Applied Materials, Inc. (AMAT) Earnings Call Transcript & Summary

Hello, everybody. Welcome back. Just a quick safe harbor. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. So with that, I'm Joe Moore from the Morgan Stanley semiconductor team. I'm very happy to have with us today Dan Durn, who's the Chief Financial Officer of Applied Materials. Dan, welcome.

Hi. Hi, Joe.

So maybe we could start with the bigger picture. You guys have the sort of statement on recent earnings calls that the future is not like the past when it comes to Moore's Law and sort of 2D scaling. What do you mean by that? And that seems like kind of the driving vision behind Applied. What does that mean for Applied Materials?

Sure. So let me share a little bit of the thinking, at least how we think about it internally. I think there's a couple of elements to the future doesn't look like the past. You touched on one of them, traditional Moore's Law through 2D shrinking. We're running into physical limits, and that's no longer working for the industry. Implication of that is there's going to be a new playbook that drives the power performance road map for the industry. It's new types of materials, new types of architectures, going vertical on the chip, finding new ways to connect chips together through packaging technologies. There's a whole host of vectors that the industry is now turning to drive the power performance road map. What's interesting for us as a company is given how we're positioned across end markets and our product portfolio within those end markets, we're extremely well positioned to drive each and every one of those vectors, and they're increasingly materials engineering enabled. The second thing I'd say about the future not looking like the past, I think we're in the middle of a handoff of end market demand in semiconductors from consumer-oriented devices to something that I think is far larger and more substantive over time. And there's a real diversification of the end markets and whether it's leading edge, trailing-node geometries, Internet of Things. There is a much larger investment wave that we're in the very early innings of that are going to play out over the course of the next decade plus that we're really encouraged by. And it's decoupling of that demand from consumer behavior to the largest companies as opposed to the largest industries because their ability to monetize those investments is greater today than it's ever been. So we're really encouraged on both elements that support that thesis that the future is going to look a little different for this industry.

And the other sort of talking point is that your PPACt framework, which I think is performance power area costs and time to market. Can you talk about what that means? And I guess as I see it, thinking as an investor, Applied has a lot more breadth of solutions that you can offer versus individual process steps. And you guys are sort of talking about the device structures a little bit more than we hear from your peers. Can you just talk a little bit about how Applied is approaching that?

Yes. So we do have a broader footprint. That's both balanced within those end markets, foundry/logic, DRAM, NAND, but like you said, the broad portfolio. And as you think about the power performance road map, I think power performance area cost has always been a part of the industry's vocabulary. It's always been there. What's different today is the introduction of the last letter, T, representing time. And I think the industry has gotten a real lesson here over the last several years on the importance of time. Not only it's the technology development, but when you have a new node coming online. And you go from insertion yield of that new technology to entitlement yield. The return you generate on the investments at that new node are heavily influenced by how quickly you can come up that yield curve to entitlement. So there's an opportunity with the breadth of our product portfolio and the way the road maps are inflecting to play a very material role. It's one of the reasons why we're very encouraged not only with the significant outperformance we see in today's environment. But as we look forward, node over node, whether that's foundry/logic or memory, our ability to increase our opportunity node over node is probably better today than we've seen at any point in the past. That product breadth, the insight we get on those customer road maps and the way the road map -- the power performance -- PPACt road map is being driven is increasingly influenced by the things that we do not only from point capabilities, the innovations we bring to market, but how we bring those capabilities together in the form of solutions to help our customers deliver that road map and do it in a very time-efficient way. I think we're better positioned today than we've probably ever been at any point in our history. And so I think when you hear the company speak and you hear Gary speak about the devices and the inflections and the way the company is stitching together capabilities in the form of solutions through our integrated materials solution offering, something we get very excited about when we see the node-over-node progressions and our opportunities set against that market backdrop, we feel very encouraged by what we see.

Great. And so before we get into Applied markets and revenue forecast, talk a little bit about just the role of semiconductor equipment in kind of the global trade ecosystem and a lot of focus in the last couple of weeks around what a supportive U.S. government executive order may look like. I won't ask you necessarily to speculate about that. But just where are you guys seeing it? It seems like a pretty big opportunity long term for equipment vendors because we've sort of shown how critical everything is globally to have access to you guys and your peers. And it also seems like a pretty big growth opportunity, to the extent that it drives different spending in different geographies.

Yes. So I'll share with you an institutional view that's got 2 elements to it. The first is how it influences the aggregate size of our markets and business opportunity and then a bit of the importance that semiconductors and semiconductor capital equipment plays in the broader ecosystem. So as you think about localization of supply chains, and I think this is going to be a trend out of the current environment and whether it's geopolitically driven or pandemic-driven, I do think there's going to be a trend towards localization of supply chains. Semiconductors is one industry, but I think you're going to see that trend play out in multiple industries. And when we think about government investments to support that thesis, from an overall equipment standpoint, I have a strong belief. Global marketplace, you will have global balance between supply and demand. The global supply statement, supply will be put in place to meet demand. Capacity will meet demand. Now the question is, is through government initiatives, it's geographic location of that supply. And so I don't think there's going to be a step-function increase in our overall capital equipment market. But what I do see is the way in which that capacity is delivered being smaller, more regionalized facilities as opposed to a very large, centrally located facility. The implication of that is, is smaller, les- scale facilities. It's less capital-efficient over time. So I do think there's going to be an upward bias in our overall WFE equipment market, but it's not going to be a step-function change because global supply will meet global demand. From a services standpoint, this is where I do think we're going to see a more substantive impact. When our customers build factories outside of their home geographies, we've got an opportunity to play a larger role in making those facilities successful because it's outside of a localized ecosystem. And we've already seen this to date, and we'll see this more over time. That is going to provide a higher service entitlement on that capacity. So we see this trend being a nice uplift from a services standpoint. That's the first element of the answer, which is impact to our semi-related businesses. The second element is I think there's a renewed appreciation with the strategic importance of semiconductors and capital equipment to overall economic growth, productivity gains within economies and national security. And so I think it is a real strong endorsement of how truly important these industries are to long-term economic growth and development. And I think that sets this industry up well to become structurally larger over time and will move our full appreciation of the importance of where things like semiconductors and hardware sits in a broader tech ecosystem that's critical to enabling a lot of the secular trends shaping the world.

Great. That's helpful. There's a lot of good points in there. So maybe if we could talk a little bit about the wafer fab equipment market. You guys have talked about -- maybe you're a little bit more optimistic than some of the forecast of high 60s. Certainly, it seems like we're going to be north of 70. And as somebody who sort of tends to model this stuff quarter-to-quarter, I'm sort of surprised. We were in the mid-30s 5 years ago. Semiconductors grew 50%. WFE grew 100%. And yet, there's pretty clear evidence that the 60 billion-plus last year wasn't enough. I mean we're dealing with shortages everywhere we look. So what do you think has driven that number higher relative to what semis have been? And again, you sort of alluded to the future looking right relative to that. Is this really the baseline that we're coming out of a recession, 60 billion WFE and we grow from there?

Yes. So I think this is a dynamic we've been fairly consistent on over time. And so I think the way in which I answer this should sound a bit familiar. What I like in today's environment is we're seeing data points that support what we've been saying for quite some time. As you think about how the semiconductor industry has evolved over time, we initially had a PC wave that developed the baseline level of demand. And then you layered on top of that consumer-oriented mobile handsets, layer in different level of demand. That gets us to where we're at today. And we're in the very, very early innings of what is beginning to take off in this third wave, which, again, I think, is far more substantive and it's nonconsumer discretionary driven. It's the largest corporations exposed to the largest industries that are finding ways to monetize these capabilities and enhance their competitiveness and serve their customers in ways that they haven't been able to in the past that makes their businesses more resilient. It's nondiscretionary. I also call it existential companies that make the investments will compete and win. Those that don't, I think, will cease to exist. And the magnitude and size, it's just an important handoff from a consumer-oriented device. And so I think there's long-term secular trends that are beginning to play out. Again, we're in the early innings of this third wave. The other observation I'd make about semiconductors is, I think, historically, you saw a tight coupling between population and consumer behavior. There's going to be an element of that, that persists. And regardless of what region you talk about, semiconductor consumption per capital, it's greater today than it was a decade ago. And a decade from now, it will be greater than it is today. And that's going to play out in just about every single geography. It will play out at different rates, but it's going to play out in every single geography. On top of that, I think for the first time in the industry's history, you're going to see a decoupling of semiconductor demand. And it's going to decouple from population growth and consumer behavior where 90-plus percent of data is being generated machine to machine and corporations' ability to monetize this and change how people live their lives and improve health and safety and well-being. There's a number of secular trends that are going to play out here in the next decade plus. Again, we're very, very early. We're just touching the surface of this, where you're going to see that decoupling, and that's going to be a nice growth adder for the industry. So these trends have been playing out. You're laying on top of that capital intensity increasing. You think about wafer size transitions. Our industry was no growth cyclical in the decade cut in of 300-millimeter wafers. Capital intensity -- WFE intensity bottomed in 2013, and it's been on an upward climb since. We think that serves the industry well. And then when you think about the change in the road map from just a 2D-shrink-driven world to something that's broader, more diverse, at a time when the demand for compute power has never been greater, it's one of the reasons why -- many of the reasons why we really encouraged by what we see and why we will say the opportunity in front of us is never greater in the history of this company than it is today as we look forward and these trends play out. We're super encouraged by what we see.

Great. And then just one other WFE question. Your China commentary, I thought, was quite interesting, talking about China sort of sovereign customers growing by a few billion. And I think we sort of know what the biggest foundry, DRAM and NAND company are doing, and they're not up that much in aggregate. In fact, they have been down a little bit in aggregate. So obviously, there's a breadth of customers there that are growing in China. Just maybe talk to that a little bit. And it seems like Applied is pretty well served when it comes to -- pretty well positioned to serve new customers when you have new and emerging customers given the breadth of your product portfolio. Can you just talk a little bit to why you're optimistic on China this year?

Yes. So what we see this year, again, is another step in the journey that's been playing up for quite some time. We call it slow, steady development of an ecosystem, investments in technology road maps, modest deployment of capacity behind those road maps. And we just see it as a disciplined evolution of the market. The other observation I would make is that the Chinese local domestic market is evolving in a way that mirrors, I think, what we see happening globally. And we've been talking for quite some time about foundry/logic demand diversifying between leading edge, trailing node geometries. And the growth profile of these specialty nodes over time is actually greater than what we see on the leading edge. And what you see is a mirror of that in the Chinese domestic market where you're seeing a broadening and diversification of the demand profile. I think there's 4 traditional customers that investors like to think of, 1 in the NAND market, 1 in the DRAM market and then 2 in foundry/logic, 1 more trailing node, 1 more leading edge. And it creates this framework around how to think about domestic China spend. What I would say is, is we're seeing a diversification of that spend that's broader than just those 4 customers. And I would just mirror back to about a year ago, the debate inside the industry around foundry/logic was TSMC had spent a nice portion of their budget earlier in the year. Isn't the foundry/logic market going to roll over? And we took a strong point of view at that time that's a no. The spend profile has been to diversify and be handed off to other market participants, and it's going to be strong throughout the year. What you saw play out in foundry in last year, you see playing out in the China market this year. There's a broadening of the customer base. There's the diversification of spend. There's spend on the leading edge. There's spend on trailing-node geometry, specialty nodes. And I'm encouraged because it's a slow, steady development of that ecosystem. And it's a broadening of the customer base in a fundamental way that I think will speak to the long-term health of that ecosystem as it continues to grow and develop.

Great. You were absolutely right about foundry. I mean certainly, to me, it was -- there was this math of the front half loaded foundry year last year, and it did not play out. And in fact, the shortages that we're seeing now in foundry are so significant. And we just finished with AMD a few minutes ago talking about their tight supply of cutting-edge foundry and then you talk to several automotive suppliers who are seeing shortages of trailing edge foundries. Is it -- at this point, it seems like we don't really need to overthink it. I mean if there's not a lot of shortage, there's going to be pretty strong capital investment. So it seems remarkable to me that in the wake of HiSilicon going away, that foundry is this tight. But it certainly seems like it is and it certainly seems like it's durable, at least through the balance of this year.

Yes. And so that would be our point of view. We talked about how well the company performed in 2019. We talked about our business being up over -- our systems business being up over 26% in 2020 -- calendar year 2020. And that's against the backdrop of a market that was up mid-teens. Let's say it land at around 60, up mid-teens. We see that strength following through in 2021. We see foundry/logic continuing to be strong. We see DRAM from a growth standpoint firing this year. NAND, we see some investments early in the year, but being a bit more muted from a growth profile standpoint but still a healthy level. And as we look forward into 2021 -- I mean, forward into 2022, we see that strength continuing. We've got an investor meeting coming up on April 6. We'll share more about what we see in WFE in 2021 and 2022 and beyond. But we'll give people a sharper perspective of what we see in that near-term window defined by '21 -- 2021 and 2022 and also give a longer-term view. The strength we see in the current environment, we don't see it as a 1 quarter, a first half, a 1-year phenomenon. This is something that's been building over time. We don't think this represents a peak level of spend. We've got a semiconductor industry that's approaching $500 billion. It's going to go to $750 billion, and it's going to go to a $1 trillion. It's going to happen in the next decade or so, give or take. And the supply -- the capacity statement to support that level of demand and that level of growth is going to benefit this industry in a material way, and it's going to benefit Applied in a very material way. The industry was $40 billion, went to $50 billion, $60 billion. Our view is over $70 billion in this year. You will see $80 billion, $90 billion. It will be over $100 billion to support the growth of the overall semiconductor industry over time. So we feel really good about the material opportunity that sits in front of us and the momentum we have as our company-specific momentum against that market opportunity. We just feel really good about how well we're positioned.

So can I quote you on the $100 billion? And when is that going to happen?

So if I were to put a time frame, I'm not going to make a prediction, but there are certain analysts that are beginning to write around 2025. I got no reason to disagree with that perspective and point of view. But we're not going to put a stake in the ground, certainly not in this forum. We'll say more about our expectations here on April 6. But if you take a look at some of the analysts that are beginning to write on what that looks like over time, I would say we're pretty much in line with the consensus view on that.

Yes. Okay. And then the strength in trailing and foundry seems pretty significant as well. And I know that's been a focal area for Applied to sort of make investments in not just cutting edge, but across the whole spectrum of process technologies within foundry. Is that going to help you this year in terms of your performance versus -- I'll get to your performance versus WFE. But is that going to be a help to Applied?

Yes. So absolutely. If you think about specialty nodes and you think about what Gary said on the earnings call, we see significant growth in that market. We sized that the specialty nodes at over $3 billion of revenue for us this year. So it's an important part of the business. And we've got a broad suite of technologies, it's key-enabling technologies. We've got an opportunity to not only drive the highest performance characteristics of devices on the wafer, but we can also influence the speed at which our customers bring those new nodes and that new technology to market. And so wherever our customers are creating pull from an innovation standpoint, we're going to meet them in the market key-enabling technologies that enables them to be successful. We've got a great portfolio of technologies to serve the specialty trailing node geometries. We've got a great suite of technologies to meet our customers' needs and drive inflections on the leading edge technologies. And so again, as that spend diversifies, we're really well positioned with the suite of technologies and the innovations we have in the pipeline to make our customers successful. So we will meet our customers where there's demand.

Great. And then just my last WFE question. I mean your ability to continue to outperform WFE, you've done that for the last couple of years. Sounds like you're pretty optimistic about the future. But just thinking, in general, is that the way you're going to drive the forecast as you did a few years ago? Or is it going to be a little bit more focused on which segments do well? There's a number of ways you can cut that. But obviously, you've done well relative to WFE. How are you thinking about the next few years?

So I think as we run our business, given the competitiveness that's embedded in the end markets, we'll always look at our business through the lens of key-enabling technology and innovation that can drive our customers' road maps. We take care of that. We will outperform the end markets. We'll drive share gains. We'll drive the profitability, and we'll drive value creation for our shareholders. So we'll always look and run the business through the lens of innovation because that -- making our customers successful is what drives everything else. We will create the shorthand for people as proof points we're doing the right thing from an innovation standpoint to talk about relative growth rates in our end markets. In 2019, the overall capital equipment market was down almost 10%. Our business was down a little over 2%, and our 2 closest peers were down 12% and 13.5%. So we significantly outperformed in 2019. When you think about the innovation we're continuing to drive to shape our customers' road maps in 2020, the market was up mid-teens, so landing at around $60 billion, up mid-teens. Our systems business was up over 26% against that opportunity. Foundry/logic undergrew the overall market, and we outgrew that end market almost 2:1. DRAM, we significantly outperformed at 27%. Our systems business is up against that market opportunity 27%. Our 2 closest peers were up a few percent and down a few percent against our up 27%. I don't know of a stronger proof point of delivering key-enabling technologies to our customers than those statements of strong outperformance in '19, 2020. And based on what I see around our innovation pipeline, the way we're serving our customers on a node-over-node basis, whether it's memory, foundry/logic, I fully expect that outperformance to continue in 2021. And again, we have confidence in the follow through both from a market momentum standpoint into 2022. We'll share more of that here in a few weeks at our Analyst Day -- or the investor meeting we're going to hold on April 6. But I feel confident that we're doing the right things based on that market momentum we see, as proof points we're doing the right thing in our innovation engine, the things that fundamentally drive success from our customers' road map and their ability to deliver that road map in a timely fashion so they're driving outsized economics for their customers and their investors.

So we've got about 5 minutes left, and there's a lot that I still want to ask you. Maybe touch on display, and then services. So display first. The markets seem pretty healthy. TV prices are seeing really strong -- OLED attach is good. You've talked about that. And from a quarterly basis, it looks like there's kind of growth, but you're still looking at kind of a flat year. Like what does it take to see that display business inflect more meaningfully on a full year basis?

So we pointed out, we see a number of green shoots, and we're tracking them. I think where you set expectations in a market like this are really important. And we want there to be a bias to the upside off of those expectations. And so I think what you'll see is from where we've guided our fiscal Q2, you're going to see some nice growth into the back half of the year. We expect a cyclical bounce to a more attractive point in the investment cycle in 2022. And so when we show strength into the back half of our fiscal '21, I would expect there to be follow-through into 2022 at those elevated levels of investment versus where the industry sits today. So right now, the setup around that business for us has been bouncing along the bottom for a while. But the setup from this point going forward looks really good in terms of adding to the growth rate, adding to the cash flow profile of the company and be a meaningful driver for the business going forward. So encouraged by the green shoots. We'll continue to watch the markets evolve. But from where we guided in Q2, I would expect to see some nice growth as we profile into the back part of the year and into 2022.

Okay, great. And then maybe talk a little bit about services. I know it's a business that you're quite positive on and investors are quite positive on, and in particular, the progress that you've made in terms of growing that opportunity relative to the installed base that's out there.

So services, and you rightfully characterized, this is a business we're really positive on. It's got a nice, steady cash flow profile, revenue growth profile. It's a real nice adder to the business. There's an intersection between the services business and our ability to influence the T part of the PPACt road map, power performance side of it. There's an intersection between what we do in the hardware side and the service side. What's great is we've grown this business over the last handful of years at 2x the rate of the growth of our installed base. So clearly, the team is doing something really right from a strategy and execution standpoint. We would expect to continue to significantly outgrow the growth of the installed base. You've seen long-term service agreements. The penetration rate of those increased substantially over that same time period. We've taken it from around 40% to over 60%. We would like to increase that percentage over time. And we think there's a nice opportunity to do it. And then embedded within those long-term service agreements, you're seeing the tenor of those agreements and what our customers are willing to sign up for begin to extend out in time. Virtually all of the agreements a few years back were 1 year in duration. Today, a full 1/3 of them are beyond that 1-year time horizon and extend out into the future. So we like the way that business is profiling, strategy, the execution in good years and bad years, our installed base will continue to grow. And we will see this business virtually grow each and every year as we look forward. And then from an increased complexity standpoint, this is, again, just another nice trend and tailwind for the business. The new nodes we bring to market have far tighter process windows and performance tolerances. And so the service entitlement on equipment we ship today is significantly greater than what we used to ship, say, even a decade ago. And so that has a nice upward trajectory as that level of complexity continues to grow inside of the industry. Not only you're outgrowing the installed base, you've got to grow in the complexity, both of those create a nice tailwind for the business, and then execution against the strategy to create more subscription-like revenue. As those trends play out, I mean, we think that business is really well positioned to drive value for shareholders and the company over time.

Yes. That's a great business. So maybe we can just squeeze one in from the webcast. The -- so I think it's an important question. Kokusai deal is supposed to close in a few weeks. Is that process kind of moving forward? And maybe just remind us what the situation is there.

Sure. So we've got one regulatory authority remaining. We've been engaged constructively for quite some time, as you can imagine. We understand all of the questions that have been raised. All of the questions have been addressed. And the way I describe it, it's just decision time now. We've got a couple of weeks left. We'll see what happens from a decision-making standpoint. I think there's a low probability of extension beyond March '19. It's not 0, but it's a low probability. I think March '19 is a key date. And all the questions have been answered. I just think it's decision time. We'll know here within a couple of weeks. And certainly, when we're in front of investors during our investor meeting on April 6, we'll have visibility on where things stand.

Great. Okay. Well, we'll wrap it up there. Dan, thanks very much for your time. And thanks, everyone. And sorry, I didn't get to everyone's questions on the webcast.

Thanks.

Thanks.

Thank you.