Great. Welcome back, everyone. I'm Joe Moore from the Morgan Stanley Semiconductor team and very happy to have with us today the management team of KLA, Rick Wallace, CEO and Brent Higgins, CFO. So before we start, just quickly, a safe harbor. For important disclosures, please see Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures.
If you have any questions, please reach out to your Morgan Stanley sales rep. And I think KLA has a safe harbor as well, and then we'll get started.
Hey, Joe, good to see you. Thanks for having us. Very quickly, we're going to make some forward looking statements today, and those statements are subject to risk. Those risk factors are outlined in our SEC filings, and you can find them on our website. So Joe, I'll just take a minute and just characterize the overall business environment here.
It will lead into some of your questions. We had a nice finish in the fourth quarter to calendar twenty twenty. Of course, 2020 had a lot of interesting dynamics. So a lot of them not great from a personal point of view, but certainly from a business point of view, we had a nice finish to the year. In the fourth quarter, we saw some improvement in the memory part of our business.
So the second half largely driven by foundry logic, but some improvement in breadth in the memory part of our business, which gives us some confidence as we move into 2021. We finished 2020 at 15% revenue growth for the company and 31% earnings growth, so really strong performance for the company and we're pretty proud of all the work that our teams did to make that happen. As we look at 2021, I think it's just taken a bit of a step back. If you go back to just in the November timeframe when we were at earnings, we saw the 2021 environment as being mid single digit, high single digit type expectations for WFE. And certainly over the three months to four months to where we are today, we gave guidance and talked about an environment that was mid teens plus or minus a few points at earnings back in early February.
If you look at just the general momentum that we've seen in the business, it's probably biased to the higher end of that range today versus the lower end. So certainly, you look at that type of growth, expect the foundry logic part of our business to be mostly in line with the market after a faster than market growth year last year, to DRAM to be faster than the overall market performance and then flash to be a little bit less. So of course, things change a lot as we move through the year, but at least that's how we see it today. If you look at our service business, our service business is had a very strong year in 2020, and we would expect that to continue to grow in line with our long term historical target of 9% to 11%, but probably biased towards the higher end of that range. If you look at the demand for semiconductors and how broad based it is, really stratified across a much broader set of end markets, parts of that business drive our service business.
And about 75% of our overall service business is subscription like contract. And so I think that because of those dynamics, we'll see that business continue to perform pretty well. And then our electronics component packaging part of the business, which is the more than more part of KLA, a big chunk of that coming from the Orbotech acquisition should have a double digit growth year as well on top of a pretty good year last year. So overall, we talked about an environment where we could continue to deliver double digit performance for the company into 2021 and performance in terms of operating leverage that's ahead of our long term target model, really driven by the overall strength of the mix of our business being more semi process control heavy, but also the overall operating leverage from the strong volume increase that we've seen overall. So pretty good environment and I'll leave it at that and happy to take your questions.
Yes, it's a really good environment. And by the way, for the listeners, you can type questions on the webcast and I'll ask those as we go. When you talk about this, there's a lot of movement from year to year, but I'm really trying to wrap my head around this idea of WFE pushing 70,000,000,000 my guess is exceeding $70,000,000,000 before all is said and done this year. And I sort of think back to five years ago when it was about 35 and then there's lot of hand wringing over as next year, 37 or 38. That big picture move from 35 to 70 seems pretty important.
And that's in the context of semiconductors growing about 50%. So good growth in underlying semiconductors, but really good growth in capital intensity and WFE. Are you guys thinking about that? Is $70,000,000,000 coming out of recession, is that a baseline we can build from? Or is that where do you think the long term range of WFE outcomes is going be from here?
Yes, Joe, I'll start and then ask Brenda to join in. I think it's a great question. I think if you look back over time, we kind of viewed it in two ways. One, there were a number of years where semi growth was continuing and there was no increase in WFE. And that was driven by a couple of factors.
So we kind of got off the traditional trend line where you would expect capital intensity to roughly keep pace with the growth of semi. And we went for quite a bit of number of years where that didn't happen. And I think there are a couple of things that were at play then. And now I think we're doing catch up. But two of the main drivers, think, one, there was a lack of real advancement in terms of design rule capability.
So essentially scaling stalled for a number of years. And what we did was we got some benefit out of multi patterning and some other things, but there was no real node driven by the traditional scaling of devices driven by EUV because EUV was so delayed. So part of what's happened, think, is as you've seen there be an increase in capability, I think we're pretty much catching up with the historical trend line of capital intensity. The other thing that's happened, I think that's also maybe a bit of a surprise. And if you went back in time, what a lot of people were predicting about the industry was that the advanced nodes, there would be fewer and fewer designs because of the affordability issue.
And that kind of went the other way. And even with our customers, they were surprised by the broadening of the number of advanced designs and the fact that many players that used to buy their semiconductors have started to design their semiconductors because that's driven additional capability for them. So part of what happened was if you think back to say, for example, the 20 to 16 nanometer nodes, there was a lot of reuse in that because there was really only one device that was driving that particular node. And when the world moved to the next device, they were able to backfill a lot of that. What we're seeing now, which is impressive is a broadening of demand.
We saw memory also go backward in that same time frame. If you think about three d NAND, the whole world shifted backwards in terms of design rules in order to do three d NAND and that had a little bit of a holiday for the semiconductor guys. But as that has caught up, those things have gotten more. So I think if you look depending on the arc of time you look over, it's kind of understandable that we would have caught up. And then you say, what's the composition of the demand now and how sustainable is it?
It doesn't seem to be much speculation. I mean, it seems like multiple devices driving designs. We have the first year in a while where handsets are growth and that was largely waiting for five gs. So you have a number of inflections happening. I also think that we had some growth and then as Bren said, COVID has been horrible for the world, but it definitely has accelerated the digital transformation.
There's no question, witnessed this conference and others that, that trend has been accelerated, which has been driving in a year where I think a year ago, a lot of people were uncertain what the semiconductor industry was going to be like in 2020 and certainly the equipment industry. So those are a combination of factors. Now you've got advanced design rules, multiple devices, multiple players. You've got strength returning in memory this year. And so all in all, it feels like the other thing we look at, Joe, is the profitability of our customers is very high.
So they can afford to make these ongoing investments as they in fact, some of them are fearful that they'll fall behind if they don't keep doing it. So I think those are kind of the trends. So I think we've kind of gotten back to a traditional expenditure in terms of capital intensity that maybe the industry was used to. So I think more of the anomaly was the four or five years where we had flat WFE in spite of growing semiconductor content. And finally, the proliferation of devices, you saw it and we all saw when the industry hit here, it seemed rational because the concern with COVID was nobody is going to be buying cars.
And you look at it now and there's a scrambling for automotive semiconductors and we've got literally lines down and we're not talking about EUV EV cars, we're talking about every car. So I think the proliferation, we're finally at a place where semiconductors have become much more ubiquitous and the economics are kind of working for everyone in the value chain. So we think this level is sustainable. We see a lot of strength. Our biggest challenge, as Bren mentioned, is right now is support of the supply chain to be able to we all learn how to do installs and things without people traveling as much in the last year.
But our customer and our profile for the calendar year looks very strong as we look out based on business that we've already booked and what we're anticipating. And Brenna, if you want to add some color to that?
Well, it's just as I said, the demand, I wouldn't say it was it surprised maybe I would say surprise is maybe the right word is the way it changed in the last three or four months. And both in terms of the magnitude of overall investment, but the slope was very steep in terms of the increase. There is a lot of pressure in the supply chain in terms of the ability to get components to support this activity. And then you've got lead time dynamics that are also challenging to manage across our own products. So we're there's a lot of focus in the company today on execution, both as Rick mentioned in the field, but also in our factories to scale up to support what looks like a very strong demand environment.
Great. That's helpful. And I guess as I think about this WFE environment, I guess the other factor has been this kind of protectionist view of the world where there's sort of China building sovereign capacity to protect themselves. You have now The U. S.
Talking about executive orders to emphasize the importance of semi hookers. It seems like that's a pretty good thing. At the same time, it seems like capital spending is largely going to be in support of demand rather than thinking about geographically. So how do you think about those geographic tensions as sort of helping the business, hurting the business going forward?
Right. It's a great question, Joe. We do look at it. And I think you're right. Ultimately, over the long period, you're going to rationalize that.
You don't have additional capacity if you don't have the additional demand. However, just in the same way, there are these bubble costs and this bubble investment to ramp up a new facility. And let's say, the Arizona project that's been talked about a lot. There is one where there's going to be some inefficiencies, so additional spending to bring that on. And one of the advantages for most of these players has been scale.
So eventually they'll have to get to scale on these new projects. And we've heard, of course, Samsung has talked about a new U. S. Fab. We talked about what TSM.
But in the end, there'll be some incremental investment over the next few years to ramp those. But ultimately, it's going to match and align to the end demand. China is a little bit different because what they're doing is what we used to call strategic. And so there may be some level of investment that they're counting on taking some market share. What's interesting is for the first time that share isn't at the leading edge.
That share is more a little bit lagging, which is a different dynamic what we've seen in the past. So when we attribute the overall WFE, most of it is 90% is supported by demand, and then there's some other projects that are going on. It's good for the big global equipment companies. Frankly, it's always been the case that a globalized customer base is better for the bigger equipment companies because those customers demand international support at the same level. They actually want people that are experienced in their home fabs to be located in the new fabs.
So if you're a smaller equipment player, it's harder. So for a company like KLA, that's a net positive to us. I do think there's going to be some challenges in the labor market because getting people to staff these facilities is not trivial. And it's not like we have an abundance of that talent in The U. S, but that's a kind of a different level problem, not really so much for KLA, but some of the challenges for them.
So I think in aggregate, it's not a bad trend. I do think that there'll be some period of time where you'll see a heightened level of investment as they transition to those new facilities. And I think we'll see them becoming ultimately scale sites. I don't think they'll leave them subscale because the economics doesn't make sense.
Okay, great. And as you think about process control as a percentage of WFE, it looks like to me you've roughly kept pace over the last couple of years, three years. It seems like there's an awful lot of importance being put right now on yields. At the cutting edge, you talked about the new entrants coming in. They're going to have to focus on yields.
And when you think about the issues at Intel or the foundry battle between Samsung and TSMC, like all of that stuff is very yield sensitive, process control sensitive. So I would think that would be a good lead indicator for process control intensity as a percentage of WFE. But what do you guys think over really the next couple of years?
Right. So as we laid out in the Analyst Day that we had in 2019, and that was September, the thesis we laid out there still seems to hold. What was different is WFE has accelerated faster than what we modeled. But we said we would gain some our view is we would gain some share and there'd be some intensity increase. And it was relatively modest.
We didn't think it would be huge, but by driven by a couple of things. One, the technology drivers, we've seen that with the increased implementation of EUV has been a catalyst to more process control intensity. We definitely have seen that. And we think even some of the memory, our position when we first went to three d NAND, that was a very relaxed Moore's Law node, if you would, when they first transitioned. But as they've caught up and as DRAM is reinvesting, those challenges continue to be the present.
So we do think that that drives process control intensity. The other thing is the proliferation of advanced designs is helpful to us because if you're running a fab with multiple process flows, multiple designs, you do need more control. So that's been a positive relative to process control intensity. So as we model that, over time, we do see that edging up. We've been relatively, I think, modest in terms of the percent that it was going to grow.
We actually gained more share in the first year than we had modeled over time. We had two fifty basis points. And so that was not we'd said about 50 per year. And we've seen some indications we're going to have higher process control intensity as we go forward. But we also have new products to get out.
And if the one tailwind we had a little bit with COVID is it's taken a little longer to get the new products out. There is more demand for process control intensity in some segments than there is capability. And we have those products being developed, but they're not fully in the market yet. And we think that will be additional catalyst to growth. We're going to see some of those later in calendar 2021.
And then in 2022, we'll see more rollout of those new products. So we think we're well positioned for it. And in terms of outgrowing, as you know, it's somewhat dependent on the memory logic, memory foundry mix, but we feel pretty good. And Brent, if you want to add some color to it because you're closer to those numbers, obviously.
Yes. So Joe, in our model for 2019 to 2023, we had about 70 to 80 basis points of WFE share. Everybody has a different denominator on WFE. So I'll just use that as the baseline. And I think one thing that's been encouraging that we've seen here is that most of the contribution has come from just the end market drivers driving foundry logic investment, driving design starts, lots of design flows through foundry logic that's been good for our position.
We have to execute. We have a number of new products that come to market that we think for the first time will give us some an opportunity to improve process control intensity in a more meaningful way. A lot of it around the reticle and the reticle ecosystem with multiple products supporting that new e beam capability that leverages the portfolio of the company to drive more relevance to our optical inspection systems and then improvement opportunities with new technologies and memory. So the good thing is, is that what's been driving the business so far doesn't include much from those products. And I think what's incumbent upon us as a company is to seed the market and start to drive contributions over the next couple of years from those opportunities.
Yes. I mean it seems like the WFE focus is sort of one thing, and it's the way people frame all of this stuff. But it seems like you have pretty good visibility into a lot of people needing process control across all your end markets. So it's a pretty healthy backdrop. I do have one question from the webcast.
Competitive landscape now, you've got Applied Materials talking about gaining in optical inspection. I think they said 44% growth last year for their new tool. You have laser tech with actinic, multiple Korean players. For I guess, twenty years, I've sort of had hand wringing about KLA's competition in optical inspection. But can you maybe talk about that?
Are you surprised by any of the competitive dynamics? And then maybe give us a little bit of a breakdown both in optical and the importance of EV?
So I'll take the first part and then let Bren fill in some of the numbers. So we have always as you alluded to, we've always had competitive pressure in our markets. We would fully expect that for two reasons. One, there is a perceived increasing need and there are a number of players in segments of our market. So we're by far the largest player.
As you know, we're four times the than anybody else in process control. But we have segment competitors in every one of our we don't necessarily in the long term view that as a bad thing because that drives innovation, that drives our teams to continually focus on providing more capability. I've often said that the biggest competitor we have is relevancy to drive the value of inspection and measurement. Most customers, if they had their way, would of course not spend money on inspection measurement, they'd rather spend it on capacity, but not if they can't yield. And so the balance has always been on us and the onus is to demonstrate capability.
We invest very heavily in R and D to make sure we have new capability when that's coming out. All of our competitors come out from a point product position where they'll have a specific answer for a specific niche at a time. And our view is our portfolio is actually what our customers really want is to be able to optimize their inspection and measurement investment in the things that will drive them to the next capability. So we gained share in a period we've held and gained share since the Analyst Day. We've been successful in driving share through our strategy.
We have a great product road map. We do expect there to be continued competitive pressure, but we also think that customers based on the engagements that we have today, we feel very good about where we are competitively. And even when there are alternatives out there based on their desire to get our next generation, we feel like there are certain gaps to the market that they know we can fill. And so we'll continue to do that. I've heard years ago, I mean, it wasn't that long ago where the people that are saying they're gaining an optical inspection, we're saying optical inspection was dead and it was all about e beam.
So from our standpoint, being able to continue to innovate in optical, it's part of the bigger story, which is the best capability that we can bring to our customers is the most effective cost of ownership to solve the problems that they have. And that's typically driving to higher capability, higher throughput at lower costs, whether it's optical and Bren mentioned we have e beam products and we have what's considered it's now very advanced, but even the laser scanning versus the broadband plasma. So the same thing is happening in reticle where we provide a portfolio and you mentioned one of our reticle players, but in aggregate, we feel really good about where we are. We think if you really want to test how we're doing in market share and in market competitiveness, look at our gross margins. You would think that if we were under a lot of pressure competitively, we wouldn't hold gross margin.
That would have been the traditional way you would check that. And if you back out the percent that our service businesses and know that we bought EPC, you can see we're doing well competitively. And that's been true over time that this noise has been out there. Our margin and share has continued to grow. So that's just some of the data.
So we feel very good about where we are. And Bren can maybe provide a little more color on the product specific questions.
That was pretty good, Rick. I think it's important, the portfolio approach that we have to solving some of these customer challenges and providing our customers the flexibility to optimize for their cost of ownership and time to results. We specifically in that market you referenced, we cover that space with multiple platforms of technology that balance that need between sensitivity and throughput. And because of the introduction of EUV and foundry logic and the scaling dynamics that are part of that, we saw similar levels of growth in that part of the business. So if you're in optical inspection, given the inflections related to EUV, it's a good market.
And it's a good market after a number of years when EUV was delaying where it was a little bit less relevant. And so it's in some ways to the earlier discussion a catch up, but also the introduction of EUV has created a big opportunity to drive relevancy of optical inspection. And then we're doing things moving forward, including e beam capability to drive more relevancy in the optical inspectors by using e beam tools to train those systems. So I think from a competitive point of view, we feel very good about where we're at. We're continuing to invest.
We invest in our business at levels of our competitors' revenue, in some cases or most cases in process control. So we're comfortable with our roadmap. And as Rick said, our gross margins speak for the differentiation that our products provide.
Okay, great. That's helpful. Maybe you can talk a little bit about Orbotech. How is that going? Should we start to see growth in some of the areas like display that they're serving?
And then generally, as you've sort of integrated that acquisition, has that changed the way you guys are thinking about M and A looking forward?
Sure. So let
me start with how it's going and we can talk a little about Display and then the longer term. One, as we lay out in the Analyst Day and in the deal acquisition, our thesis around Orbotech and what is now the EPC Group was kind of this more than more area of growth of semiconductors, which has been talked about for a long time. But the packaging, the world around packaging and the innovation that was we kind of had a flavor of that with our IQOS business, just looking at the back end. But there's more technology that's coming in and more and more of our leading edge customers have advanced packaging efforts that are following. For a while, there was really one and now I can count at least five.
And our engagement at those levels is higher than Orbotech could have achieved on their own because those are the same customers we deal with in the front end. And what they wanted was somebody that they could work with that would have the resources and that they could hold accountable for the results. And that is a message we send that team. Part of that business that we get is this expectation that we'll deliver. And so we feel good about the opportunity.
We have some advanced development going on, leveraging some of the historical KLA technology enable some of the capability that is going to be needed in packaging. So we think that's a driver, a growth vector that's going to be faster than the rest of the semiconductor industry because of the increased investment that's going on. We think that there's a long way to go there, too. I think that we're early days in terms of some of the work and we've realized the cost synergy. We think that there's a lot of expectation around revenue synergy that we haven't seen yet, but we think will happen in the next couple of years.
I'm particularly excited about some of the interaction we have between those divisions where we see that there is capability in one that can be applied to another. And I think we're pulling that together. Part of why we think it works, and this was our thesis going in as the KLA operating model, that was the question we had, and we feel like we've answered that. The systems and the way that we've run our business historically have been mapped now to the Orbotech businesses, and we feel very confident that we're driving results as a function of that. One of those is, for example, this concept of focus on gross margin.
We talk a lot about differentiation as a company, and we've historically always said, you can say what you want, but look at the gross margin and that will give you an indication of how differentiated your products are. So I'm happy to say we've made improvements in there. We never reviewed the display market was one that we didn't that was not the basis of the acquisition. We thought that was opportunistic if we could drive some capability. There are some leading edge technologies that are interesting in that.
I feel like the team has done a really good job of dealing with the cost basis of that business. And we think there's an opportunity for some growth. But I have to say it's a very small percent of KLA today. And while I think there's some upside to where it's operating at today, that was not part of our thesis. The last thing I'd say is to your question on more M and A, it's not something that we feel we need to deliver on the results over the next several years, but we certainly are going to be opportunistic about looking at opportunities, especially given the success that we've had of applying the KLA operating model to these businesses.
In a period of COVID, doing integration across regions, we feel like we've tuned that up. And so we will look at that as another way to deploy cash, but we always measure it against, would we be better, would our investors be better if we bought back our stock? And that's the measure that we're going to always have is what's the best investment of that capital that we're generating, whether it be buybacks, dividends, acquisitions, that's how we weigh that. So Bren, I know you've done a lot of work on the details of it, so I'll let you weigh in here.
Well, one aspect of our thesis on this was that we could take market leading positions and get more out of them in terms of incremental leverage on the business. And so the Orbotech transactions allowed I think demonstrated that we've been able to do that. Every business has different margin profiles. The margins of EPC are closer to 50% than where they are in Semi Process Control, which are 64% plus if you look at our segments. But the incremental margins on the growth are consistent between the two businesses.
And our business model of delivering 40% incremental operating margin, 40% to 50% is our objective in that business. It's our objective in the semi process control business. And if we can do that with our capital deployment activities, then we should be able to drive earnings per share growth that's 1.5 times our revenue growth rate. On the synergy front, we talked last quarter that we think we're probably around $80,000,000 We had publicly disclosed a goal of $50,000,000 So we've seen considerable growth in our expectations for synergies. And I would think as we move forward, as we start to do more work around facilities and put systems in over the next year and a half or so that there's more upside there.
So I think the applicability of the operating model to all aspects of their business from product development to channel and customers' customer management to general just operational effectiveness is playing out. And we're just getting started on opportunities in and around the service business. Also part of the thesis was can we take businesses that have complex systems integration in a service tail and be able to drive stronger service performance. It's been great for KLA. It's driven a level of predictability overall.
And we certainly have the infrastructure worldwide to be able to support broader revenue stream. So that's an aspect of the deal that is still forming. And yes, the go to market is a little bit different for different markets as it relates to service, but certainly the infrastructure and efficiency opportunities underneath are opportunities for us as well.
Okay, good stuff. Unfortunately, we're out of time. So I had a couple more webcast questions, but we'll follow-up those by email. But thanks guys for participating. It's really helpful and we'll talk to you soon.
Thanks, Paul. Thanks.
Thank you, Joe.
Thanks for having us. Thanks, Paul.
Thank you.