term loan in the quarter and expect to retire our outstanding $250 million 2.375% 3-year senior notes upon maturity in November.

Finally, during Q3, we paid an aggregate of $100 million in regular quarterly dividends and dividend equivalents for fully vested restricted stock units.

In conclusion, today's results reflect our continued market leadership, the critical nature of process control in all customer segments and our industry-leading business model.

This, fueled by expected strong total backlog as of the end of the calendar year, position the company for continued growth in the strong and stable overall WFE industry environment.

Given consensus expectations for WFE growth in the mid- to high single digits next year, our expectations for another year of strong memory investment, improving process control intensity in memory, contributions from new products and continued high single-digit annual growth in services, we are modeling total revenue in the $4 billion range in 2018.

Against this revenue backdrop, operating performance is expected to be in line with the $3.9 billion to $4.2 billion annual revenue range outlined in our public business model.

With that, to conclude, our guidance for the December quarter is shipments in the range of $945 million to $1.025 billion, revenue between $930 million and $990 million and GAAP diluted EPS of $1.58 to $1.82 per share as well as non-GAAP diluted EPS of $1.59 to $1.83 per share.

The December quarter EPS guidance range assumes an 18% tax rate in the quarter.

This concludes my remarks.

I will now turn the call back over to Ed to begin the Q&A.

Okay.

Thank you, Bren.

At this point, we'll open the call for questions.

(Operator Instructions) All right.

Christine, we're ready for the first question.

(Operator Instructions) Your first question comes from the line of Farhan Ahmad from Crédit Suisse.

My first question is on the outlook for 2018.

Can you just talk about what segments have strengthened since your last outlook?

And in terms of foundry versus memory, where do you see growth and where do you see kind of flattish?

Yes, Farhan.

So I think since last -- it's Bren.

So since the last call, certainly our expectations for next year has strengthened over the last 3 months, and we're seeing that in our outward forecast, which I provided some details to you in this call.

I think when we look at the data, I think with the investments that are happening in 3D NAND, I think we still continue to expect that to be very strong.

DRAM also looks like it's probably stronger now than what we thought a few months ago, and I think foundry/logic is flattish.

I think logic will be slightly higher.

But overall, I think the foundry/logic segment will be largely flattish and probably more second-half weighted as we see it flow through our shipment profile over the course of the year.

And then one question on the print check.

Have you guys sized what could be the potential opportunity for print check?

And what is the customer interest that you are seeing right now?

Farhan, you mean specifically for EUV?

Yes, specifically for EUV in terms of print check technology, how large that market could be.

Yes.

Okay.

Well, let's start with development and then think about HVM.

For development, there's a lot of work.

In fact, a fair amount of capacity of the Gen 5 systems are being used for print check and development.

So I'd say that I don't think it consumes a tool right now, but it consumes part of the use case of a tool for us.

When we get into HVM, it gets a little different.

It depends on how big we think the adoption of EUV is and when.

But we would expect that, that market grows because of just essentially the available technology for inspecting reticles.

And so that will drive the print check market higher because there won't be an actinic tool for in-fab reticle for the first wave of tools.

So I think you'd see it grow in size from them.

We haven't sized it exactly, but I think for every company that adopts EUV in production, as the EUV scales, the applications for print check will grow commensurate.

So maybe it gets to a couple tools per customer that are driving that.

Your next question comes from the line of Harlan Sur from JPMorgan.

Just kind of follows up to the last question and based on some of your prepared commentary, it looks like you guys have seen some momentum on your reticle inspection platforms.

And we've heard that at 7-nanometer EUV mask inspection, it's Teron platform plus Gen 5 print check.

For 5-nanometer and below, though, it's kind of interesting because I think at the recent SPIE technical symposium, I think the KLA team actually talked about a multi-beam e-beam platform for EUV reticle inspection targeted at 5 nanometers and below.

Can you guys just give us some more insights on this new multi-beam tool that you're developing?

What are the benefits versus current generation?

When is it going to be available?

I know 5 nanometers is still a few years away, but I would imagine maybe sometime in 2019?

Any color there would be great.

Sure, Harlan.

So the strategy for reticle inspection, as you mentioned, Teron right now is serving the need.

And I think for customers, as they implement 7, whatever degree, 7 is going to be implemented, Teron with the combination with the Gen 5 for print check is the most economical and efficient solution for our customers.

But you're right to point out that a 5-nanometer, our view was we will need more sensitivity on the reticle.

Rather than design an inspection tool -- an e-beam inspection tool and modify a wafer tool and make it for reticle, we decided a couple years ago, several years ago, to target specifically the reticle market with a platform optimized for reticle inspection and being developed by a fair number of players from our platform that develop reticle tools for us.

So we think we have a lot of expertise.

We feel very good about that progress of that, and we are targeting introduction of that tool in the 2019 time frame to intercept 5-nanometer.

Hey, Harlan, it's Bren.

One other point I'll make, and I said this is in prepared remarks, but one of the uplifts that we're seeing, the onetime uplift we're seeing to our operating expenses here in the December quarter, is driven by some prototype investment that we're expensing for this program.

So Rick's characterization of timing seems to make sense and program's progressing according to plan.

Great.

And then your commentary on China suggests that you guys are seeing some of the activity on some of the China domestic memory R&D pilot programs.

I think they're trying to show up in your pipeline.

Are you going to be shipping some of these tools here this quarter?

Or is this more kind of the first half of next year?

And secondly, on the domestic China business, you guys have always said that process control intensity, in general, is about 100 basis points higher versus the mainstream guys.

Is that kind of how it's playing out?

So Harlan, on your first part of your question, so for the new memory projects and there's about 3 significant ones that we're tracking, we booked some orders back in the June quarter for a DRAM project that we will start to ship at the end of this quarter, right at the tail end of this quarter for installation into next quarter.

The -- we also booked in this quarter significant orders for a 3D NAND project, and so we'll see those orders probably ship out towards the June quarter time frame.

So to answer your question, we will see that business start to come to the P&L next year.

This year was more memory weighted as I described and probably a little over double in terms of the bookings from -- in China versus last year.

And as we look at 2018, the order profile looks like it's flat to up from there.

So a significant amount of activity there.

Obviously, based on success with these Phase 1 investments, we play a pretty significant role early on as they're ramping the fabs.

And so process control intensity is higher.

Share tends to be better also because of the -- just the experience we can bring to help these customers navigate their yield and ramp these fabs quickly.

And obviously, in memory, being able to progress the technology road map matters a lot to the economics of these projects.

But just to add to that, I think -- remember the greenfield, in particular, these are not at-scale fabs.

So we're at the front end where the intensity is higher as Bren was saying.

I think the other advantage we will have in the memory market in China is some of the platforms that we've been developing for memory will come to maturity in time to support those, so we'll see process control intensity higher.

But the investors in China and the fab managers are very focused on efficiency in all their investments, so we don't see them necessarily higher.

It's more that the fabs are subscale and at the beginning of the ramp.

Your next question comes from the line of C.J. Muse from Evercore.

First question on gross margins.

Can you walk through what's driving the uplift in the December quarter?

Is that simply just the greater revenues?

And as you think to 2018, you outlined gross margin in line with your target model, 62%, 63%.

Though if I look back in the last 4, 5, 6 quarters, you definitely hover around 63%, if not higher.

So curious what we should be thinking about mix-wise that might drive that to the mid-point of your target model or whether you're just being conservative a bit.

Yes, on the last part first, C.J., I think that as we look at next year and look at the expected mix of business, we're likely going to see sustaining tailwinds that we've experienced in gross margin and we'll see that continue into next year.

So as I'm modeling next year, I'm modeling it between 63% to may be as high as 64% over the course of the year.

So pretty -- I think it continues, and what's driving it is a lot of the same factors.

I mean, clearly, the product positioning and the differentiation and value we're adding to customers is playing out in the product pricing.

I think the introduction of new products over the last 12 months or so -- the engineering execution has been great, so we've been able to climb up the cost curve pretty quickly on warranty and installation and some of those kinds of things.

We've been able to get a fair amount of leverage out of our service business as we've added tools into Korea and Taiwan.

We can leverage the resources pretty well.

So all in all, I think there's a number of factors, and I think that they're fairly sustaining as we progress into next year.

Very helpful.

And I guess as my follow-up, if you think about the greenfield activity that's flowing through today, typically, that would be the ideal time for you guys in terms of process control intensity.

So curious to hear your thoughts on what you're seeing both from a logic foundry perspective as well as from a memory perspective.

Well, I think Rick talked about from a memory perspective and what we're seeing in China.

I think for the established customers, I mean, to that point, I mean, there is a participation -- a higher level of participation as they outfit a new fab on the front end.

And then over time, the customer will probably scale back as they ramp that fab, as we generally see across a fab ramp.

China is unique.

Rick talked about that.

And on foundry/logic, I don't think that there's anything different.

I think it sort of follows the same model.

There's a fair amount of development activity that happens up front, and we participate above what the normal ramp level or the normal intensity levels are.

And then over time as customers move from defect discovery to line monitoring, they tend to try to optimize a little bit more for fab productivity.

Your next question comes from the line of Edwin Mok from Needham & Company.

So first, I guess, just kind of look a little longer term.

I think you guys provided some color for 2018, including kind of outlook for second half of '18.

Just curious, that's by a few quarters away.

It seems like you have better visibility now than before.

What gives you that better confidence?

What's improving in visibility for you guys?

Well, so the statements were about the shipment profile.

And as we look at this point and we look at the shipment profile into the second half of the year, it looks like based on the backlog we're bringing into the year and our expectations for orders in the first half gives us pretty good views into shipments into the second half.

So we feel, at least at this point, it looks reasonably balanced across the year.

And we're in this range of about $1 billion of shipments plus or minus $50 million as we look at our plan over the next several quarters.

So it's still a ways out, and things can change.

But based on how customers are slotting and how we're planning our manufacturing operations, that's how it looks.

Okay.

That's fair.

Second question I have is on the service side of the businesses.

If I look at some of your peers and some of other equipment manufacturer, they're actually seeing pretty strong growth in service, double-digit kind of growth rate, which is higher than historical.

Do you see that being possible for your service business?

Or you think that kind of a mid-single digit is a reasonable growth rate for your service business?

Well, service looks like in calendar '17 will grow about 10%.

Now its historical growth rate has been about 8% if you go back to, let's say, 2000 or so and play that through.

Now it grows every year, as it should, as the installed base is growing.

So we are seeing an acceleration in service, and we were open with this in our last discussion of just our modeling for the business over the long run that we think service can maintain its historical growth rate.

So we think service runs probably 7% to 9% over time.

And part of what's driving it obviously is the growth of the installed base, but also the mature and legacy fabs and how they're running those fabs now to meet IoT and automotive demand and so on.

So we're adding new capacity to some of those tools.

They're running the tools harder, so they're running them longer, and so the contract and billable streams are nice.

And one thing about our service business is it's about 70% to 75% service contract.

So it's a very predictable stream that we can build our cost structure around and move forward with.

Your next question comes from the line of Toshiya Hari from Goldman Sachs.

Can you remind us how you view process control intensity in memory going forward over the next couple of years?

And I know you guys are working on a couple of initiatives in memory from a product development perspective.

How big could the opportunity be with those initiatives over the next, say, 12 to 24 months?

Yes, so on process control intensity for in and around memory, I think at planar NAND versus 3D NAND and DRAM, DRAM was always around 9%.

Planar NAND was probably 8% or so.

And if we look at across total memory today, we're in that 9% to 10% range.

Clearly, as we've been talking about for several months, we think there are some opportunities not only for more inspection in terms of various defectivity in 3D NAND but also opportunities in and around metrology as metrology's much more intensive for 3D structures than planar structures.

One thing that's also been driving our business is the increasing specs around flatness and wafer flatness for 3D NAND.

And so we've seen our wafer customers investing more new capability to be able to produce wafers for the IC customers to be able to meet that spec challenge.

That doesn't show up necessarily in some of our pure IC numbers from memory, but it's certainly a derivative effect that's positive.

So we think we make some progress on these 2 offerings that we think that, that can probably drive to an opportunity that can be $200 million to $300 million across both platforms.

I don't think you'll see that in '18.

There's an adoption time line that's associated with that.

But over time, it could be -- if we can solve the problem, I think there's an opportunity that's in that magnitude.

Okay.

Great.

And as a follow-up, you guys have talked about your business with some of the wafer manufacturers and even some of the mask shops.

I guess you made some comments in your prepared remarks.

I guess these are 2 customer groups that perhaps some of your peers don't necessarily have access to.

So I guess I was curious, how big is -- how big are these 2 customer groups as a percentage of sales today?

And what's kind of the outlook into 2018 when you think about their spending patterns?

Well, we don't break it out, but it is significant.

And it is often the case that when there are new specs is when -- and new products is when we see new adoption of technologies.

So it isn't something we break out, but it is significant, especially for certain of our products.

Bare wafer inspection we mentioned and also flatness we mentioned as being critical.

And then of course, in the mask shops, it's the reticle tools both for overlay -- registration, but also for defect inspection.

But pretty significant, but usually, the investment cycle on those is a little bit more in waves and less of an annual basis.

Your next question comes from the line of Jagadish Iyer from Summit Redstone.

Rick, I have 2 questions.

First, on -- you had a nice uptick in the patterning segment in the September quarter.

What is the real driver for it, please?

Yes, Jagadish, this is Bren.

So that was reticle inspection.

And so we had a couple of tools that we booked in the quarter focused for -- on EUV development activities from captive mask shops.

Okay.

Fair enough.

And then just as a follow-up, I wanted to understand, now as EUV starts to make initial insertion next year, how should we think about the overall kind of the process control market given all the puts and takes about how multi-patterning could be changing at least in '18 and 2019?

Any high-level thoughts here?

Sure.

I think there are a couple of things to think about.

The net of it all is we're not, at this point, capable of really seeing a big change in process control intensity, but it does move to different parts of our business.

For example, right now we know there's even more pressure on wafer manufacturers for flatness for EUV.

So there's an example where we'll get more.

And what seems like perhaps unrelated, we also think that the Gen 5 application and adoption will be higher to support EUV print check but also initial production.

There are fewer layers.

So some of the films areas, we'll reduce -- the number of steps, we'll reduce, but there are new defect types.

So right now when we model, we don't see a lot of difference in terms of overall capacity, and we have a number of flows that are what we believe the process flow would look like depending on the success of EUV.

Where it will start to matter more and it's already impacting some is in reticle inspection, both with the Teron and then when we talk about later as we get to new capability 2019 for the gen -- or for the 5-nanometer and beyond, we'll see increased adoption as we go forward there.

Your next question comes from the line of Patrick Ho from Stifel, Nicolaus.

Maybe as a follow-up to the last question on EUV, given the lack of actinic inspection tools at least at the outset, is there the potential, as it goes to high-volume manufacturing, that you could see increasing intensity for optical inspection at least at the outset to offset the lack actinic inspection?

Patrick, no.

I don't think so.

I think that the high-volume if it's done in the current generation of EUV, will be served by the Teron platform.

I think when we get to the high NA production for EUV, which will come at the sub-5 probably in maybe 2022, 2023 time frame, that's when there'll be a need for the EUV reticle tool, but also potentially actinic tools at that time frame.

But we don't see a need for them before that.

We think we're fully covering the market in the next few years with the toolsets that we have.

Great.

That's helpful.

And my follow-up question in terms of China.

You talked about some of the memory fabs that you guys are tracking and where you're seeing the activity already start in terms of process control buy.

How do you look at, I guess, some of the logic and foundry fabs that are also starting to develop in China and some of the opportunity that even though it is going to be a little more mainstream and legacy nodes, what are some of the opportunities there for you given that even the 28-nanometer node does have high levels of process control intensity?

Well, we've already seen some of that, right, with established players.

And they do invest and you have to, again, look at scale because those fabs tend to be bigger than the giga fabs elsewhere in the world for those nodes.

So process control intensity, by that definition, is higher.

We do see increased interest in those nodes.

And those new facilities, of course, when they're greenfield and they're coming up, they don't have the option of reusing any other equipment, so we do see opportunity there.

So we think that there will be continued opportunity.

And initially, we didn't see, I'd say a year ago, so we weren't envisioning the memory that we're seeing now in China, but we actually were anticipating logic.

So it's something we've been prepared for and even have, in some cases, restarted old lines to support them with products that are more appropriate for those design rules.

If I look at the mix -- yes, just one other comment.

If I look at the mix for calendar '18, it looks like it's slightly more foundry weighted than memory.

Calendar '17 was more memory weighted; '16 was foundry.

So I would expect more foundry investment in '18 from that region, and that's the native China business.

(Operator Instructions) Your next question comes from the line of Farhan Ahmad from Crédit Suisse.

Just a question on the dividend.

If I look at your dividend payout right now, it's somewhere around 30% to 32% for this year, next year earnings.

And your long-term model, you've talked about 40% to 50% range.

So how should we think about the dividend going forward?

Are you planning to stick to the 40% to 50% range?

Yes, Farhan, it's a good question.

I mean, clearly, we raised the dividend back in June, and we raised it almost 10%.

Our model -- and the business has strengthened and the performance has been very strong.

So clearly, we've outpaced that target.

I mean, that's -- when we talk about that target being in that sort of 40% to 50% range, that's a [3-cycle] target over time.

We think about it that way because you're going to have periods you may be under it when business is strong, but business (inaudible) obviously want to build it to be able to not only just maintain it, but also raise it in weaker times.

So over time, we target the growth rate of the total dividend payout versus the growth rate in earnings.

And so if you look at our model right now where we see the top line growing 6% to 8%, we think we can grow the bottom line greater than 10% on that, that we'd grow the dividend payout in that greater than 10% range.

So there's a couple different ways you can look at it.

I mean, clearly, as the business continues to grow and we've raised it now with 6 or 7 years in a row.

So this is a very consistent process on an annual basis, and we'll run through that process again as we complete our strategic planning process for the year.

And also, Farhan, as you can imagine, if there were a change in tax policy and we had access to more of that cash, we would revisit the payout ratio.

Your next question comes from the line of Harlan Sur from JPMorgan.

Just kind of following up in terms of capital return.

So it looks like you guys are going to get to your target gross debt level of $2.5 billion sometime very soon.

I think you guys are already at your target leverage ratio.

So in a pretty short period of time, you're going to have a lot of strong excess free cash flow.

You obviously have had a strategy of, number one, investing in the business; and then secondarily, a strong focus on capital return.

So how do you guys think about uses for this excess free cash flow, i.e., something like turning on a stock repurchase program?

Well, so we started buying back some modest levels of stock a couple quarters ago.

And you're right on all accounts, right?

As we get through next quarter, we retire this note next month.

We will be consistent with the commitments that we made back when we restructured the capital structure of the company back in 2014.

So given the strength of the business, we're below our leverage levels.

And so there is -- it's not necessarily a compelling reason for us to continue to delever.

And we would redirect, I believe, in the shorter term towards share repurchase.

Now there is the U.S. and offshore cash dynamic in terms of cash flow.

And given the strength of some of our product lines, I think that our U.S. cash flow into next year is probably at the lower end of the 60% to 70% target range that we've referenced in the past.

But you're right, there will be some cash flow that will become available, and we'll run through our process.

And first and foremost, we invest in our business, and we're doing that.

And then secondly, as part of our -- capital structure is part of the company strategy.

And so as we look at growth opportunities juxtaposed against shareholder return alternatives, it's part of our DNA and will be something that we will do.

So I'll announce something when we have something.

There are no further questions at this time.

Mr. Ed Lockwood, I turn the call back over to you.

All right.

Thank you, Christine.

Thank you, all, for joining us today.

This concludes our conference call.

Have a wonderful rest of the day.

This concludes today's conference call.

You may now disconnect.