KLA Corporation (KLAC) Q1 2014 Earnings Report, Transcript and Summary

Good afternoon. My name is Mike, and I will be your conference operator today. At this time, I would like to welcome everyone to the first quarter fiscal year 2014 earnings call. [Operator Instructions] Thank you. I will now turn the call over to Ed Lockwood with Investor Relations. You may begin your conference.

Thank you, Mike. 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 first quarter results for the period ended September 30, 2013. We released these results this afternoon at 1:15 p.m. Pacific Time. If you haven't seen the release, you'll find it on our website at www.kla-tencor.com or call (408) 875-3600 to request a copy. A simulcast of this call will be accessible on demand following its completion on the Investor Relations section of our website. There, you'll also find a 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, 2013, and our subsequently filed 10-Q reports. 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. More information, including factors that could cause those differences, is contained in the filings we make with the SEC from time to time, including our fiscal year 2013 Form 10-K and our current reports on Form 8-K. We assume no obligation and do not intend to update those forward-looking statements. However, any updates we do provide will be broadly disseminated and available over the Web. I'll now turn the call over to Rick.

Thanks, Ed. Good afternoon, everyone. I will lead off the call today with summary highlights in Q1 and the current business environment, and then provide guidance for the December quarter. Then Bren will follow with more details on the results. Q1 2014 was a good quarter for KLA-Tencor, demonstrating our market leadership, the strength of our core markets and process control and solid operational execution. New orders in Q1 grew 11% sequentially to $790 million, above the range of guidance and setting a new record for total bookings for the company in the September quarter. We experienced good momentum in order activity in the quarter, and demand was strong across each of our end markets, highlighted by record memory customer bookings in Q1. Bookings activity was also robust in foundry, where we witnessed pull-in orders to support 20-nanometer development. The solid demand affirms our ongoing focus on providing superior value to customers, both in terms of meeting market requirements and delivering superior competitive offerings. In fact, a key contributor to our strong bookings performance in Q1 was the success of new products, particularly with memory customers, with whom, we achieved quick acceptance of our latest generation plasma inspection tools, as well as a key mask inspection win. With the increasing cost and complexity associated with leading-edge memory, we are seeing higher adoption of process control for memory customers, a trend we believe will continue to benefit KLA-Tencor over time given our market leadership position in that field. As we look ahead to the December quarter and into 2014, we expect continued strong order momentum, high levels of business activity with our customers. And we believe this sets the stage for a positive calendar year 2014 for demand and capital equipment industry overall, and for process control in particular, giving the increasing importance the process control is playing in helping our customers achieve their growth and cost objectives at the leading edge. With KLA-Tencor's strong backlog, our market position and the forecasted growth of the segments in which we participate, we believe we are well-positioned to continue to successfully execute our strategies to deliver superior growth and financial performance relative to our industry and return significant value to stockholders. Looking ahead to the December quarter, we expect the strong demand environment to continue as momentum for new technology development continues to strengthen across multiple customers. We're encouraged by the pace of this investment and how well we are positioned with new products to address this increasing cost and complexities that are associated with competing at the leading edge. Now for guidance for the December quarter. December bookings are expected to increase approximately 10% at the midpoint and be in a range of $800 million to $950 million. Revenue for the quarter is expected to be between $670 million and $730 million, with non-GAAP earnings in the range of $0.67 to $0.87 per share. And with that, I'll turn the call over to Bren Higgins for his comments. Bren?

Thanks, Rick, and good afternoon. This was a solid quarter for KLA-Tencor in terms of financial performance and operational execution. Revenue for Q1 was $658 million. Fully diluted GAAP earnings per share was $0.66. Non-GAAP earnings per share was $0.68. In our press release, you'll find the GAAP to non-GAAP reconciliation of the $0.02 difference. My comments on the quarter will be focused on the non-GAAP results, which excludes the adjustments covered in the press release. As Rick mentioned, Q1 results are highlighted by new orders finishing above the upper end of the range of guidance at $790 million, with our order forecast for both Q1 and for Q2 strengthening as we progressed through the quarter. Memory investment is focused on 3D technology development and new capacity in NAND and technology upgrades in the DRAM. Foundry demand in Q1 was stronger than expected with upside to our original forecast coming from pull-in orders to support 20-nanometer development and pilot activity. We see foundry orders increasing in Q2, with ramp in investment in 20-nanometer and strong customer acceptance of our latest generation products driving order growth. The regional distribution in new system orders and quarter-to-quarter change was: the U.S. was 22%, down from 38% in the June quarter; Europe was 8%, up from 6%; Japan was 10%, down from 12%; Korea was 18%, up from 10%; Taiwan was 26%, up from 10%; and the rest of Asia was 16%, down from 24% last quarter. The distribution of orders by product group was: wafer inspection was approximately 52%, reticle inspection was 8%; metrology was 19%; service was 20%; storage, High Brightness LED and other non-semi was approximately 1%. Finally, for new semiconductor system orders, the approximate distribution by customer segment was as follows. As expected, memory was strong once again at 47% in new system orders in September compared to 44% in June. Foundry came in above the original forecast at 45% in new orders, an upside from sales directed at 20-nanometer compared with 33% in June. Logic was 8% in new orders versus 23% in June, which was in line with our original forecast. Our continued heavy investment in research and development is enabling us to maintain our strong position at the leading edge as evidenced by the fact that investments by our customers at 20-nanometer and below constituted roughly 76% of the orders we received in the September quarter. Total shipments in the quarter were $637 million, down 17% from the June quarter and slightly under the midpoint of the guidance range. In total, we ended the quarter with just over 1.2 billion of backlog comprised of $964 million of shipment backlog or orders that are not yet shipped to customers and expect to ship over the next 6 months, and $251 million of revenue backlog or products that have been shipped and invoiced but have not yet been signed off by customers. We expect shipments in the December quarter to grow 30% at the midpoint compared with Q1 and be in the range of $800 million to $860 million, with a number of newly released products shipping in volume in the quarter. Turning to our income statement, I'm proud of the operational execution in the September quarter with revenue, gross margin and earnings finishing at or above the upper end of our guidance ranges. Revenue for the quarter was $658 million. This level is down 9% quarter-to-quarter and down 9% versus the same quarter last year. For the December quarter, total revenue is expected in the range between $670 million and $730 million. Our revenue guidance for the December quarter reflects the impact of high shipment levels of newly introduced products during the quarter. In general, our revenue recognition policy requires the completion of formal customer acceptance on new products prior to recognizing revenue. This formal acceptance process typically takes up to 90 days to complete. As a result, revenue recognition on several tools shipping during the quarter is expected to be delayed into the March quarter. Once the acceptance pattern is demonstrated for new products that ship to the same facilities with the same specifications, our policy generally allows for revenue recognition at the time of shipment. As a result, we expect strong sequential revenue growth in the March quarter as we begin to recognize revenue from the December newly released product shipments and as these shipments begin to convert into revenue more rapidly than in the December quarter. Gross margin was 58.2%, up 40 basis points from the June quarter despite the lower systems revenue level. Our gross margin exceeded guidance due to the favorable product mix and lower inventory reserve charges in the quarter. We expect gross margin to remain roughly flat in the December quarter as we support higher levels of manufacturing in shipments. As revenue grows over the next several quarters, we expect gross margin to increase consistent with our 60% to 70% incremental gross margin model. Operating expenses were $228 million, up $5 million from the June quarter. Operating expenses were $8 million higher than the midpoint of guidance for the quarter, primarily due to a pull-in of investments for some of our engineering programs and timing issues associated with cost-sharing milestones of certain projects. R&D was $131 million, up $5 million from June as we continue to invest in key research and development activities and customer collaborations for next-generation technologies. SG&A for the quarter was $97 million, about flat quarter-on-quarter. Going forward, we are sizing the company's quarterly operating expenses in the $225 million to $230 million range for the next several quarters. The timing of engineering investments and milestone achievements in our cost-sharing programs will lead to some fluctuations within this range quarter-to-quarter. Other income and expense for the quarter was a net expense of $10 million, down slightly from the June quarter. We expect a similar result in the December quarter. The tax rate was 21% in the quarter, lower than the 23% planning rate due primarily to nontaxable increases in our deferred compensation savings plan and the tax benefits resulting from divesting of employee equity awards in the quarter. At the 23% guided tax rate for the quarter, earnings per share would have been $0.66. Going forward, you should continue to use the long term planning rate of 23% for modeling purposes. Net income was $115 million, or $0.68 per fully diluted share. Turning to the balance sheet, cash and investment ended the quarter at $2.95 billion, an increase of $33 million over the June quarter. In the quarter, we repurchased 61 million of stock at an average price of $58.31. As of September 30, we have 4.8 million shares available for repurchase under our current authorization. We paid a dividend of $75 million in the quarter. In July, we raised our dividend per share $0.05 to $0.45 per share. This was our fourth consecutive annual increase in our quarterly dividend and reflects our ongoing commitment to returning capital to shareholders. Cash from operations was $177 million in the quarter, comparable to the June quarter. Net inventory increased by $26 million quarter-to-quarter to $660 million to support the expected shipment ramp. Inventory turns on GAAP COGS were 1.7 versus 1.9 in the June quarter. Accounts receivable finished the quarter at $441 million, down $84 million from the prior quarter. Day sales outstanding based on shipments were 63 days. Net capital expenditures were $22 million to support facility expansion activities worldwide. Fully diluted shares ended the quarter just under $169 million and are expected to remain flat for the December quarter. Full-time headcount ended the quarter at 5,913. Finally, as we commented earlier, we're encouraged by the strength of the business environment at the leading edge. We believe the factors we described earlier that led to what looks to be a strong second half of calendar year 2013 will continue into 2014, giving us confidence that we'll see industry CapEx growth in 2014 of about 10%, if not a few points higher than that. Given process control adoption during no transitions in our market position, KLA-Tencor is well-positioned to deliver market leading revenue growth. With that, to reiterate, our guidance for the quarter is: bookings are expected to be within the range of $800 million and $950 million; revenue between $670 million and $730 million; and earnings per share of $0.67 to $0.87. This concludes our remarks on the quarter. I will now turn the call back over to Ed to begin the Q&A.

Okay, thank you, Bren. At this point, we'd like to open the call to Q&A. [Operator Instructions] We'll do our best to make sure everyone has a chance for follow-ups in today's call as time permits. So, Mike, we're ready for our first question.

[Operator Instructions] Your first question is from John Pitzer with Crédit Suisse. Farhan Rizvi - Crédit Suisse AG, Research Division: This is Farhan, calling in for John. You're guiding to a very strong growth in the January quarter. And I just wanted to understand what's the mix that you're expecting. Is it going to be mainly foundry-driven? Relative to your shipments in this quarter, how do you expect the mix in the January quarter -- sorry, in the March quarter?

I think you mean the December quarter. We're not going to guide the March quarter. This is Bren. So when you think about the mix for December, we see memory at about 30%, and that's 50-50 NAND and DRAM. Logic at about 10% and then foundry would be 60%.

And your next question is from the line of Krish Shankar with Bank of America Merrill Lynch.

Rick or Bren, first question, when you look at all the memory-related bookings you're getting, are these primarily technology-related? In other words, I'm trying to figure out does this mean that they will slowly start weaning off for you guys as the quarter progresses because the technology buys would be done by then?

Well, as Bren mentioned in his prepared remarks, the memory -- there's 2 elements in memory. One is the work on 3D, which is a new technology that's going to be ramping over the next several quarters. And then the other is on DRAM, there are technology buys. And I think we're well-positioned in both to see increasing adoption, which we laid out that case at SEMICON West and that's starting to be what we're seeing now. So I think we're in reasonably good shape. And also, as we mentioned, there is the element of the technology buys at the front end of ramps, and we hope to continue to see that over the next couple of quarters.

Got it. And then if I could just -- for the follow-up, December quarter looks like you might probably do $400 million in bookings on the foundry side. Can you help us understand how many foundries is it going to be? Is it going to be 1, 2, or 3, and is it all 20-nanometer?

Yes, this is Bren. And I'll start and then Rick can follow up. So most of the activity we expect in the December quarter is focused on 20 and below. Multiple customers, investing. There will probably be some 28 investments as well that will happen in the quarter, but that's generally how we see it.

Right. I think everyone is in the mode right now -- all the foundry players are in the mode of needing to make investments to support the FinFET work that they're doing below 20, but we're also seeing, as Bren mentioned, some of the guys are maybe a little bit behind or also investing to get yields up as they anticipate 20-nanometer demand. There is even some 28 spending that is still going on.

Your next question is from Harlan Sur with JPMorgan. Harlan Sur - JP Morgan Chase & Co, Research Division: Given your view on WFE spend for next year and sort of given your pipeline and visibility, Rick, can you just give us an idea on the weighting of that spend, maybe first half versus second half and some color on how the different market segments, logic, foundry, memory will kind of shake out from a mix perspective next year?

Why don't I let Bren start and then I'll follow-up?

So I think -- I don't think know if we can comment at this point around whether it's first half or second half. I mean right now, it feels pretty evenly distributed. But we'll see how that goes as we get closer to it. When you think about the composition, I think memory is probably somewhere between 30% and 40%. And so I think you'll see logic foundry there at 70 with, let's say, the logic pieces around 20 and foundry at the difference. So as I said, pretty evenly distributed. I think we're, right now, we're sizing the company given our comments in the prepared remarks around WFE, up 10%, maybe a few points better. So that, I think, puts us in a position where we're sort of sizing the company around $800 million to $850 million a quarter in terms of new business levels. Harlan Sur - JP Morgan Chase & Co, Research Division: Got it. And given that most of the activity today is around both migration and given the significant step up in complexity for this next-generation processes, I know that you guys have talked about increased process control intensity at the front or the ramp of a new node. But are you guys also seeing a bias towards usage of predominantly leading-edge tools, like for example, a bias towards your 29xx platform versus your 28xx platform? Any way to quantify, let's say, for 3D NAND or 16-nanometer FinFET, what percentage of tools are skewed towards your leading-edge tools, is it 70, 30, and so on?

No. It's more -- it's higher than that if you're talking sub-20-nanometer. And really, the mix is not so much 29xx versus prior generation. The broadband plasma, I think, is almost all going to be the 29xx if you look below the 20-nanometer. It's really a blend of that versus the laser scanning tools that we have. And again, as we laid out at SEMICON West, you'll see a mixture of that probably at the front end of a ramp more heavily loaded toward the broadband plasma and then later in the ramp, the laser tools will come into play. But you're right, process intensity going up. We think in 2014, probably somewhere 15% to 16%, and that really depends, as Bren indicated before, what -- how memory ends up being as a total percent of the mix. While memory is up, it's, as we know, it's less intensive for process control than foundry and logic is.

The next question is from Tim Arcuri with Cowen and Company.

First thing, Bren, just a reference to your answer that you just gave. So you're sort of implying that March quarter revenue, given all the deferred, is sort of an $825-ish million range, is that right?

Well, I'm not guiding it, but I think that given these new tools actually get acceptance and we move into the March quarter, there is going to be a delay there. So I'm expecting strong sequential growth, somewhere in probably the 10% to 12% off of the midpoint. So I'm not guiding March. $825 million probably feels a little high, but we'll see how we do when we get there and we'll see what happens in terms of our performance relative to the guidance ranges that we gave. The higher end, I think it makes it more possible, but we'll have to see how it plays out.

Okay, got it. And then can you just give us the September breakdown in memory bookings by NAND and DRAM? And then I wanted to ask Rick, if I look at R&D expense and I just compare it relative to the last time you were sort of an $800 million range. I'm just looking at the March. It's like a good $15 million, $20 million, arguably even $25 million higher. So can you just sort of address what that increment is? Is it related to $450 million versus last time or is it related to an increased competitive environment?

So first on the NAND versus the DRAM in the September quarter, it was 60% NAND, 40% DRAM. As far as the increased R&D, there's a couple of factors. One, just the advanced technologies are requiring more investment and just the materials alone, if you think about bringing those in, are up. But there are 2 programs that are -- so I would argue absent EUV work that we're doing in the reticle space and some of the 450 we're doing, we'd actually be at a lower level. But -- so the efficiency keeps going. But that's not really fair because we have to do 450, and there's some element of investment that's going on with that. And then the EUV, while it's not at full rate, it is certainly taking investment that is now -- probably nothing comparable to that back then because we were at the end of the 6xx peak investment at that point. So I think those are the probably the 2 biggest single elements to that.

I guess the only other thing I'd add to that is that we are investing in some other markets that we weren't back then. So that's probably a factor in this as well, right? So more TAM [ph] opportunity but some investment requirements to participate.

Your next question is from Srini Sundararajan.

The additional spend from Intel on solving the defect density, do you have an idea on what that could be and whether you will see that from other vendors such as other foundries such as TSMC and others?

Well, as you know, we're not going to comment on specific customers. But in general, it is absolutely true that once you go sub-20 or really, it's about FinFET, that I think that process control and complexities really increase in 2 areas. And I'd say one is, no question, there's more metrology opportunity there, especially as EUV is being delayed. I think there's more opportunity for opportunity in process control around just the patterning space. And then the other one is defectivity, and we are seeing advanced opportunities for us as people go sub-20. And as I mentioned earlier, the need to go to our latest generation broadband plasma tools and the latest generation laser scanning tools. So I think all those factors are contributing.

Okay, just a quick follow-up. Rick, do you have any comments on the Tokyo Electron, Applied Materials merger potentially?

No, not really. I mean I think that it kind of makes sense, I suppose, from their perspective. From ours, it's -- the one perhaps advantage is we've remained very focused on an opportunistic time period where we've got a great market. And I think one of our competitors is going to be a bit distracted. But we're not kidding ourselves. All competition is going to be intense and we got to continue to execute.

Your next question is from Edwin Mok with Needham & Company. Y. Edwin Mok - Needham & Company, LLC, Research Division: Rick, I have a question on the memory booking. And If I look at a number and also triangulating with kind of increased memory intensity that you guys talked about on this -- during SEMICON, it seems like memory booking is picking up a lot faster. Do you see that the memory intensity is actually expanding faster than what you guys have talked? Well, I think you guys talked about only 140 basis point on the next 2 node. Do you expect that to kind of accelerate as you see these new memory orders?

We don't know. It's early. But I would say 2 factors. One, there was a long period in which people weren't investing. And I think that there's a bit of catch-up going on relative to that. And then the other factor is I think some of the guys that are ramping new technologies have recognized the value of accelerating their investments. So getting inspection metrology capability in place at the front end. And some of those guys have logic experience now and they recognize the value of that. So I think there's some catch-up. I think there's front end loading, small degree. And I think that I'd still stick with the model that we showed at West. Y. Edwin Mok - Needham & Company, LLC, Research Division: I see. Okay, that's helpful. And my follow-up just quickly on account FinFET versus 20-nanometer planar gate, right? You said a step-up in process control intensity. Based on the comments from the last question, it sounds like you believe there is. And in that way, if that is, is there a way that you can kind of quantify that?

Yes. We definitely model that as we laid it out at West. But I think the process control intensity goes up both in the metrology space but also the defectivity. We did quantify that. I don't have a...

About 30%, if you went from, let's say, 28 to 20 and below. A little hard to parse out what's 16 versus 20. I mean, they're doing the lithography at 20, FinFET at 16 and below. But generally, about a 30% increase in terms of dollars per wafer start. It's the same tool set, generally, for the 20 and below. So we capture it that way.

Your next question is from Mahesh Sanganeria with RBC Capital Markets.

Rick, a question on consolidation in the process control. We still have several players in the process control. In visual processing side, you're seeing a pretty good consolidation. What are your thoughts on consolidation, especially in the metrology segment?

Well, I don't see a lot of impetus to change for us. I think we're well-positioned with the products that we have. It may be true that some of the smaller players that we compete with will find that there's some benefit in merging or doing some M&A. But I kind of don't think that makes a lot of sense because a lot of what they've done to be able to have a business at all is focused on a segment. And I actually think that they would run the risk of losing some focus if they were to try to put together companies that didn't naturally go together. I don't think anybody else has the same kind of opportunity to put together a KLA-Tencor kind of company based on just geographies and technologies. And I think we're uniquely positioned to do that. So it may happen, but I don't see it likely that it happens.

And then one more question on market share. I think, earlier in the year when the markets were a little bit weaker and your orders were weaker, there were a lot of discussions on market share losses in several segments. Now that you are coming to closer to the end of the year, do you have a good estimate for where you will end up in market share on some of the segment, particularly on the wafer inspection, considering that there are claims being made that you're losing market share?

Right. We don't see it. We haven't seen that we are losing. We've seen, in fact, that we have some strengthening going on in the broadband plasma adoption, which is very high-value capability for our customers, strong performance for us. I think that it's always hard when we have competitors who have a small part of their business in our space. It's hard to refute their claims that they're gaining share. But I think if you look at our overall relative performance, we seem to be doing well, and I think that's a function of both process control adoption but also our share. We think there's actually opportunity, I mentioned this at SEMICON West, to see some gains in market share through this year and then into 2014. And we feel very good about the position we're [Audio Gap] that we had in the quarter are associated with new technologies, new capabilities. So we think that that's strengthening our position as we go forward. So we feel very good about the market share.

Your next question is from Vishal Shah with Deutsche Bank.

Rick, I just wanted to understand your outlook in 2014, kind of 2014. If you're talking about increasing process control intensity and greater process control share in the memory space, is it fair to say that you guys should see better-than-expected revenue growth, better then industry revenue growth in 2014? And also, can you talk about how you see the use of e-Beam tools, sub-20-nanometer node in logic?

Yes. I'll let Bren take part one and I'll take the second part.

Yes. I think you know we covered this a little bit in our prepared remarks. But I think given the compelling nature of these node transitions and how customers are now actively moving that direction that we think we are positioned to see the kind of dynamics that we saw, let's say, in 2011 and 2012 around relative growth for process control in KLA. So I think, in contrast to most or the earlier part of this year where maturing yields in 28-nanometer, the last sort of 1/3 of capacity being added that most of the spend was really focused on capacity buys. So as we move forward here with what we believe is a strong technology environment, given the strength that we're seeing with our new products and multiple new products across all our major core segments, we feel pretty good about an opportunity to outperform the industry.

In terms of e-Beam below 20-nanometer, I probably meant it this way. I think the e-Beam per review of optically detected defects continues to be a good segment to be in. And we have a very strong position in that segment. We gained share recently and have a great product offering there. In terms of inspection, it remains a niche application. I do think there are opportunities for it but it's a relatively small niche. And I think it would grow on the order of similar kind of growth rate to what we see in optical inspection, overall, probably not as fast. But I think there is some opportunity but it's pretty niche. So I don't see that there is a particular gap that, that technology is filling. I think it's more of a specific niche. Review is where we see the most opportunity and we've got a strong position there.

The next question is from Stephen Chin with UBS.

Just wanted to follow-up, again, on the logic order question at 14-nanometer. Do you get the sense that higher logic spend at 14-nanometer to address this defect density issues is all accounted in your 2014 view, again with the process intensity growing to 15% to 16%? Or do you get the sense of growth beyond that as we get into 2015, which is when the foundry has really hit the FinFET spend.

In 20 -- or just in terms of logic spend, we had a very strong first half of the year in logic. Now we tend to get more lead time in logic. So it's not -- it's consistent with our forecast that we haven't seen a lot -- we didn't see a lot of orders in the September quarter and don't expect that much in December. I think as we progress into next year though, I think as the 14-nanometer ramp starts to play out and planned out, I think that we'll see it pick up then. I mean, it's always interesting whenever we hear these little data points or these or announcements around, well, CapEx is up or CapEx is down or we're having these issues. We work pretty closely with our customers, and so within -- around the logic space, we're pretty tight in terms of their plans around yield management and process control. So we don't necessarily see these blips in business associated with an error there.

Okay. Maybe I can follow up then on the gross margin progression that you're going to likely see on all these new tools. I mean how should we think about modeling cost of goods sold from these products shipped but not recognized for sales? Is there perhaps a step function increase in gross margin in the March quarter when the revenue show up?

No. I don't think it's going to be all that different than what we have seen historically. And that's why I made the comment about, from an incremental gross margin perspective, the 60% to 70% model is how you ought to be thinking about it. Certainly, in any given quarter, we'll have mix issues that will drive us over or under that range or the higher end or lower end of that range. I mean, one of the things that's encouraging is we are starting to see a bit of a pickup in the mash-up segment of reticle inspection here going forward. And that's a more favorable mix product for us that has been pretty weak over the last several quarters. So I think as that comes in, I think we feel pretty good about the mix overall and the margin performance. And assuming we can manage our product transitions well and minimize our inventory exposures, I think we should be fine.

Next question is from Jagadish Iyer with Piper Jaffray.

Two questions, Rick. If I look at your foundry orders based on the guidance for the fourth quarter, it looks like your foundry orders for calendar '13 could be down somewhere in the vicinity of, say, 15% year-over-year. We've all known that the foundries are supposed to be the highest in terms of the process control intensity. I'm just trying to reconcile where are we wrong here and how was the trend going to shake out in 2014, given all the industry inflections in the foundry next year? And I have a follow-up.

So this is Bren, I'll start, and then Rick can come in if there's anything to add here. So yes, foundry in calendar '13 was a little bit weaker for us so I think your numbers are about right. We're seeing the orders here starting to pick up now here in the second half. And I think it bodes well for what we expect in '14 where we think foundry should be solidly up next year to drive the 10% kind of growth that we expect. I think when we think about next year right now, memory is probably flattish for us, right? Maybe a little bit up, but with foundry picking up and logic may be up as well. So...

And it's probably true that if you looked at bookings, obviously, that would be true. If you look at revenue, probably a slightly different story because a lot of what we saw in '13 was revenue that came for foundries from orders that were in '12. And so we -- there is kind of this node component of it so that at the beginning of 28, we would see some strength and then you kind of hit an air pocket until you get to 20, which is what we're starting to see some of now we think that picks up. So it's a little bit hard to dissect it on an annualized basis. You really have to look node to node. And that's how we expect, and we are already seeing signs that, that will strengthen as we go into '14.

And then as a follow-up, [indiscernible] backdrop that we all know that spending is going to be up at least 10% year-over-year. [indiscernible] subsegment between inspection metrology and reticle inspection going into next year, given the spending mix that you see, particularly with 3D NAND being extremely process control intensive. so I'm just trying to get a sense from that.

Can you repeat the first part of the question? We didn't hear it.

Yes. The first part of the question was that with spending going to be up 10% year-over-year, how should we think about movements that trends between your various segments, i.e. between your wafer inspection metrology as well as reticle inspection between 413 and 414, and given the spending that you're going to see particularly from 3D NAND, which is process control intensive. I'm just trying to get a sense of how directionally the wafer inspection metrology and reticle moves.

Well, so reticle inspection tends to be a little bit lumpy. So we can probably expect to see that somewhere between sort of the 10%, maybe as high as 15%. But generally, most of -- that is a business that has been a little bit weak. It's starting to pick up again. But -- and I think we'll see investments happen in a given quarter and then it will fall off on follow-on quarters. So I think you can think about it in terms of 10% to 15% range. I think metrology generally is about 15% to 20%. And I think wafer inspection is around 1/2 the business, and that includes high-end pattern inspection on pattern e-Beam and so on.

Your next question is from Mehdi Hosseini with SIG.

Rick, what is going to be driving the variance in your December quarter booking? Is that by the type of a device manufacturer, foundry versus memory? Or is it more driven by product mix?

You mean how do we have the range?

Yes.

Both sides of the range or...

Yes. What's going to drive having -- you having the low end versus the high end.

Yes. It's less about particular device types and more about customer activity. And as you know, we often have some products that are pretty big in terms of the big impact because the dollar value is high and with those come in or if they don't, that'll swing it. So I think it's less about any particular segment and more just the general variation we see in the bookings environment, given fewer customers that place bigger orders and just if they come in or not. Nothing in particular in terms of device type or projects that we're counting on.

Got it. And then going back to prior commentary on R&D. You emphasized some 450-millimeter-related projects. But how about the EUV en masse inspection? Could you elaborate on your plans? Or how should we think about incremental spending that is needed to get the R&D up and running?

Well, we have investment going now. We'll continue to make that investment. We believe that eventually, there's going to be a need for an [indiscernible] inspector so at EUV wavelengths. We have the capability now in our 6xx platform to satisfy market needs, probably through the 10-nanometer node for EUV. Although it's not ultimately what people want for volume production. It will have -- we'll start seeing delays in our forecasted timed market if we don't ramp inspection investment through 2014. And right now, we're working with customers to determine exactly what the demand is. It's a bit of a waiting game because our customers are feeling, I think, more and more concerned about the insertion point of EUV production. And I'd say in the last few months that the sentiment has been pushing out EUV in terms of its high-volume manufacturing, which means that pushing out and giving sort of more time for the investment. So on the one hand, that'll dampen our investment in our reticle tool, but on the other hand, it creates great opportunity and patterning process control and we're going to benefit from that. So I would say, it's still stay tuned because we haven't secured customer commitment toward the need and the sharing of some of the investment risk of that. And I think that's going to take some time.

Just to make sure that I understand, that incremental investment that you would need to make, that has to do with sub-10 nanometer or you're still not sure on 10-nanometer?

No, we think that the tool set that we have, the 6xx, can support -- we think that 10-nanometer will be pretty light for EUV production. And so EUV is likely pushing out of 10-nanometer, in which case, customers will be able to, by and large, do what they need to do with the capabilities that we have. But that's not high-volume EUV manufacturing. So we'll need to invest significantly more if we want to have a tool that intercepts high-volume manufacturing EUV, but that continues to be delayed.

The next question is from Patrick Ho with Stifel, Nicolaus. Patrick J. Ho - Stifel, Nicolaus & Co., Inc., Research Division: Rick, can you just remind us what the incremental increase in capital intensity is in the transition from planar to 3D NAND in terms of process control? And maybe you could go, well, one step below on the inspection side as well?

Yes. I don't have it committed to memory. I think, in general, that what we see is that there's a significant CAGR on the increase if you go to 40 nanometers down to some of the 3D stuff. But remember, the adoption rate was low, I think, in the mid-8s. And we're talking about a 30% increase in that adoption as you go to the 3D. And we're seeing some evidence. But again, nobody has really done volume 3D, so nobody really knows for certain. But the initial activity we see from our customers support the thesis that we're going to see increasing demand for process control.

[Operator Instructions] The next question is from Ben Pang with Northland Capital Markets.

For the foundry node, can you refresh, again, on what the differences are in the capital intensity between 28, 20 and FinFET? And for 2014, for foundries, do you think your opportunity is bigger for 20-nanometer or for the FinFET development? And the last question is, what should we look for, for the tax rate for 2014?

All right. I'll start here on the easiest one. I think the tax rate does fluctuate a little bit, but I think for modeling purposes, you should just go ahead and assume 23%, now bridge it to the actual against that number as we go through. But I think that overall, as we model it out, that's a reasonable percentage.

So on the question of process control intensity, what we laid out at West was that in the logic foundry space that the 2x nanometer, 4x, it was about, we average 15.8% intensity. At 2x, it was 17%. And we saw that at 1x going to 18.4%. So that's really the difference between the 2x would be the planar work and what we see going into the 3D. And so 18.4%. For memory for 3D, we saw that increasing as you go from 8.8% as we said up by about 30%.

Do you think in 2014 for the foundries, are you going to sell more of your equipment for 20-nanometer capacity or for FinFET development for foundries?

Well, I mean I think a lot of it, to Rick's point, the intensity goes up on a per wafer start basis. So it really gets back to how much capacity ultimately gets added in terms of how many tools are actually purchased, how much we actually end up shipping. So -- but node to node, as Rick mentioned, we see a 30% increase in dollars per wafer starts spent on process control.

There are no further questions. I will turn the call back over to Mr. Lockwood.

Thank you, Mike. And I'd like to thank everyone on behalf of the management team for joining us here today. An audio replay of today's call will be available on our website later this afternoon. And once again, we appreciate your interest in KLA-Tencor.

This concludes today's conference call. You may now disconnect.