Lam Research Corporation (LRCX) Earnings Call Transcript & Summary

We are now live.

Great. Good afternoon, everyone. Thank you for joining us today. I'm Stacy Rasgon. I cover the U.S. semiconductor sector here at Bernstein. A few housekeeping issues before we begin. First, just for your reference, I wanted to call your attention to Lam's safe harbor statement, which is currently on screen, which covers any forward-looking statements that may be gone over in today's discussion. Secondly, I'd like to say a little bit about how our interactive Q&A is going to work. We're using a service called Pigeonhole. There's a link on the side of your screen to access it, it says Pigeonhole. When you click that link, it will open up a new window in your browser. The video will continue in the prior browser. You can submit your own questions through Pigeonhole in the box at the top. You can also vote on questions that have already been submitted by pressing the up triangle next to any one of them. So please do go ahead and click on that link now. You can submit questions for Lam beginning right now and at any time during the presentation. And some of this may get folded in. We'll also leave out time for Q&A at the end of [indiscernible] Finally, we're working with our partner Procensus to do live polling on the presentation after the session ends. A link to Procensus is also on the left of your screen. Again, this will be a live poll with immediate access to the results. For those who choose to submit, so you can get sort of real-time feedback on sentiment on Lam. Please do take 60 seconds after the presentation to fill it out. So with all that out of the way, it's my great honor to introduce our guest today, the President and CEO of Lam Research, Mr. Tim Archer. So the semiconductor capital equipment industry's been enjoying a renaissance over the last many years. And even given some of the current cyclicality, I'd say, things look good. Total wafer fab equipment spending remains well above historical peak levels. I'd say the much anticipated end of Moore's Law, far from reducing the need for customers to spend on equipment, seems that it , in fact, made contributions from companies like Lam more important than ever as materials-driven innovation moves to the forefront of process development -- process technology development. I think there are obviously some near-term controversies around the trajectory for wafer fab spending given the current COVID and other situations. But over the longer term, I do think there are controversies around the changing dynamics that could allow the industry to remain a secular rather than cyclical grower. What continued innovation in memory might mean, especially for a company like Lam, new opportunities and new applications for leading etch growth, and of course, the consequences of some of the recent geopolitical shifts. So to answer these, and hopefully, many other questions, it gives me great pleasure to welcome Tim to our session today. Tim, thank you so much for being with us today.

Pleasure, Stacy.

So in this format, I do like to stay away from near-term things. I will not be able to quite do that 100% given where we are [indiscernible] . But before we dive into that, I do want to start higher level and more open-ended. So you've been CEO at Lam for about 1.5 years. You've been with the company since 2012, since the Novellus acquisition. You've been in the industry more than 25 years, so you've seen a lot. You've seen a lot of cycles, a lot of different trends. So, I mean, near-term things aside, I'd say that to me at least, the structural attractiveness of semi-cap has continued to improve through all of that time. I think the value-add of the industry, 2 semis, seems to be growing by leaps and bounds, especially as Moore's Law is getting more difficult every year. Cyclicality, it's still there, but it's certainly reduced, it's less. Companies are profitable, not quite profitable even at the bottom of the cycle. And I would say everything seems just, in general, broadly healthier.

I was hoping you can give us just a minute or 2 of your thoughts on where semi-cap has come during your career, where you see it going and at least at a high level for a moment, and then we can dive in, like how you see Lam fitting into that story.

Okay. Well, great. Stacy, I appreciate the opportunity. And I think it's great you pointed out just how long I've been in this industry and how many different things I've seen. It has been a long time, and the industry has changed a lot. I mean it's a -- for -- in many ways, for the better, as you just described, regarding profitability and perhaps less cyclicality, and we'll, I'm sure, dive into some of those things for questions today. But maybe I just wanted to start, I think it's a very fair game to talk about current events. I mean, they really are occupying a tremendous amount of management time right now as we think about keeping employees safe and restarting all of our production capabilities. But I did think it'd be helpful to start today's discussion with a little bit of context and remembering that it was really only about 2.5 months ago that we hosted our Investor Day in New York City. And at that time, obviously, none of us knew exactly how this would transpire over the next 2.5 months. But at that time, we laid out a little bit of history of Lam's recent performance kind of 2013 through '19, which was a period in which Lam outperformed and outgrew WFE by nearly 2x. And we did that by executing on a multifaceted growth strategy that combined product competitiveness to gain share. I think some very strong capture of very important technology inflections like multi-patterning, 3D NAND, a strong focus on etch and deposition. Really, a decision we made to keep most of our R&D attention squarely focused on what really have become 2 of the fastest-growing segments of WFE. And then not to forget, really a focus on growing the revenue we could derive from our installed base. And that allowed us to grow quite rapidly until 2019. 2019, right after I took over as CEO, we got the first downturn in the industry in about 7 years. So there was one challenge I got to face. But the reality is that the structural change is not only in the industry that you talked about, but also in Lam, in our operating model. And a lot of the work we had done while I was COO over those years, Lam delivered in that down year, a year in which memory spending was weak. We delivered the second best EPS in the company's history. And so we sort of joked about it, but it's like if that's as bad as the downturn was, maybe you're right, maybe cyclicality is not what it used to be. But it was not a bad year, but we came into 2020 really optimistic, not only about the future of Lam's growth story, but also about the fact that the memory tool is going to come out of a maintenance spend year, memory spending will be up this year. And we put forth a plan just 2.5 months ago that said between '19 and 2023, 2024, we could grow the company another 50% to $15 billion when WFE would be about $60 billion. And really -- doing many of the same things. We launched a new etch platform in March. That's really -- we think we'll take equipment intelligence and the use of data to kind of another level in terms of helping to drive technology and productivity in the most advanced fabs. We talked a lot about the launch of our new enhanced ALD product that we think has the ability to disrupt older markets, things like PVD and CVD and furnaces and spin-on films with much higher quality ALD films. We think that will expand our market, served market. And then we talk about growing our installed base business more than 40% between now and 2023. And so as we talk today about a lot of current events and some of the uncertainties around COVID-19 and China trade, I just want to make the statement, we don't see any real change in that long-term outlook or our opportunity. And in fact, what I would say, many people are talking about, work from home and many of these items, and even in our case, remote support of customer fabs, is really enhancing the importance of a lot of the technology that we're working hard to build and enable. So I think that while a lot has changed, one thing that hasn't is our long-term outlook for the growth opportunity of the company, so.

Yes. Got it. That's helpful. In near term though, I know you've talked about -- yes. I know you talked about -- I mean, a couple of things. You didn't give guidance, but you did give a view of where you saw things were going. And that view at least seemed reasonably positive. I know you talked about both supply as well as demand disruptions. Maybe could you talk a little bit about just what you're seeing now in terms of the order patterns, maybe both those supply and demand disruptions and just in general, what you're doing about your supply chain in order to kind of make it more resilient and keep it robust?

Yes. As I first said, I mean, when the -- as we saw the situation worsen there towards the end of the first quarter and ultimately culminating in shelter-in-place orders going in, in California and Oregon, that's when we withdrew our guidance because we realized we had to put the safety of employees first. We stopped manufacturing for about that two-week period while we relaid out our manufacturing, obviously, implementing a lot more controls around social distancing, PPE. We did the same thing with our critical suppliers. And I would say we've definitely seen improvements since that time. One of the -- maybe the most important statement is that our manufacturing and lab operations around the world are up and operating, albeit not at 100% pre-COVID capacity. And why that is, is because within the same space, our implementation of social distancing, to ensure we are keeping employees safe, means people are further apart. And so they're -- and some of the workstations are further apart. And so the output we can get from the same space isn't quite what it had been to that implementation. But it has come a long ways. And what we said was that over the next several months, we're making changes where, what we're really doing is using additional space. We're expanding into what was previously office space and some storage space and making more manufacturing area, more gown rooms to get more people in. And I think that we're feeling very good right now about our progress. First goal was to stabilize output, get tools to customers and then now ramping up to meet what we see is strong and growing demand still in the near term from customers. And so not only do we have to get back to where we were pre-COVID, but we anticipate having to get to even higher output levels in months to come. So that's something that the team is working very hard on.

What do you think is driving that demand? I'll be honest, I've been a little surprised at the magnitude. And you're not the only ones that are seeing that even the like of all of this. Are you seeing any evidence of pull forward or stockpiling? Or do you really believe that it actually is sort of true structural [ antimatter ] would you be able to tell the difference?

Well, I think that the answer to the -- would we be able to see pull forward, we would be able to. We'd go back and we set as our reference, the conversations we are having with customers about investment plans 6 months ago before anybody sort of knew about these things. And for the most part, I would say we see customers performing right through those plans. You'll recall at the January earnings call, we said mid- to high 50s for WFE. And that was strongly up from '19 with a lot of bias towards memory. And people were surprised maybe by that strength but we're seeing customers, at least in the near term, executing, for the most part, to those plans. Now we also said now on our more recent earnings call that supply constraints and other things could bring that down a little bit from the mid- to high 50s to something at least in a scenario we're considering, low to mid-50s. But it's still strongly up from 2019 and I think that's just because there was underinvestment in memory in 2019. We exited the year with customers really needing to not only put on capacity, but continue to invest to move technology nodes forward. And I think that with most customers, just as I talked about us kind of looking through this to our long-term objectives, I think most customers are doing that as well. And that would be my best guess as to why we're still continuing to see strong impact.

You don't see that just as a memory statement, but you see that across all your customer types?

We see it across all the -- all types. I mean, I think that we have been looking for where we might start to see that. Obviously, we had expected the biggest year-to-year shift in gain to come in memory, which is I think why we've been looking at that one most closely. Foundry had been strong, and we said even in our January call that we expected it to remain strong into the first half of the year and that's [ what we continue to see ] .

One more quick question on COVID. This is a question that actually, they're asking us to ask every management team.

Yes?

As you think sort of like through and beyond the pandemic, how are your priorities shifting, if at all, especially as they relate to things like cutting costs through increasing levels of investment or the areas where your investment -- like what kind of like changes are you making to your priorities in how you're running things that may be more permanent?

Yes. So it's something we're giving a lot of thought to. And you'll have to understand how Lam -- what our strategy really has been over the last few years. And part of that is a more distributed manufacturing, more distributed R&D capability globally. And it wasn't -- at the time we envisioned it, it wasn't because of a global pandemic. It was because we're a strongly customer-facing organization. It's kind of how we try to drive our business and so we announced at the end of last year, for instance, a new technology center we were going to build in Korea to put some development capability right next to a couple of our biggest customers. We announced a large manufacturing facility that we're building and we'll open middle of next year in Malaysia because, again, the majority of -- vast majority of our business is in Asia. Our supply chain is there. We can take some costs out. So I think what you're -- what we found through this was that regionalization, strong regional capabilities, actually are helping us -- helped us respond. Because different regions actually got hit at different times and also recovered at different rates. And so I think we will look for smart ways to distribute our capabilities a little bit more globally. And I think the other thing that we will do is we're investing in what will definitely stay permanent is, we're investing in a lot more remote and virtual support capabilities. That's everything from -- we used to build physical training centers, and we would send people to those physical training centers. We now have completed construction of virtual training centers around the world where we use VR to train engineers on how to do maintenance on tools. It saves tremendous productivity, time and travel from those regions and also saves us a lot of capital from having to stock up these training centers. So VR -- now we're starting to actually get customers to allow us to use some of these connectivity and support products inside their fabs. And that used to not be -- because of IP concerns, that was always really a difficult hurdle. But because of the challenges of COVID-19, customers started -- have started to open up to that and I think they're now seeing also the power of -- instead of us having to put a person on a plane who will show up in the fab 2 days later, you can remote in, and in fact, we can guide some of our on-site people to do very advanced tasks through remote supporting. So I think we will continue to invest in those types of activities. And that's -- it's better support for the customers, and it's also lower cost ultimately for us. And so...

Got it. That makes a lot of sense.

Those things will continue.

So you were talking a little bit about different regions. Let's talk about one of those regions, China. So obviously, there have been some shifts and further shifts in the geopolitical landscape, and actually, I would even go to the Pigeonhole question that's the top one on the list. Commerce Department rules, Huawei, maybe we take the Commerce Department rules first. In particular, I think the military end-use restriction, to me, seems to be at least to have the potential to be the most disruptive and it's the first time that I've actually seen the Commerce Department specifically call out semiconductor capital equipment directly in the press release and in the regulation. So how do these rules impact you? It's -- you have a little bit of time now to evaluate them. How should investors be thinking about, I guess, maybe the current, like, direct impact and like a cost check for impact as we -- as these things evolve.

Yes, it's -- it unfortunately, probably like, I would imagine most of the companies so far unable to give a specific quantified answer. But what I can say is it's been helpful that finally the rule has been finalized and published, and as you said, the military end-user rule is the one that places an obligation on the equipment suppliers to determine whether or not we need to apply for a license to ship to certain customers. And therefore, we're working right now through a process. We have to complete that. This rule goes into effect on June 29. So we have to complete that process. But it includes our own diligence on those customers . It includes us turning to some third-party input as well to provide us additional input as well as the customers themselves certifying that they are not military end-users. And we're in the middle of that process right now and so I really can't comment on kind of what we'll see but we're making good progress through the process and I would anticipate that no problem was -- for us to arrive at answers here in the next short period.

Okay. How much of your manufacturing is U.S.-based versus international?

We don't disclose exact percentages, but we have large manufacturing facilities in California and Oregon, but also a large manufacturing facility in Korea, manufacturing in Taiwan, in Austria, and we just said building a very large facility in Malaysia that will come up middle of next year. So from the standpoint of, if somehow global manufacturing plays some role in this, I mean, just our strategy we've been employing for many years and the one I just talked about, global distribution of resources for business continuity, I think, we're well equipped to satisfy demand from wherever it's most appropriate to satisfy it from.

Got it. And then the final, final question, obviously, Huawei. Any thought on how that through TSMC or through SMIC, demand potentially migrating from Huawei to other customers? And just how does -- how do you see that impacting you directly and maybe indirectly?

I think that, again, these -- we look at them as shorter-term perturbations, if there is any, that directly affects us from the standpoint of how it may affect those customers' demand. And obviously, it's a much better question for those customers, how much of their ongoing demand is within those very specifically defined, produced or developed by Huawei, as it's kind of defined. It isn't -- it does segment it down to certain types of devices. And I don't know that, that's the vast majority of any of our customers' business at this point. But obviously, there could be some short-term demand as the supply chain redistributes in other supply parts or products that might have been produced for or by Huawei. So it's not -- again, coming back to -- in the context of our '23, '24 model, we don't see it as a significant impediment to continuing to execute to that model.

Okay. That's good. Let's move on.

Yes.

... from there. So I want to talk about wafer fab equipment spend and capital intensity. So we talked a little bit about this last year when we...

Yes.

But I'd love to hear your kind of your most recent thoughts about the prospect for structurally higher wafer fab equipment. I would love to hear your view on drivers, where there's capital intensity or capacity versus technology spend node migration. Drivers on memory spend, in particular, I want to dive a little bit more directly into memory. But just more broadly, what are your thoughts on drivers of capital intensity increasing in the industry?

Well, I mean, capital intensity, it is -- it's going up as a result of just, obviously, increased complexity of device manufacturing. Structures that we are talking about, whether it's on the foundry/logic side and you're looking at the 5-nanometer or 3-nanometer device, you're looking at the types of equipment that have to be employed, whether it's, obviously EUV or now our -- obviously, Lam's very advanced etch and deposition ALD equipment. Capital intensity is rising. But fighting against that, actually, what I would say is rising capital intensity isn't necessarily as good as it might sound in the long run. There is an affordability element to this. And I think what you've seen is when processes like, let's say, 2 dimensional scaling, whether because EUV wasn't available or the alternatives were more expensive, I think you're starting to see the technology road map slow down and kind of log jam. And now that you're getting more affordable, more production worthy solutions becoming available, you've seen scaling speed up again. And so Lam's interest is and always will be finding a way that blends both technology and productivity together to make the customer and to motivate the customer and their end-markets to adopt more of it. And so in fact, one of the things you'll see is, you may have seen when we announced our dry resist project working with ASML and imec, one of the objectives there is to ensure that EUV, through the use of the right layers, whether they're masking layers or dry resist, is a highly productive manufacturing process that enables continued scaling. And so you'll get capital intensity improvements, but at the same time, there's some of us working very hard on productivity improvements to keep this -- keep the economics of the industry going. But generation by generation, you've seen it step up. And perhaps more importantly for Lam is not overall capital intensity but etch and deposition capital intensity. So I mentioned, we've been very comfortable putting a lot of our R&D and focus into that area because etch and deposition, when you're talking about building vertical structures, which is true -- every device now, whether it's logic or foundry or NAND, they're vertical structures. We need a lot of etch and deposition to manufacture those structures. And at every node, our SAM continues to increase. And so maybe the reality is, as etch and dep intensity increases, then maybe something else stays more flat, which is how the industry becomes affordable.

Sure. And we saw that in memory, obviously. And like to -- maybe to dive in there a little bit. So obviously, as the industry went from 2D NAND to 3D NAND, I mean, I don't know the capital intensity within the NAND flash must have gone off -- up by several multiples, I would think, over the last 5 or 10 years on that transition. How do you create the value for your customers to keep going there? Because, I mean, they're getting less and less. Obviously, as you went from -- this came up in another presentation, but you go from 32 layers to 64, obviously, you're getting double the output for doubling. You go from 96 to 128, it's less, it's 30% or something like that. And so you're buying less and less improvement for the same sort of like incremental slide of capacity. I guess given that framework, how do you think about drivers of NAND CapEx intensity, particularly given how much you play there and how you are continuing to sort of drive value to customers to make sure that they stay on that road map?

Yes. It's -- I mean, first of all, I mean, even where we have -- we own the application, we continue to drive tremendous productivity node to node in terms of wafers per hour output, where we can, cost of the system, but that's more challenging. It's much easier to drive productivity. Things like our self-maintenance that we implemented, where instead of having to have people come over and open the tool and replace a part and have that tool down for 24 hours every couple of weeks, we now have the robot replace that consumable too. It's done in a few hours, no people involved. One of our customers estimated that when they've converted their entire fleet over, they'll save on the order of 45,000 labor hours. And so those are the kinds of things that, again, we look at ways in which we can bring technology to bear on the cost problem and help the customer drive their cost down node to node. Because obviously, you can't sell the next generation of -- on a per-bit cost basis, it's worse, and we know that, and we do want them to transition. So it's things like throughput, it's things like productivity of the tool. And then with the launch of our new etch tool, as an example...

What was the name of that one, by the way?

Called Sense.i, which is kind of a play on sensors and equipment intelligence. And the idea being there, that if you can use the data coming off the tool and you can incorporate the ability to sense and respond within the tool and incorporate self-maintenance and other features, you can actually make these very complex tools. You can reduce the cost of ownership for the customer. And therefore, that tool is designed to bring both technology and productivity into the marketplace. And because it can be essentially configured to bring the right amount of technology and cost to any particular application, we've talked about that as an opportunity for us to do even better in the semi-critical space than we have in the past where productivity plays a bigger role. I mean, I think nobody will dispute that Lam today, we bring the industry-leading etch technology to the most critical applications. And often, if we ever are going to lose, we lose often where the tool doesn't quite scale down in caustic appropriately. But we think Sense.i will have -- will cover the full range better because of its focus also on technology and footprint.

That's interesting. What data streams are you pulling off of that, that you're feeding back into the process? Just out of true curiosity.

I mean it's -- there's hundreds of data streams available on our tool. But obviously, a lot of them are fairly simple things like -- I mean, to not give too much away, simple things. I mean, it's powers, pressures, gas flows. I mean everything that looks for -- it's feed -- not only feeding back into the tool, but one of the biggest challenges you have today in very complex manufacturing is that you need these chambers to match each other extremely well. And so there's a difference between being able to do an application and when you're at the maturity point where you're focusing on getting those last few die to yield at the very edge of the wafer, that's where the data becomes important. Data is not important just to do the -- the job correctly on one wafer. The data helps you do the job correctly on a million wafers per month or more, which is what we're processing through our tools through these applications. And so you want all million of those wafers to look very much like each other, even though they come off vastly different chambers in different fabs and et cetera, et cetera. So.

Got it. What scaling techniques, I mean, is it -- would you call this one the one that you're most excited about? I mean you've talked about dry resist as well. There's a number of things you've mentioned. Like how, what do you think is most exciting like for you as a technologist?

Well, it's obviously -- the new etch platform is the first new platform we've built for etch in about 20 years. So it's pretty exciting and the use of the data is really exciting. But I started out as a deposition person, and so I'm pretty excited also about the ALD, the atomic layer deposition that we talked about. And that product also is very exciting for the reason that, again, as I mentioned, bringing technology to bear on maybe old problems, right? And so the point is, I think people have known that things like PVD and step coverage challenges, things like a spin-on or SACVD with maybe some film quality challenges, those have been in the market for a long time and they've been good enough to do the job. And ALD has always been out there. And everybody knows ALD films are really high quality, but the productivity has always been part of the challenge. And so then what Lam has done is we've brought kind of the focus in on how do I enable this extremely capable film deposition technique like ALD in a productive form factor and platform? And so we've designed our ALD system so that it actually has productivity that is sufficient for customers to insert it. And they get good productivity and they get much better film quality than they were getting [indiscernible]

By productivity, you basically mean, like high etch rate with good productivity with good --

In the case of the atomic layer deposition, it's more about, in many ways, the structure of -- the layout of the tools such that you can pack in lots of deposition stations into a relatively small footprint. And so it's about -- in atomic layer deposition, it's a little bit harder to speed up the process.

Yes. Well, that's why I asked.

Yes. And so therefore, often the means there is through a more creative architecture of the tool. And there's a few other tricks that we haven't talked much about, but it has to do with the process that we call -- that helps to inhibit the surface during certain stages of the atomic layer process. So it's a -- anyway, technically, it's a very interesting and exciting process. And I don't know, we have a lot of those in our pipeline right now.

Got it. I'm interested. I used to be an etch engineer, so I [indiscernible] The question of -- the impact of the EUV always comes up. But again, you actually have a resist product that I think goes hand-in-hand -- maybe if you can elaborate a little bit more on how the drivers that sort of fits into everything?

Yes. Well, I joked that for at least 2 or 3 years, I was always being asked how will the introduction of EUV negatively impact Lam? I joked that I got so tired of answering that question, we decided to develop an EUV-related product. But did have a -- we had a great idea for how to use our deposition and etch knowledge to think about how can we put a resist down that is -- that helps to enhance the sensitivity to UV photons in a way that you accomplish the exposure at lower doses, which helps to speed up the productivity of the EUV process. That's good for everybody. And also to use the fact that you can do dry deposit and resist to essentially gain significantly more control over the process. I mean, you think about the spin on -- any spin on process, they've been around for a long time. And a lot of that is the -- engineering of the chemical, and then you basically end up spinning it on. And we talked a little bit about this. There's a lot of waste in that process, a lot of the materials are spun off. And so we just thought that, in general, if you can go dry, it's a better, more controllable process. We're still in the early stages of development here. I mean, we're engaged with kind of all of our top EUV transitioning customers, but we're a little ways away from that being put into full-on production yet, quite far away from that. But the promise is there, that the film properties are really strong and that dry deposition, you can control the process very well. So kind of in the 3- to 5-year time frame, we feel very optimistic about that. And in fact, at the Investor Day, we said it will build with time, but in the next 5 years, about $1.5 billion of cumulative revenue from that product is our expectation. So it'd be a nice adder to our portfolio.

Got it. Got it. So if I take a step back a little bit, like I used to tell people, especially like a year or 2 back, like when people are looking at the semi cap space and what I think of Lam. Like, the first question was always, well, do you like memory, do you love memory, right? That sort of [indiscernible] Lam isn't -- it's not just a memory company anymore, especially now where the foundry piece has been growing actually quite a bit. I was wondering if you could talk a little bit about how you sort of see long-term growth potential, both in memory as well as logic/foundry and how do you think you're positioned in each form today, as well as where you think that might be different, say, in like 3, 4, 5 years from now.

Yes. Well, I think that a couple of statements is, clearly, I mean we're -- we do love -- we love memory. We love foundry/logic as well. But it's memory -- the strength in memory has been a great foundational underpinning for the company. And it's something that still occupies a lot of our attention to ensure that we are satisfying our customers in that space and delivering what they need. And I think that if you look at the big trends over the next 5, 10 years, this point of data, storage and transmission -- memory's going to have -- I mean, memory, I think, plays a tremendous role in the data story. However, foundry and logic does, too, from a computation perspective. So we do want to -- we're stronger in memory, I think most people know that. But we talked about in the December quarter, record revenues for the company in foundry. We have made -- in foundry and logic over the last 4, 5 years, we've made significant progress in both market share as well as just the SAM expansion through our product development. And looking forward kind of next 5 years, I mean, memory is likely still to be our strongest suit because etch and dep are just so critical in those devices. But in foundry and logic, products I just talked about, ALD, strengthens our position in foundry and logic. Dry resist for EUV, almost exclusively a foundry/logic play for us. And so we are -- and what I want you to take away is, we're investing to improve our foundry logic business in a pretty meaningful way while at the same time, launching products like Sense.i, which will serve both, but really, the very first product it's launching for us to shore up and make sure our memory etch product positions are very well defended. So we have a multi-faceted product pipeline and growth story. And I think 5 years, the company will look bigger and a little bit different in terms of our mix across segments, but -- yes.

And I guess to build on one, right? If I take the kind of the drivers of growth maybe from your Analyst Day, like whether it was just growth in WFE -- services installed base growth, just the SAM, the amount that you're tacking going up as well as like share gains. I guess, what do you see is like the biggest driver around market share? Like is that a goal that you target? Is that sort of like an output of everything else that you're doing? I guess, how do I think about those growth drivers as we're looking forward?

Yes. It's interesting. So just real quick on market share because, I mean, I think your point -- I think you hit on it right, which is many ways, it's the outcome. I mean, obviously, we aspire to grow market share. We've committed to gain 4 to 8 points of share across etch and deposition. But that's -- I mean that just happens when you actually develop best-in-class products and you position yourself into the right inflections. And when you're in the fastest-growing segments of WFE, it actually also just happens naturally from the standpoint of you growing share overall as a company. And -- but within those spaces, we think we have a strong story there. But it's not something that we go out and we say, this year, we must, at all cost, gain share. We think about profitability, we think about the installed base, we think about the customer long-term relationship. So I would say that of those 3 drivers, I mean, share is one that will come through our execution, the SAM expansion comes through the product pipeline I just talked about, and the installed base comes from really just focusing on satisfying the customers' needs for all the tools they bought from us for all these years. We have over a 60,000 tool modular installed base, and we talked about how -- by delivering more value-added products, we've increased the capture of revenue per tool on average by -- it was almost -- it was about 50% from, say, 2012 to 2019. Between 2019 and 2023, we think we can take that up roughly from the 1.5x to 1.7x. And so again, it's just kind of continuing to build on this revenue capture by delivering more value to customers. So I think all of them are drivers of our business. And we actually have different people on my staff. I mean, each of those reports to me and each one drives with their own objectives to go build those pillars of the company.

Got it. I guess to drill a little bit actually into that installed base, that services business, so this is something that you've been talking about more and more. It's been growing. I think even in the down years, it's still been growing or pretty close to it. You said 50% over the last like 5 years in terms of revenue per chamber. Like how -- do you think about that in terms of like the lifetime value, like over the lifetime of the chamber? Like how does that -- do you try to grow that lifetime value? Is it -- what's the best piece of this, is it sales? Is it services? Is it upgrades -- like what's actually driving that kind of the continued growth? Like how do you look at that?

Yes. It's all -- it's kind of all of the above. I mean it's -- we've seen growth across spare services upgrades. Again, it depends on the customer and the device. I mean, we engage with the customer usually how they want to be engaged, more or less service. I mean, generally, people need spares -- we try to be very competitive on our spares' pricing and delivery and support. Upgrades are a great way for customers to make better use of the stuff -- the assets they've already got employed in their fabs. And so that's been a -- upgrades has been a strong driver as customers look to get more out of their fabs. So each of them has grown. I think that going forward, we are investing more in, again, back into these data solutions, where we are helping customers in some ways either do more of their own service or -- because we really don't help them all that much by just sending an engineer and having them pay for our engineer plus something, right? That's not a great deal for either of us. So some of these data tools that help our engineer -- our customers maybe troubleshoot their tools more effectively themselves or perform some of the routine maintenance or maybe even have a third-party who can do it much cheaper than us perform it. And what we end up selling is the know-how, maybe the -- we have one product, which is a tablet-based system that helps our customers perform routine maintenance with very high precision and repeatability without having to pay Lam for that service itself. They just pay for the tablet and subscription to that capability. And so we're continuously looking for win-win ways to help them lower the running cost of the fab and kind of continue to create this ongoing revenue stream for us. In terms of lifetime value, we do think about that. And in fact, the only data point we've given so far is that there are -- definitely some of our tools in our installed base and in our product portfolio, they have as much revenue potential over the life in the aftermarket as they did at new tool sales. So done correctly, and if we satisfy the customers, we have a lot of potential to make revenue from those tools for 20 to 30 years.

Got it. Got it. I want to go to a few of the audience questions. We're in the last like kind of eight, 8 or 10 minutes of the session. Do you believe there's a new dynamic emerging with TSMC taking leadership where U.S., Korea, Taiwan, China, were all spending more aggressively didn't pass the bill. If I was going to ask this question myself, I'd also maybe fold in like some of the new moves with TSMC building factories here in the U.S. But what do you think about that dynamic? Maybe that folds into just the rest of the geopolitical situation as well? But how do you think that sort of like plays out as TSMC is now becoming the premier semiconductor manufacturer versus maybe some of the U.S. names that might have held that title in the past? What does that mean?

Yes, I guess, until -- maybe I'd honestly say that until we really start discussing geopolitical issues. I mean, we haven't distinguished all that much. I mean, we're a global company with resources and employees all over the world. And as I just mentioned, that's something where TSMC's has been a partner of ours. I mean, and every semiconductor equipment maker for many, many years. And so I don't know that it changes the dynamic all that much. We focus -- obviously, we focus on our customers, but we really focus on the problems that foundries are facing and logic companies are facing and memory companies are facing. And we try to deliver solutions to those problems. And done correctly, actually, a lot of customers will implement those solutions, right? It's like we build an etcher that's capable of doing high aspect ratio etch with the hopes that every customer will actually adopt that. And so I don't know that we -- that our business fundamentally changes depending on any moment in time like one customer or another is doing something.

Okay. Okay. A question asking about atomic layer etch.

Yes.

We talked about atomic layer deposition. Can you talk about atomic layer etch?

Also an equally important application that we've been working on, and we talked about a couple of years ago. We do -- we -- that's actually a little further along, it's often integrated into perhaps some other process. But we've talked about it over the years. I mean, it will find applications just as atomic layer deposition did. It's just, say, ALD is -- has some more visible disruptive properties right now given the markets it's attacking. But ALE, we've looked at it for things like line edge roughness smoothing with EUV and several other applications. But I think there's a couple of things that I think you can take for certain going forward. In the future, there will be stronger and stronger demand for atomic level manipulation, whether it be etching or deposition of some materials. And there will be more importance of selective etching and selective deposition. And so really, you're getting down to -- our ultimate goal is to be able to manipulate any particular atom on any particular surface and add it or remove it. And we're not quite there yet, but that's what's coming. And so ALE is part of our portfolio and one that we're investing in.

Got it. I did a PhD thesis on etch and line edge roughness, so that...

Okay. Well, there you go. So I knew we're -- we're getting closer to solving that problem.

We didn't quite solve it. Anyway, question on just memory supply and demand, what do you think about the current supply demand situation in memory around your customers and when do you think it might get back to normal?

Well, I mean, I guess we said we exited 2019, we felt quite undersupplied. We've seen investments in memory through the first half of this year. And if I know exactly what's going to happen with current events, I'd be able to tell you, but I believe that it will continue to normalize. And I don't think -- I think you're hearing, like, I think maybe even at your conference here, there was some commentary about memory that sounded positive. So therefore, I would say that my comment, memory plays such a critical role in the data economy that I actually feel like you can't remain undersupplied for very long. And so I think we expect memory investment to continue through this year.

I guess along those lines, how serious should we take China's memory aspirations? Like are the Chinese players actually competitive? Like how do you see them evolving? I'll put the geopolitical concerns at the moment. I'm not [indiscernible]

Yes. I think it's -- well, it's dangerous for me because it's -- there really are only -- there's only really one customer reach. And so I don't want to comment on specific customer capability. But more broadly, I mean, I think that if you're committed to investing and you're committed to the length of time it takes to learn and mature in this business, obviously -- so maybe the key word there is the commitment feels like it's there and therefore, it just takes time. As I mentioned, with our tools, you get a lot better if you've run millions and millions of wafers. And so I believe there is a learning curve in many of these that people have to come up. But that's -- over a period of time, it can happen. So we have to see how it plays out. But I wouldn't say there's a fundamental reason why people can't succeed in [indiscernible]

Got it. I want to ask you a question about ESG investing. So this is a topic that my clients are obviously becoming more and more interested in every day. Is this something that you're spending a considerable amount of your time on thinking about it? Like what is Lam doing around ESG? And I guess more broadly, what you think the semi-cap industry could be doing to improve the most in terms of ESG -- in ESG investing.

Well, I guess -- I mean, obviously, we define ESG in our sense as -- we issue a corporate social responsibility report. And a lot of that, I mean, when you say -- am I spending a lot of my time on that, there's a certain element that I'm spending a tremendous amount of time on. And that is our employee engagement. I mean, we -- our business is a knowledge business. I mean, it's our people who are making these products and creating the innovation. So I spend a lot of time trying to figure out how we keep our employees highly energized and motivated to do a great job every day they come into work, how they can do a better job every day to come in to work through how they feel about the culture of the company. And so I definitely -- we've ramped that activity up significantly in the company. And I think with good effect through -- that we're seeing through our engagement surveys with employees. And I think that we're getting -- it's interesting. We did an employee engagement survey right in the middle of the COVID crisis, to work from home. And we saw an improvement in the scoring across every single category in our employee engagement survey. And so I think it's important. I mean, obviously, I'm a believer that CSR is important for a lot of those reasons. It drives business results because your people feel better about the company and they work harder towards your objectives. So spending a lot of time on that. Product-wise, ether, the drivers of this product, right? We talked about the fact that it saves a lot of chemical waste. And so we do have customers continually asking us for power reduction, water reduction, chemical waste reduction, and every time we get a chance to build a new tool, we factor those things in. But it's -- it takes some time because it's hard to make some of those changes dramatically on existing installed base systems. But it's a focus.

Got it. So we're down to our last 30 seconds. I'm going to give your soapbox.

My soapbox?

Why should investors buy Lam stocks?

Boy! I thought that was the last 49 minutes of this. But no. No, I think very simply summed up, I mean, I think I said, I think we're in a great industry. It's at the center of -- technology is at the center of just about every way in which our life is going to evolve. So the industry is great. I think our markets, etch and deposition, like I said, they're among the fastest growing, and we see lots of legs for that. And then I think that we've built a track record of new product introduction as well as operating model resiliency through our installed base business that I think just gives us confidence to invest kind of not independent of current events, but you can see these investments through, and that gives us high confidence. I mean, our model we introduced -- we'll just come back to that. We would take earnings per share from $14.51 in 2019 to $31 in 2023, 2024. And over that time, we also said we would return 75% to 100% of our free cash flow. And we said we would increase our dividend annually. So it's -- I think those are -- at least, our management team's statements of our confidence in the build -- business and our story and probably the reason why people should at least consider our stock.

Got it. That's wonderful. Thank you. I think we're going to close it here. For everybody on the line, I wanted to remind you, we're doing live polling with our partner Procensus. So if you could please click on the Procensus link on the left of your screen right now, you should see a new window pop up with a short poll that will take 30 -- 60 seconds and you'll benefit from real-time tracking of investor sentiment on Lam. With that, we'll close it. Tim, thank you so much for joining us today. Really appreciate the time.

Great. Thanks for your time, Stacy. Thank you.

Take care. Bye-bye.