Lam Research Corporation (LRCX) Earnings Call Transcript & Summary

Great. Why don't we go ahead and get started? I'd like to welcome everyone to the fireside chat. It's my pleasure this morning to introduce Tim Archer, the Chief Executive Officer of Lam Research. We have about 30 minutes in this session to go through Q&A. I'll kick things off. If you have any questions from the audience, please feel free to e-mail them to me, and I'll try to incorporate them in. But with that, Tim, I wanted to first thank you for participating again at this year's conference. I wish it wasn't virtual and that we were in the desert, but such is life in 2020.

I always kind of find it useful as sort of a first question in these fireside chats to help you kind of set the tone, to give you kind of the opportunity to talk from a broader perspective. What is Lam's sort of overarching strategy? And most importantly, what's the value proposition that you provide for investors?

Great. Okay. Well, thanks for having me this morning, John. I think most of the participants are looking at the Safe Harbor here, so we will reference that. Everything I say today, obviously, covered by that safe harbor. So maybe we can take that down, and then I can get into your question. The overarching strategy for Lam is -- I guess when you're inside the company, you think it's pretty simple, but I'll quickly run through that. And we laid it out at our Investor Day earlier this year, and it's about how we grow the company across, really, 3 different angles. And the first is we really like the markets we're in. Etch and deposition have been some of the fastest-growing segments of WFE, really because of this trend in the industry to convert all device architectures, whether it's in logic, foundry, DRAM, NAND, even now we see advanced packaging to 3D. And that move is increasing the capital intensity of etch and deposition. We see our SAM growing at every device node across every one of those segments, and that gives us a nice tailwind for expanding our addressable share of WFE. So the markets are good. And we talked a lot about SAM expansion as part of our growth story. We've demonstrated that over the last 7 or 8 years. We also talk about share gains. And in fact, we do that. We continue to innovate on platforms, and we talked -- we have an objective to grow another 4 to 8 points of market share through innovative new platforms. And you've seen one of those this year in this Sense.i etch tool that we've launched, the first ground-up etch system from Lam in 20 years, and we think it brings in a lot of the new advanced capabilities that are offered by big data and machine learning. We call it Lam's Equipment Intelligence solutions, and it really goes to the heart of what our customers need at this point, which is productivity delivered for those very advanced complex manufacturing nodes. And so we have an angle of growth around share gains. And then the one piece we've started talking a lot more about and we had some additional disclosures this year is the power of our install base to really serve as a platform for us to build an entirely, I wouldn't say separate, but kind of somewhat separated portion of our business, which delivers services and spares and upgrades and really value to our customers on the 60,000-plus chambers that Lam has sold in the past, and our customers still want to extract more and more value from as time goes by. And just in the past quarter, we had our first $1 billion revenue quarter from that install base business. And so when we look and we combine an increasing served market in etch and deposition, share gains from new products and then this powerful platform for building an installed base recurring revenue stream. We just think it's a business that one has been performing well and has a great future in front of it.

That's a great intro. I wanted to try to organize my questions today, Tim, into 2 main buckets. The first is kind of some of the top-down drivers for WFE and the ecosystem. The second half, I really wanted to talk about some of the bottoms-up drivers for Lam. But starting with sort of the tops down. You and I have had discussions about this just about every quarter. We've seen it in the public stock market, massive multiple expansion of SACs. I would argue that you and your semi-cap equipment brethren had benefited to less from that multiple expansion. There's been a lot of skepticism really about whether or not we should think about this industry as cyclical or actually as growth. And you had a great chart at your Analyst Day earlier this year that looked at WFE over a very long period of time from 1990 through I think about 2020, and that chart is either a thing of beauty or it's terrifying depending upon whether or not you're looking at it through a purely cyclical lens or one that's a little bit more structural. To the investors out there that are worried that at current WFE, things feel frothy, why shouldn't they be worried in your opinion?

Well, John, I think that chart really is great for somebody who wants to look at kind of the power of this -- the semiconductor equipment industry over a long period of time. And there's no doubt from that chart. Those who, unfortunately -- I wish we could show the chart here, but it shows tremendous growth. I think it was almost like a 10x growth in the WFE spent over that period, which means growth for the equipment business. And inside of that were a whole lot of small cycles that you saw occurring, especially kind of in the late '90s and mid-2000s. But stepping back, you see this is a growth industry, without a doubt. And all of our companies were much bigger now. I've been in this industry for nearly that entire chart. I joined in 1994 to the equipment industry. And so when you're in it, you feel those cycles. But when you step back, you realize the companies are getting -- the opportunity just continues to get larger. Now I can't really speak to the multiple expansion and when that will be recognized, but you know, I come back to the quality of the business because if you look at how companies like Lam have performed over that period, again, you don't find many 40-year-old technology companies that are actually today performing at their absolute peak. I mean, it's -- and that's the power of the equipment business, is as the electronics industry and semiconductors expands, that business really aggregates into the demand for semiconductor equipment and demand for us to continue to innovate. And I think that's why, on that 30-year chart, you just saw the equipment business being a steady -- I would say steady grower if you looked past those cycles. But I think a couple of things have happened. And maybe just within that chart, that's a little bit hidden. First, in 1990, when you go back to the beginning, I looked it up because the penetration of PCs into the U.S. market was only about 15% at the start of that chart. We kind of know now that's like 90%, maybe 90% plus. 1990, I mean nobody knew about smartphones. Nobody knew really about the cloud. Nobody knows any of these things, artificial intelligence, machine learning, big data. And so you've seen an explosion of applications over that period. And I think today, we see that just accelerating into the future. And so I'm not saying there won't be some element of cycles. In fact, 2019 was a bit of a downturn. But again, the quality of our business and the dependence on semiconductor equipment meant even though WFE was down about 10%, Lam delivered its second-best earnings per share in its company's history in a downturn. So I would say that the cycles, while they may still exist in the short term, the long-term picture is extremely bright for growth.

Well, Tim, I'd like to try to drill down specifically to some of the investors' concerns at the industry level. And I think one of the big ones is from that period between 2000 and 2015, as we were kind of transforming from 200-millimeter wafers to 300-millimeter wafers, there just wasn't a lot of WFE growth out there. I'm kind of curious as to whether or not you believe that given that 450 is not going to happen, if you look at long-term cost of capacity or capital intensity curves, they tend to reset every 10 years with that wafer start, with that wafer size transition. The fact that we don't have one today, what does that mean for cost of capacity and capital intensity in your mind?

Yes. Well, first, let me take the first part of that, which is during that period of kind of what you called no WFE growth, there was a lot going on. One was the 300-millimeter wafer transition you talked about. That delivered tremendous productivity simply because of the additional efficiencies you gain from the larger wafer sizes. As you said, right now, 450 doesn't appear to be happening. And so that productivity boost that we saw in the 2000s, that really meant that the equipment companies were sort of running in place, introducing new products, but not really seeing a lot of growth in end spending from our customers. That doesn't -- that's kind of played its course now. There's not a wafer transition on the horizon. The second thing that happened during that period was there was massive consolidation in the industry. If you look at the memory market, significant number of players exited the market, and that demand consolidated into much more efficient, larger-scale players. And so there was a productivity boost from that aspect. Same thing on the IDM side and the foundry outsourcing. And so there are really 2 big productivity drivers. It was the consolidation in the 300-millimeter wafer transition, in my opinion, that in the 2000s drove this lack of WFE growth. Those are behind us now. And what you're seeing is now an acceleration of complexity of manufacturing. You don't have those productivity tailwinds, and you've seen now WFE really start to rise and capital intensity rise as a result. But make no mistake. Productivity is extremely important. It's the #1 conversation I have with customers. And so that's why I mentioned Sense.i. It makes great use of the customer's fab footprint. It's a -- we think the most productive etch tool that's going to be in the marketplace, uses Equipment Intelligence to lower maintenance and service costs, faster start-ups, faster ramps to yield using big data and machine learning algorithms. Those are the things -- that's how you're going to get productivity in this new era when you don't have wafer transitions and industry consolidation helping you.

Well, I would argue without that wafer size transition, tool productivity becomes even that much more important to your customers.

Correct. And that's why if you look at much of our market share gain objectives where another thing investors often worry about is when you have some of your elements, you're growth-focused on market share gains, just say, how do you do that? We -- our strategy is one of technology combined with disruptive productivity, and the productivity really comes from this innovation around platform design and how you make the equipment innately more productive through the design of the equipment. And that's something that I think Lam has had a long track record on both the deposition and the etch side of delivering productivity for our customers.

Tim, the other concern that I'm sure is not new to you is just the emergence of China as a major player in WFE. And to be clear, new sovereign entrants has never been -- it isn't something that's new in the industry. You had Japan in the '80s, Korea in the late 80s, early '90s, Taiwan in the '90s. But I think one of the things that investors are worried about is a new sovereign entrant tends to have different strategic objectives than just financials, and they tend to go through periods of overbuilding. Can you talk a little bit, just given the strength we've seen in China at the industry level and specifically for you, how rational do you think this spend is? And to the extent that investors are concerned that U.S.-China trade relations might be causing your Chinese customers to buy ahead, how big of a concern is that in your mind?

Yes. I mean it's not a large concern in my mind. We've said that from the standpoint of spending today, we've generally said that we don't believe that the pull-ins is a big part of this, meaning we spend ahead. But to your point, I mean, we would acknowledge -- it's a little bit what I just talked about in 2000s. As any new entrant comes into the market, and it doesn't matter where they're located, there's a learning curve. I mean there's something that -- when we look at players that have been in the market for a long time and have built up tremendous learning, there's an efficiency that is gained in that. There's efficiency also from scale. And so I mean to the extent that any new player has to, you might call it overspend, but effectively, you're spending because you're still getting your yields up to those mature levels. You're still getting the equipment to operate at maximum efficiency in a fab that might be still ramping to the scale that the larger players have in their fabs. And so you might look at that as an overspend, but I look at it as it's the necessary spend for where our customer is at the point in their cycle. And we've seen that just play out over and over in the decades as new players come into the market. So I guess the reason we're not worried about it is we think we've seen this. We've seen this story, and we know how it plays out. And therefore, our views on -- and the model we've built for 2023, 2024, anticipate new players becoming more efficient and therefore, spend kind of, I wouldn't necessarily say moderating, but effectively modifying according to their progress.

You can't talk about China without talking about regulatory concern. And there's a lot of different aspects to this. I mean specifically on the near term, any more comments you can make about some of the actions the U.S. government has taken towards SMIC and the way it might impact your business? But I guess the broader question is, how do we think about the regulatory risk into China? If geopolitics comes into play and China becomes a more difficult geo to do business, is that a zero-sum game? Or do you think a lot of that capacity will need to be reconstituted in other geos?

Yes. I think that -- I mean it's -- there's not much more we can say from what we just said on our earnings call about 6 weeks ago. As we've said, we've applied for licenses -- for the types of equipment and parts that require licenses. We're waiting to hear the response to those applications. Not much else we can say there. I mean it's difficult to speculate on what might happen. As you say, I mean, geopolitics is not -- probably not any semiconductor equipment CEO's area of expertise, but I think in general, we look at the market, and you probably just laid it out best. We've seen over the years both demand and manufacturing kind of move globally. And whether it was Japan or it's Korea or it's Taiwan or now it's China moving into this, for the most part, they're all intending to satisfy what we look at as the global WFE global semiconductor demand. And so what our answer has been and still is, is there may be near-term disruption. I mean you can't move capacity overnight and activity overnight. But the demand we're talking about, the digitization of the global economy will be satisfied by some player, some manufacturer somewhere, and it's only a matter of how you get that demand to be manufactured. So it's -- in the long term, again, come back to that 30-year chart, come back to the 10-year chart. It's a growth industry because of the digitization of the global economy period.

Tim, the glass half-full part of this geopolitical tension concern that doesn't get a lot of talking about is this has really shown a light, I think, on the strategic importance of semiconductor production and the fact that Europe and the U.S. is fairly dependent upon Asia right now as a geo region. And you're seeing the CHIPS Act go through the U.S. Congress. I've talked to people of the EU who are contemplating similar incentives to build domestically. You've seen TSMC talk -- announce the fab in Arizona. How do you think the geopolitical tension plays out from a distribution of capacity over time?

Well, I think as you said, it certainly highlighted the importance of this industry and something that those of us in the industry have probably always believed. But the strategic nature of manufacturing, it's more than just about jobs. It's about having that very important high-technology base within your shores. And you see countries now competing and continuing to compete aggressively to bring manufacturing into their country for a whole lot of reasons. I mean, again, they're great high-tech jobs. They spawn education. They spawn all sorts of things that are good for countries' economies. And now you add on top of that the strategic importance has become so obvious lately. I think you just see incentivization. It's difficult for certain countries that haven't done it in a while. But now you see, as you just mentioned, the CHIPS Act and other things, I think you'll continue to see this move. And for us, I don't know that it changes. Again, we're a global player exposed to all geographies. And so I think we will be ready for wherever the manufacturing is occurring.

My last sort of industry question. We on Wall Street tend to focus exclusively on the bleeding edge of Moore's Law. But as you mentioned earlier in an earlier comment about the memory consolidation that happened from 2000 to 2015, one of the things that, that provided was cheap capacity for trailing-edge buyers who bought some of that equipment at pennies on the dollars. [indiscernible] not anymore. In a world where IoT is consuming more and more silicon that's not on the leading edge, how do you think about trailing-edge WFE both at 300-millimeter and, quite frankly, 200-millimeter? Are we going to go back to a world where you're actually building brand-new 200-millimeter equipment again to satisfy some of this demand?

Yes. Well, it's -- in fact, we just had an announcement just in about the past quarter or so on a new 200-millimeter tool, reporting some of our more advanced photoresist strip capability back to 200-millimeter to satisfy demand. Our Reliant business, which serves a lot of the specialty technologies, which you call the off leading-edge market, that business has had record after record after record revenues. And it's simply because, as you said, there's this explosion of applications, whether it's the wearables or it's all sorts of electronics that are now into our daily lives. And that's -- that -- we see it growing. It's expanding. There aren't a lot of tools out there to be repurposed as they were there for a long time. And so we sell both refurbished as well as new equipment into that market. We just see demand growing. And it's really, again, the growth there is interesting because it's not about necessarily big projects. It's about everything you touch is starting to contain semiconductors. It's -- and people forget that for every one advanced like smartphone application processor that gets so much press because it's manufactured with the latest foundry node, there's like 30 or 40 trailing edge -- what you call trailing-edge chips that have specialized capabilities that have to surround that in order to make that one leading-edge chip function. And so I think that's an expanding market. It's material to our business. It's a good business. It's growing. And we tend to continue to compete there with, as I said, both used and new equipment.

Wanted to transition to some of the bottoms-up drivers for Lam specifically. And at the Analyst Day earlier this year, you really hit upon 3 growth drivers. You [ wore ] them as install base share growth and SAM growth. I'm going to mix it up a little bit. I wanted to start a little bit with kind of your share growth. And if you look over the last 10 years, you've more than doubled your share of WFE from 7% 10 years ago to what's just under 15% today. But in the vein that Wall Street's always asking what have you done for me lately, how do we think about share growth from here, especially given that there's some concern that you had some tailwinds over the last decade like the move to 3D and NAND, like the delay of EUV that accelerated multi-patterning that might not be reoccurring today? What's the share growth expectation? And what drives that share growth in the years to come?

Yes. No, it's great. I mean -- so let me just take for a second, I'd like the ability to thank you for acknowledging the success we've had. It does get harder. There's no doubt. You doubled it, and I know there's big expectations out there for continued growth. But that's why, again, we're really focused on a couple of things. One is you just mentioned big tailwind, 3D NAND conversion. No doubt, Etch and deposition was the right place to be. I just mentioned everything is going 3D. We're seeing -- you look at logic foundry transistors becoming more 3D, go to gate all around, much more 3-dimensional structure. 3D DRAM, while still pretty far out there on the horizon, it's on the horizon now. You look at 3D chip packaging. All of those things are driving etch and dep capital intensity to a higher level. And so I think that you'll see continued tailwinds for etch and deposition from those aspects. At the same time, you're talking about share of WFE, which means that kind of pulls in some of the SAM expansion. And we talked at the Investor Day about a couple of things where we look at part of the parts of the SAM -- parts of the market we don't compete in today, and we think about how we disrupt those with etch and deposition solutions, places where we're really strong technologically. And we've highlighted, too, the wet resist market that we think is ripe for disruption in the EUV transition to a dry resist. And we don't do wet resist spin on equipment, but we do dry deposition equipment. And so pulling that SAM in helps us grow our share of WFE with that technology. Really happy with the launch and initial engagement of that product. And then enhanced ALD, where we don't compete in certain parts of market like furnace, but we do compete very effectively in atomic layer deposition. And so there are segments of that furnace market or the SACVD market or some of these other technologies, even the spin on dielectric market that can be pulled into our enhanced ALD SAM by -- through technology innovation. So we're just looking at many different ways, and we think that there are a lot of opportunities for us to grow our served portion of WFE and also the share of that WFE that we're capturing. And so I would say story is not done with the doubling that we've had in the last few years.

I think one of the things that was very impressive at the Analyst Day was your commentary around the dry resist market and your expectation that, that could actually be over a 5-year cumulative period kind of a $1.5 billion incremental business opportunity for you. How do we think about the ramp of dry resist for you? And how do we equate it to kind of EUV systems that are being shipped out in the field?

Yes. Well, it's -- we gave that number, $1.5 billion cumulative revenue over a 5-year period. Obviously, we have just started shipping the first of those systems into the EUV leaders. Now I think most people familiar with the foundry logic world know it's a pretty long cycle to get qualified into R&D and then go to the pilot line and finally get to HVM. And so you're really talking about revenue impact that's pretty back-end weighted on that. I mean it's the 2022 and beyond type of impact. But what it will do is it will scale. Effectively, it starts to scale with EUV layers. And we think that EUV is at the point where it is a very solid solution for a lot of the, again, cost and scalability challenges that the industry has to solve. And Lam wants to help enable that because scaling is good for us. Again, as nodes move forward, you need more Lam equipment. And so we want to make those transitions happen. But it also -- EUV, the dry EUV resist, it's a perfect example of what I talked about with disruptive productivity. It will lower the customers' cost to implement EUV and therefore, should actually accelerate some of the EUV adoption. It makes it easier for the customer to pull it in for more layers and take the benefit of EUV. And that's why we think it's an attractive solution and it's meaningful for Lam from a revenue perspective.

I want to switch gears to the services business. I think the investment community applaud the fact that you're finally breaking that out on quarterly reports. It's a great business. I think that the overall growth in the business is well understood. Perhaps what's less well understood is how well you've done just growing your revenue per chamber over the last several years and your expectations going forward. What's driving that? And how do we think about the growth in services versus the growth in WFE for you?

Yes. Okay. So it's a great point. And that business -- I mean I'm glad we're now disclosing it simply because it makes it easier for us to talk about what we think is a really important part of our business going forward. I mentioned this idea of -- the install base, it's a platform. I mean it's 60,000-plus chambers and growing. As our new system business does even better, we add to that install base. It gives us more chambers on which to generate revenue. And like you said, we have a strategy to increase the revenue derived from each of those chambers. And the way we do that is we partner with customers and we look at how we can lower their cost of manufacturing through productivity upgrades, things that make the tools run faster, require less service, consume less parts. Those productivity upgrades are -- we've been doing extremely well in selling those to customers based on the results they deliver from a cost perspective for the customer. So it's a simple ROI evaluation that the customer does. Technology upgrades allow the customers to take those 60,000 chambers and continue to advance them forward and get more life out of those assets that they've already invested in. And so we've seen very strong demand always for how do you extend the life of my tool? I've already purchased it. I want to [indiscernible] the technology upgrades. And then the services side, I mentioned the advanced services, Equipment Intelligence, this idea of how do you install things on the tool that allow them to ramp to yield faster, require fewer people and fewer labor hours, less time to match all the chambers. Now you're not going to install 50 to 100 chambers trying to do the same thing inside one of these mega fabs. If we use big data and machine learning algorithms and other types of advanced database services, we can help the customer ramp, and that's, again, it's a productivity benefit for them. And so as I mentioned, underlying almost everything we do is about how to bring technology to bear on this node-to-node cost-reduction challenge that the customers face. And much of it is just because as you started out without wafer transitions, without consolidation, you have to get it through innovation around the equipment. And the services business is just one of those areas that customers look for productivity by definition. And so it's -- but maybe the final comment on this. When I talk about this as a platform for growth, what's great about this business, we're really the only provider of those services and upgrades because we're the experts on the equipment that we've sold. And so it becomes a large sea of tools, sea of opportunity for us that's only limited by how much we can innovate around what upgrades and services we can add for that 60,000-plus chamber install base.

Tim, we're coming up to the end of this session, but I wanted to ask one question to at least try and at least on a qualitative basis, maybe steal some of your thunder from the January conference call. And that is, how should we as investors be thinking about WFE for next year? I mean clearly, you saw another strong year overall this year led by foundry, led by a recovery in NAND in the back half of the year, DRAM, still kind of at cyclical lows. But how do we think about sustainability or even further growth next year in broad strokes?

Yes. I mean in broad strokes, because I mean we normally use the January earnings call, as you noted, to give our annual WFE outlook. So I probably won't give the actual numbers. But what we [ have ] said and you're hearing from our customers about their optimism and positivity of demand, so that's a pretty good sign of -- for us. But we've said foundry logic, I mean, the explosion of demand, we see foundry logic continuing to be strong into -- as we go into 2021. NAND has definitely picked up in 2020. And we've said that we still exit the end of the year, though below what we believe is long-term demand on a supply basis. So we think NAND is still entering the year strong. And DRAM, I think the comment we've made is, like you said, cyclical low. It's 2 years of relatively little investment, and we think 2021 is setting up to be a much stronger DRAM year. And so if you have good foundry logic into the year, good NAND and DRAM recovering, I think it tells you that we're fairly optimistic about 2021 WFE without previewing our January '21 earnings call too much just yet.

I had to try. With that, Tim, I'd like to thank you for joining us this morning and having this conversation. I want to express all of our wishes that you, your immediate family and the extended Lam family stays safe and healthy in what's been a very challenging and trying 2020. But thank you very much.

Thank you, John, and same to you. Thanks.