Lam Research Corporation (LRCX) Earnings Call Transcript & Summary

Why don't we go ahead and get started. I'd like to welcome everyone this morning to the fireside chat with the management team of Lam Research Corporation. To my immediate left is Tim Archer, President and CEO. To his left is Doug Bettinger, Executive Vice President and CFO. Also in attendance is Ram Ganesh and Tina Correia from Investor Relations. I appreciate everyone coming to the session today. We've got a pretty straightforward fireside chat. I'll start things out. There is a mic in the middle of the room. It's not being passed around for COVID protocols. But if you do have a question, please stand up and just walk over to the mic, and we'll try to get that to you. First and foremost, Tim, Doug, I really appreciate you guys coming out. It's great to finally do this face-to-face back [ again ]. I think the last face-to-face big investor event was your Analyst Day in early 2020. So it's a nice little bookend there.

Tim, I always try to make my first question a little bit open ended to give you the opportunity to talk a little bit about the value proposition you think Lam represents to the investment community. You're coming up on your third anniversary as CEO, but your contributions to the company goes back, I think, 40 years with Novellus -- at least 30 years with Novellus. So you've been around for a long time. During your tenure, stock's are up 5.5x, massively outperforming the S&P. I guess my simple question to you is what's the second act look like for the people in the room?

Hopefully better than the first act, but -- now first of all, let's set the record straight. I have not been here 40 years. But it is 27 years in the equipment industry. So close to 30 years. But yes, thanks for having us here. And you're right, it is very nice to be back in person. And I think that -- what was interesting is when I think about this, I spend a lot of time right now thinking about the sustainability of our success. I mean I appreciate your comments about nice performance the last few years. I mean, really, we're quite proud of that. But we really want to make sure that this track record of success continues. And so we're thinking a lot more about how we deliver value, not just to the investment community, which is incredibly important, but to all stakeholders because this business requires us to think much more broadly. And so when I think about that from a stakeholder perspective, we've been spending a lot of time thinking about Lam's culture and how it is that in all of our decision-making, we prioritize success of not only our customers, but also company and investors, but also our employees because there's a tremendous war for talent out there. Our -- everything we deliver fundamentally is off the intellectual property of our people. And so we spend a lot of time thinking about that and I'd say our culture is in a great place from that perspective. Value proposition for customers. I think we've got a fantastic technology portfolio that really addresses where we think technology is going. It's -- In some parts of the market, we have incredibly strong and sustainable and defendable positions like our position in memory. But what's really exciting for us is that going forward, we have an opportunity in foundry logic that's a little bit unencumbered from the past. And that means that we can position a lot of new products without the fear that they're going to disrupt big segments of our own business. And so I'm sure we'll talk about some of those, but it's when you -- we're here we're thinking about things like ALD and selective etch and dry EUV resist. And so that's an exciting second act for us is how to grow that part of our business very aggressively. And it's -- and then maybe the other thing for investment community, but also for Lam is just this point of, we exist in a market right now where things are getting more complex. And so Lam's value proposition in that case is we're a company that has always taken on the challenge of the most critical applications. We -- that's where we really thrive, is solving technical challenges and differentiating ourselves for our customers in that way. And so the fact that the business has become -- processes have become more complex, dye are getting larger. There's no wafer size transition on the horizon. So customers are looking to companies like Lam to deliver technology and productivity. And I mean, ever since the days of Lam and Novellus, both that's really -- it's a great marriage of technology and productivity. And so I think we kind of think about some value proposition for kind of every one of those stakeholders.

No, that's a great answer and it makes a lot of sense. I have a lot of Lam-specific questions, but I want to kind of get the bigger sort of industry WFE question sort of out of the way. We were discussing this a little bit last night and this morning. If I go back 5 years ago, I think WFE was sitting at about $40 billion, $42 billion. And most of Wall Street was terrified that that was going to be a peak and things were going to fall down. Over the last 5 years, we've seen WFE go up 155%, pretty consistently. And despite that, we find ourselves in the worst supply shortage situation that the industry has ever seen. So what's happened? And I guess, more importantly, to your sustainability, how do you think about kind of capital intensity and WFE growth from here?

Yes. It's -- we were -- you mentioned we were last together in kind of one of these events in March of 2020. And at our Investor Day, we put out 2 models. One was $60 billion WFE in 2023, 2024. In the high end that we got a lot of criticism for $70 billion. So yes, it's -- I don't think any of us really quite saw this WFE future. But the reality is it's a buildup of things that, as you look at them, are not that surprising. I just mentioned a couple. I mean, you've got incredible increases in process complexity as technology moves forward. You've got a broadening of applications that really are drivers of WFE, we never talked about before. Again, a few years ago, you wouldn't have thought of automotive, and the increasing semiconductor content in automotive is an actual driver of new WFE spend and yet that's got a lot of attention, and that's a tremendously strong part of the business, those trailing edge technologies. I mentioned larger dye size, no wafer transition. These are things that are again just -- each of them makes sense. Yes, those will be drivers of WFE. And I think when you put those all together and you combine them with, now you're starting to see this -- the effects of government intervention and to talk about adding additional WFE into the marketplace. I think it's just a driver of a sustainably growing WFE environment over many years to come.

So Tim, Doug, to put numbers behind that. I mean you've heard me talk about 2000 to 2015, the WFE and if you really didn't grow all that much during that 15-year period. It's our view that semiconductor revenue growth rates are probably accelerating along secular lines. Should we start to think about semi cap equipment as at least semi-type rev growth rates? Or is it semi plus? Or how would you guys think about that?

Actually, John, when I think about it, I think it's semi plus, frankly. And you have to think about the secular demand for semiconductors is driving what's needed for all of us, right? I mean just look around, everybody's got a digital device. Rising capital intensity layers on top of that when you think through and look at what is happening with the growth in wafer fab equipment spending. So when you put the 2 together, yes. I mean things are good. And when you layer on top of that the space in WFE that we sit, that's grown even faster than WFE. So it's a multiplier when you look across the totality of our business and why we've done so well.

No, that makes a lot of sense. One last near-term question before getting into Lam specific. You had one of your larger peers report a few weeks back and they clearly talked about supply constraints. And so one of the questions that I keep getting asked by investors is how do you think about the current supply environment? It's sort of an odd dynamic that the guy that makes the equipment to make chips, can't get the chips to make the equipment. How are you guys sort of characterizing the current supply and especially with your footprint in Malaysia, which is unfortunately had more COVID flare-ups than perhaps other goes across the world?

What I'd say is maybe throughout the COVID period, we've been dealing with different constraints. First really was about labor and getting enough people in under the COVID constraints to do the work. And second was about expanding our own physical capacity. And with the opening of Malaysia earlier this year, a second factory we've just announced in Oregon and practically doubling of our capacity in Korea. We feel really good about our own physical capacity. Now we move on to the next thing. And now we're on supply constraints getting enough parts in the door. We've got a fantastic supply chain team. And it's not to -- I would note their job is incredibly difficult. But I think we have a great track record right now of sort of addressing the issues that are coming up, seeing the next issues were they might be. We're partnered very closely with our key suppliers and I think it's going to be with us for a while. But I'd say so far, it appears as though we've been able to work through the issues as they've emerged.

And Doug, is there a way to quantify sort of the cost headwinds it's represented into the P&L?

Yes. I mean we've been talking about headwinds in gross margin, I don't know, I think since COVID began. I haven't quantified it, but I've been pointing a couple of things out to people. First, I'd go through it. As we ramp Malaysia, it's a bit of a headwind that will become a tailwind as we ramp volume. So I think that's intuitive. Freight logistics is one of the bigger things that sticks out because we have inbound materials coming from Asia to our factories in the U.S. and then outbound going back to the customer base in Asia. That's the biggest headwind. And so while we haven't quantified it, I've been pointing something out to people to help you think about it. If you go back to that financial model we put out in March of last year, the implied gross margin in there was 47.5% to 48% and we just guided you to 46%. And at these revenue levels, I would have expected us to be closer to the model. And so that's a way to think about it. It's inherent in that delta right now.

No, that makes a lot of sense. So I want to be very careful how I ask this next question because I think it's a little bit silly to make you apologize for your success in memory. But clearly, I think one of the concerns that the investment community has is that you're more memory levered play and less a logic foundry levered play. And the memory part of WFE tends to be a little bit more volatile. So a couple of questions sort of around that. In sort of your opening comments, you talked about unencumbered opportunities in logic and foundry. I'm wondering if you can flush that out a little bit more. And I know you hate to talk about specific customers, but I'd like to think that you came to Arizona just for me. It's been much easier to get CEOs of equipment companies to come to Arizona given all the building activities going on. So maybe you can talk about the opportunities with the 2 guys here in Arizona that are about to embark on a pretty big expansion plan?

Sure. One, I'm not going to apologize for our strength in memory. I mean it's -- to the Lam-specific story, it's a great foundation on which to build. I mean you see rising capital intensity. Our position in the memory space, really creates this foundation for reinvestment in the business from which we can then either attack additional opportunities within memory, prepare ourselves for coming inflections within memory like eventual move to vertical DRAM or invest in foundry/logic opportunities and create those new products that are going to drive growth there. And so the memory base that we've worked so hard to build, I mean, that's just a tremendous asset of the company, the one that we think just keeps getting stronger. We update this number occasionally on the 3 most critical applications in NAND. We now are up to a 45 million wafer gap to the next nearest supplier through those applications. And so again, these moats are incredibly valuable for years and years to come. Now what do we want to do in foundry/logic and kind of what are those opportunities? You're right. I mean there are -- there's a competitiveness. I mean just like in every segment, but our customers are competing to deliver better and better technology at the leading edge and the customer that can get there with the best solution and the most productive solution is likely to be the winner. And so they -- I would say we're seeing an engagement with the supply chain, companies like Lam to work on those solutions. When I say unencumbered, it means we don't have to enter in sort of like sell an extension of a tool that's been around for a while. We can come in with a free mind and really collaborate with those customers. And I would say we're doing it with every customer, regardless of whether they're going to be in Arizona or in other parts of the world. The challenges are similar. I mean there are needs for new applications, new tool types. We talked about one, selective etch. As you move towards nanosheet or gate-all-around type structures, you need to evolve to ultra high selectivity, so you can remove and recess certain materials without touching even an angstrom or more of the active silicon part of that device. Those are technologies. Actually, we've always had the surface integrity group, our clean group. The new stuff to do the same thing on the active junction area, and that was always one of our strengths since we've been able to adapt those types of technologies. And I think we're going to have a great selective etch process. We're engaged deeply with all of those leading customers on those types of technologies. Dry EUV resist, disrupting the -- if we're looking to disrupt sort of the track in resist and develop market, again, with a better solution for making EUV cost-effective and more extendable. That's a technology we're working on. ALD metals. The -- as you make the transistor better, the interconnect has to get better as well or else you end up with RC challenges that don't allow you to extract all the device performance. And so we see ALD metals being an inflection that may inflect some of the current PVD or CVD-type processes that are being used there. So Lam can enter and engage with customers with that kind of open slate. And a lot of the ideas come from the customers. I mean [indiscernible] [ best ] guide when you only have a few customers to let them tell you what it is that they need. And then our team of engineers go off and create that great solution.

And I know this came up at dinner last night, but to the extent that we love numbers on Wall Street, is there an ability to predict what the mix of your business might look like between memory, logic/foundry over time? Or is it just too difficult of a target to try to hit?

I think it's difficult in any period of time. And what we know is we would expect our business in both memory and foundry/logic to grow. I mean that's obviously the mix. It could be an element of time, but the reality is, as we grow our foundry/logic business, our memory business will also continue to grow because the intensity of etch and deposition as you double the number of layers in NAND, for instance, rises quite fast. Not only for the existing pieces of equipment used to build the stack but also for new steps. We've talked about one of those, which is the -- as you build more layers on to the front side of the device, you need to deal with the stress of the wafer. And so we've recently introduced and had great success with a tool that balances that stress by doing something to the backside. And so those kinds of new opportunities that come up. Since Lam is so deeply embedded in that development process with the customer, we get first access to those opportunities as long as we execute, then we end up adding to our portfolio. So I think you're going to see improvement in foundry/logic, but you're going to see memory continue to grow in intensity as well.

Well, I'm curious because if you look amongst your peers, I think you guys were a little bit less vocal about giving detail for WFE next year. And I think that the way that some interpret it on Wall Street is that, "hey, their memory exposure is making you guys a little bit less sanguine than maybe your peers on next year. Can you dispel that a little bit or give us a little bit more color on how you view WFE for calendar year '22?

I'll start and then I'll let Doug. He usually tells me what we can and can't say about the next year. I'd tell you the whole story. But it's -- no, I think that in general, there's -- we're sticking to what we have to focus on. I need to make sure we have enough capacity that we have enough resources that we're ready to meet customer demand. A precise number for WFE for us it's something that -- it's a little too early for us to feel that we have to do that. What we said is it's going to be a strong year. We feel good about all of the markets fundamentally in terms of the trends I just talked about, greater complexity, greater intensity, Lam's opportunity. And so I don't think there was any -- there was certainly nothing we're trying to mention -- message. It's just we've kind of stuck to that we give that kind of guidance in the January call. And to start to get out of that cycle, source say maybe we're actually trying to message something differently. But we feel quite good about 2022.

Yes. That's the only thing I'd add. There's a cadence to what we say, when we say it in the calendar year and I fact, I think at earnings we said a little bit more about the next year than we normally would have because we see a very strong year. And we'll give you all the normal color that we do when we announce earnings next quarter, John.

And then, Doug, I wanted to give you an opportunity because I know you've thought about this a lot, just the volatility of memory CapEx diminishing over time. I think one of the interesting things about the memory cycle last time around. I think it was the first time ever we saw CapEx get cut before gross margins actually peaked. And typically, these guys spend at 1,000 miles falling off the cliff and they just didn't do it. Is there an argument in your mind that the industry is just healthier looking at CapEx in a different way? I know you look at not only CapEx to rev but you look at CapEx to sort of profitability.

Yes, that's the most important metric in my mind is -- everybody likes looking at capital intensity and CapEx [ versus ] revenue, something we've been trained over, I don't know, decades to look at. Frankly, it's a wrong matric in my opinion. CapEx as a percent of operating profit is the right metric or EBITDA, however you want to measure it. Because we buy CapEx in this industry out of profits, not out of revenue. And so keep that in mind. Then when you look at that metric, actually, it's not at a high watermark. It's actually had a relatively low watermark because the industry is extremely profitable right now. And I think that's sustainably where we are. This industry has matured. Everybody proactively manages their business, the cadence at which investments are made are measured and thought about. And I just view what's going on today as a prudent way of running companies and this industry has matured.

Tim, you've looked at your value proposition to customers along at least 2 lines. One is, can I give them the technology they need and the other has been around productivity. I think one of the things that has surprised people this time around is how much WFE is actually getting spent on less bleeding edge, more mission-critical devices. Can you talk a little bit about your leverage to that part of the market and kind of how you feel about the growth prospects overall for that market and then for Lam specifically?

Yes. So the -- a lot of -- the leading-edge technologies get a lot of the attention and we spend a lot of our time talking about where we're going to win. But our business in the trailing edge, one, we're well exposed. It's a market that we see is growing. We've said kind of 2 to 3x faster than overall WFE growth primarily driven by the fact that it's this broadening of applications in that space. People every day are finding new use cases, which is part of what's creating this supply shortage, too many good ideas for how to use semiconductors. And then you're hearing a lot of talk about semiconductor content, increasing content, really adding value to those end products, whether it's consumer products or it's automotive. And that space, just we look at that as a grower, much like our installed base, where it's a little less subjected to the ebbs and flows. It's more about just the permeation of semiconductors into all kinds of products that didn't use to have those, all kinds of industrial products, consumer products. Things like image sensors exist within that space, for instance, Everywhere you look, image sensors are increasing in quantity as well and complexity. And so I think our business there it will be -- it's one of those foundational elements of every year we just anticipate that business kind of continuing to expand. It's a little harder to go back and introduce new products into that space. For the most part, many of those processes are well established. But we are actually reengineering some products to meet the needs because we used to service that market with refurbished equipment. And today, the vast majority of that needs to be sold as new equipment, and I think we'll just continue to see growth in that space for many years to come.

And Doug, the margin implication as that business continues to be healthy?

Yes. Listen, CSBG is a broader part of the business to think about, which is refurbished equipment, spare parts, service and upgrades, right? When you put all that together, we began disclosing that, I guess, a little over 1.5 years ago, maybe a little bit dilutive to gross margin, but nicely accretive to operating income because there's not a huge R&D investment required in this segment of the business. And the great part about this part of the business is, it's as close to annuity as you get in this industry, frankly. And it grows every year because chambers grow every year. I mean it's got a very long tail to the revenue generation because our chambers really never go away.

And Tim, this came up at dinner last night, but part of the model, financial model you gave out in March of 2020 had included kind of this idea of more dollars per chamber over time. And you actually said that relative to where you thought you were going to be then, you're a little bit further ahead. What are the dynamics that are driving that?

I think that the -- so it was the revenue capture per chamber. We said as that install base grows, we're also going to introduce new products and services that help the customer extract more value from the installed base. People love to get more out of what they've already purchased from you. And so we've seen that. We've introduced more services that are focused around data and equipment intelligence. That's -- we can capture some value for that and revenue from existing chambers. The other thing is productivity upgrades, especially as we look at those trailing edge applications. I mean, today, the customer can implement a productivity upgrade that helps them get more output out of the existing installed base, it's highly valuable for them. They can sell everything that they can make at this point. And so we can -- we've seen business in the productivity upgrade section and then the same thing for technology upgrades, which is you've already. The fastest way to bring on that next technology node quite often is, rather than putting in greenfield capacity, it's to go in and upgrade what you have. And so we're just seeing strength across all segments. Spare parts, which is another part of that segment as well, revenue capture. As the processes become more critical, spare parts consumption actually increases for some of those applications because the chamber conditions must be kept more pristine to deliver this angstrom level uniformity you may need from the center to the edge of the wafer to secure yield or the defectivity you may need to secure foundry/logic yield. And so the spare parts business, again, perhaps on a revenue capture per chamber is, I'd say, tracking well to our growth model.

Any questions in the audience? I want to make sure I don't monopolize the conversation. If you want to step up to the mic, please.

I mean maybe you can just talk quickly about all those architectural changes happening in the semi manufacturing space, so basically gate-all-around, replacement gating, more packaging technologies and so on. How does this impacting you as Moore's Law is slowing down and you have this kind of vertical efficiencies? Because I heard some arguments that you are less exposed than some of your competitors as we go more vertically to [ transact ], for instance, gate-all-around. If you can just talk about it quickly? How do you benefit from those architectural changes?

Well, I would say that I think that -- I would say that, that actually -- we would think of ourselves as the most exposed to the transition into 3D and vertical devices. So I actually -- those are transitions that we see as really positive for Lam. The first, of course, which occurred was the transition to 3D NAND, which obviously established an incredible success for us, but also established in our customers' minds the fact that Lam really is the company most associated with high aspect ratio etching, putting things together in a 3-dimensional way with metallization and other structures. So things like -- processes like advanced packaging, very positive for Lam, both from an etch and deposition perspective. Gate-all-around, I just talked about some of the new products, but essentially transitioning foundry/logic into a 3D space introduces new processes. In fact, that selective etch machine that I talked about very likely will be used both for gate-all-around, but also will probably play a critical role in vertical DRAM or 3D DRAM because again, you'll have similar requirements to do this kind of highly selective recess process in an etch tool. So I think that as the industry transitions more into the vertical space, which you see as you see intensity shifting away from certain steps, which are used to make very small devices, into steps like etch and deposition, which are used to make tall high aspect ratio devices. And so I think those trends bode very well for our product portfolio.

Tim, another really important technology trend that you guys, I think, are better levered to than people that don't appreciate is sort of the chiplet tile and the need for advanced packaging. Can you spend a few minutes just kind of discussing the products you bring to market there and kind of the leverage you have?

Yes. I think that, again, back to the construction of 3D structures, I mean, generally, it's -- it requires 2 key elements. One is the ability to etch those through contacts and also then to fill them with some sort of metallization. We have an incredible position that we've built over the last 20-plus years in copper electroplating and plating of other materials. And those play an important role in the chiplet heterogeneous integration structures. And of course, etching plays an important role there as well. And so we bring those to that, I'd say our engagement today is very strong. It's still a market that's in the early stages of growth. I mean, you hear a lot about it, but I think more designs have to migrate to the chiplet or heterogenous integration approach. But I think that's coming as customers look for ways to bring productivity. That's a means of bringing productivity and dealing with their -- some of their cost scaling challenges.

Doug, you brought up the financial model, a couple of questions ago, and you talked about some of the headwinds to that model. I know there was a lot of thought and a lot of sort of assumption that went into that model. And we on Wall Street try to oversimplify everything. And so we look at the headline and go, well, wait a second, WFE is a lot stronger than you thought and the earnings power is not what you predicted. And I guess my question to you is there was a key element around time in that model for market share in SAM expansion. And so if you had to adjust the model for time, how do we think about how you're doing today versus what you would have expected the model to yield?

Yes. I think, John, you're right. There's a key time component to it. It's SAM expanding as these 3D device architectures become more prominent and the stack gets bigger and so forth as gate-all-around comes in and our SAM is bigger in foundry and logic, the advanced packaging is a piece of it. So there's a clear time component to the addressable market. There's also a clear time component to our market share objectives as we bring new products to bear like Sense.i and dry resist, actually dry resist is a SAM expander in addition to share. So that -- there's a time component to that. In addition to, from a profitability standpoint, as we ramp the Malaysia factory and save dollars in a lower-cost geography and have less freight and logistics because we're closer to the customer base, those time components are still happening and are spot on where I would have expected them to be as we sit here today. It's just WFE is much stronger right now. So that's still the right way to think about it, the financial model.

And then a softball for my last question, but just use of cash and capital allocation. Is large M&A in the space largely behind you? And if it is, at some point, you're going to be buying back so much stock that I'm not going to be able to invite you back on stage because you will no longer be a public company.

I'd be while before that, but I'll let Doug add as well. But yes, obviously, our first priority is we look for opportunities to reinvest in the business. I mean I feel really good about the strategic investments we made. And what's very exciting, we're producing what we think are our outstanding results today. And yet, during COVID, I mean, as challenging it has been, we've built a huge facility in Malaysia that's going to pay dividends for us for many years. So we invested in the business there. Next spring, we're opening our new Korea Technology Center, that was a strategic investment in a new facility located very close to a couple of our largest customers, a very strategic investment there. As I mentioned, an expansion in Oregon factories, expansion of Ohio factories, expansion in Korea factories. So we're putting money back into the business as fast as we think we need to, and we're still left with a lot of cash. And so I think at that point, we look for small M&A technology acquisitions that can accelerate our own development cycles. But I think to your point, large M&A, we don't see it right now, at least on the horizon. We're always looking, but it seems very challenging. And so therefore, we had talked about returning 75% to 100% free cash flow to investors. We've increased our dividend. We have about $3 billion left on our authorized buyback. And I think that this game plan is working for us, and I think we're just going to stick to it for the foreseeable future.

If it's not broken, don't fix it. Perfect. With that, we've run out of time in this session, but I want to thank both Tim and Doug for joining us as well as everyone in the room, this was a great conversation. Thank you.

Great. Thank you for having us, John.

Thanks.