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

Well, good afternoon, good morning, everyone. Thank you for joining Evercore ISI's Inaugural TMT Conference. My name is C.J. Muse, semiconductor, semiconductor equipment analyst here. Very pleased to have Gary Dickerson, CEO and President of Applied Materials. I've known Gary for a long time and I very much look forward to our time together. So Gary, welcome.

Well, thank you for inviting me, C.J. Really excited to be here.

Good stuff, good stuff. So I've got a lot of questions here in front of me, but perhaps I'll just kick start with one of the questions where you can kind of take it both near term and long term. This is probably the first time in the history of the equipment industry and perhaps the semiconductor industry where we're hearing CapEx budgets that are looking out 3- to 5-year kind of time frame. So curious, what are your thoughts around that? What are the dynamics that could perhaps change in the equipment industry as a result of it, perhaps lead times start to extend or commitments become contractual? Would love to start there and hear your thoughts.

Well, thanks for the question, C.J. So if I look at the overall industry, semiconductors are really at the foundation of the biggest inflection of our lifetime. And certainly, this last year, in the pandemic, we're changing how we work. We're -- my 11-year old twins, how they learn, retail has been transformed, and that's accelerating. You see the automotive industry, the content increasing of semiconductors and electronics and the very nature of competition, health care, everything is really transforming and that's accelerating right now. If you look at data, data has increased about 150 times over the last 6 years. And 2018 was the first year, machines generated more data than people. If you go to 2025, the projections are it's 99% machines, 1% people. So every industry is inflecting. And again, it just creates an enormous opportunity, I think, for the entire semiconductor industry and certainly for Applied Materials because this is at the foundation of a multitrillion dollar race for the future and certainly, national security, economic, employment growth, everything is really impacted by this transformation. Relative to the way our customers are thinking, if you look at the wafer fab equipment intensity, that has been rising from a low of around 9% to almost 14% right now. But if you look at wafer fab percentage -- as a percentage of profits, that's actually improved between 30% and 40% over the last several years. So there's certainly tremendous value being created as the complexity is rising, and this is really at the foundation of every industry. When I talk to my customers, we certainly see chip shortages today. Certainly, automotive is one that's front and center, and many people are looking at factories being idled because they don't have enough chips. And my customers are telling me, they are entering into longer-term supply agreements. So the whole thinking around just-in-time inventories, obviously, are not working. But I do think, C.J, that this is really the tip of the iceberg what we're seeing in automotive today. I think it's every industry. Every industry as we work from home, learn from home, health care, retail, everything from the etch to the cloud, 0.5 trillion, trillion connected devices that is really at the foundation. So again, I think you see our customers announcing multiyear investments. You see governments around the world announcing large investments and you see our customers entering into longer-term supply agreements, those are things we've never seen before.

No. And if I -- just one question near term. If I think about kind of the progression for wafer fab equipment estimates, clearly moving higher through the year. And I think you guys are on record that next year should be better than this year. But curious, is there kind of a magic number where you just capped the capacity that you can bring to bear from the [ loss from Singapore ] is done? Or do you think that you could supply whatever demand is out there this year?

So C.J, we talked about this year, next year, being over $160 billion relative to overall wafer fab equipment. I think the -- we -- I think leadership is a lot about having a point of view encouraged to act on that point of view. And so our -- we have had a very strong point of view about technology transforming every industry data being equivalent to oil, powering the industrial revolution. So if you wind the clock back 18 months ago and you talk to people inside Applied Materials, I was talking about '21 being a big year. And it was really on the foundation of the fundamentals I talked about earlier. Now of course, we didn't know about the pandemic at that point in time, accelerating this transformation, but we had strong conviction. And you can see our CapEx increasing to put in place the operations and supply chain capabilities. So we could ramp Applied to a higher level because if you go to our Investor Day meeting, what we showed there was back to 2012 and 2013, the 2-year moving average has been moving up and to the right. And certainly, with this explosion of data, the content increasing capital intensity increasing, complexity increasing, we have a strong point of view that this is going to continue the secular growth, not every year is going to be up. But if you take that 2-year moving average, we have a strong point of view. We did have a strong point of view about '21 and beyond. So we are putting that capacity in place now. We certainly cannot respond in a week or a month, but we're continuing to invest. So Applied is positioned as this industry keeps going larger. And I would say, C.J, also in talking to my customers, they feel Applied is performing and responding better and faster than our peers.

That's great. You talked earlier about rising cap intensity. I think about node migration. So you kind of covered the secular growth part of the story. But I think one of the most interesting questions that investors have is who pays for all of this, right? And so I think you talked also about how profitability for chip makers has moved out to the point where WFE as a percentage of EBIT margins is actually quite healthy. So obviously, they have some room, but there is a problem, right, where you can't sell a smartphone for more than X dollars, right? But obviously, there's also evolving business models beyond just consumer electronics. So how do you see the evolution there and the opening up of new markets and how that enables or what will be a hindrance to continued growth in semis and semi cap?

I think it's really at the foundation. And so really, C.J, when you think about every industry, and we showed this also in the investor meeting, whether it's a smartphone or a data center or automotive, the content is increasing. And it really is at the foundation of the competitiveness of every one of those different industries. So it's really about power, performance, cost, faster than others. And so the people talk about maybe $10 trillion of economic value being created with this inflection around AI big data in this decade, and it really is at the foundation of winning the future in every single industry. So when you think of it that way, if I have access to that data on the etch with tiny ears and tiny eyes that are, again, 99% of the data is going to come from machines, not from people downloading videos or those types of things. It's in everything around us, C.J, getting smarter. And again, it is at the foundation of competition. And that's also why, again, our customers, the percentage of investment versus profits is improving, that is really, I believe, the economic opportunity is at the foundation of global competitiveness. That's why you see the same thing from a government perspective and that focus. So -- and then for Applied, the other aspect of that is when you think about how do I win that future, it's about new materials, new structures, new ways to connect chips together, application-specific computing workload, architectures, new ways to shrink. Even our largest customers talk about how they double energy-efficient computing every 2 years, it's those 5 things that enable winning that data infrastructure for the future. And that's where Applied is very well positioned going forward with those foundational technologies and combinations of technologies that are really unique in driving power, performance and cost faster than anyone else. So that's the way I think about it. I think our market opportunities have never been bigger and better. That 2-year moving average we saw, we see that continuing to go forward. And it's really about capturing and enabling that future infrastructure, that's really where I spend all my time and focus.

Excellent. So that's the next section of the questions I have for you, technology inflections, your focus and opportunity. And early, you just mentioned a PPACt. And so obviously, that plays an important role. I wanted to kind of create a runway of how to think about your positioning here. So very near term, I wanted to try to drill down first on conductor etch. That's been a real source of strength for you guys over the near term. And I think you've talked about that prospectively as a positive. So can you drill into what's driving your success there? And what kind of runway of kind of growth above and beyond the market, should we anticipate for that part of your business?

So if I look at -- I talked about, I think on the last earnings call, conductor etch or etch and CVD, around $7.5 billion, growing 200-plus basis points relative to overall market share. And really, we've grown a significant amount in, especially memory and DRAM, over 20 points of share gain in conductor etch over the last several years. We're growing in foundry logic with our Sym3 platform. And also with our selective etch, the Selectra product, where we can remove materials with 1,000 to 1 selectivity and shaping these structures like gate all around or the capacitor formation through high-temperature etching, the capacitor scaling and memory. Those are the areas that we're focused on. One of the best products in the history of Applied Materials is our Sym3. If you go back to around 2012, C.J., we had 18 different etch platforms, and we made a bet, and we made a decision to focus on this new chamber technology that has symmetrical design, larger conductants, so we can remove the byproducts, as we're etching, all the materials that we're watching, faster out of the chamber, so they don't redeposit on the sidewalls of the structures and on the chamber to rain particles down on the wafer. We have unique coatings inside those chambers that have helped us gain significant share in foundry logic, and you'll see this as these future technology nodes ramp. So we have really great technology there in that conductor etch Sym3 platform and our Selectra platform. But the other thing, C.J, is really the cooptimization where we can optimize, we have more materials in films than anyone, the optimization of those materials with our removal technology and the other thing is our e-beam technology. So as we're optimizing and dialing in all of these different knobs on our etchers, on our deposition systems, we have, by far, the highest resolution e-beam platform, we're #1, over $900 million, 50% growth in that part of our business this calendar year. That combination, C.J, is very unique. So we have very strong technology in the unit process. We can cooptimize with a number of different technologies and creating materials, removing, modifying and then the time to market advantage with clear leadership in e-beam technology to map out those fingerprints and optimize it better and faster than others. And so again, the opportunity, we have seen significant growth in our etch business, there's a tremendous opportunity to continue that momentum going forward into the future.

So good segue on the e-beam conversation. Obviously, you've made great strides with your e-beam metrology tool. Curious, what are your ultimate goals for the optical business?

Well, again, PDC, overall, we -- I talked about 50% growth this year on top of 25% growth last year. E-beam being over $900 million. And again, clear leadership there, tremendous growth opportunities. In the optical business, we introduced a new product, Enlight, that really gives our customers about a 3x advantage in finding defects at more efficiently than other technologies. So we're seeing very strong pull from leading foundry customers, especially putting that technology into more points in the line to increase their learning rate to drive the T the PPACt. So that technology on the optical is also combined with our leadership in e-beam technology with something we called the extract AI, artificial intelligence, where we combine our leadership with the highest resolution, about 50% higher resolution with this e-beam leadership so that we can take the map from our Enlight optical system, 1 million different potential defects and look at that in real-time using artificial intelligence to get down to actionable insight the map that really matters for the customer in yield and accelerating their R&D and technology. So that combination is really helping us grow our overall PDC business. And the other thing I would say, C.J, that a lot of people miss is that certainly, 50% growth in that business is something that's a driver for us. We see that continuing to grow faster in the future. But optimizing these new materials, these new structures that gate all around, the 50% reduction in wiring, the high-speed memory with the logic-like structures in DRAM, that optimization with our e-beam leadership, mapping out those windows, so you get a higher yield with bigger process margins faster that's worth billions of dollars in combining those different technologies together beyond the 50% growth rate that we're seeing in PDC this year.

Very helpful. I guess segue over to the cooptimization side, I'm curious, I was hoping to talk about integrated solutions and see if in your mind, that is something that is different from cooptimization in the sense that as we get to these finer line with, you can't run a system in isolation. And you need feedback, feed forward loop, you need sensors, you need to bring AI to the table, data science. You have to focus on thermal, have a good idea signal versus noise. So how do you think about the integrated solution requirements looking out over the next 5 to 10 years? And is that just simply adding on to what you're talking about in cooptimization today? Or is that something distinctly different and another focus inside Applied?

So on the integrated solutions, this is really important, very, very, very important. If you look at -- go back to our Investor Day, we showed the leading logic, enabling technologies for the future there. We had 1 tool that has integrated PVD, CVD, ALD, surface preparation, integrated metrology inside 7 different technologies that enables a 50% reduction in wiring resistance. And that, C.J, the number of steps are increasing, I believe it's something like 40% between 5-nanometer and 3-nanometer. So that 50% reduction in the wiring resistance is enormously important. It's all about managing those interfaces, the electrical properties of those interfaces under vacuum. That is very important. And in some of these films, C.J, are a few atoms thick. When they go to atmosphere, they oxidize. The same thing happens with the transistor as you're building the next-generation FinFET or gate all around, those combined technologies, those integrated solutions are enormously important. So that's really what gives you power, performance and cost leadership in enabling these inflections faster than anyone else. The T part of the PPACt, time to market is really what you're talking about with these sensors, the metrology, the e-beam, combining all of that together, and I personally believe we're scratching the surface of that. When you think about it, if I'm building the new wiring or I'm developing my next gate all around transistor, it's really complex. There are tens of knobs on a single tool. There could be hundreds or thousands of knobs that I need to turn to get the best electrical properties and the biggest process window. That is enormously important. So using technology and AI is a very, very big opportunity and changing the way we were faster and better. And again, this is where Applied, I think, is also in a great position because we have many of these unique capabilities. No one has this unique eyes with our e-beam leadership, 50% better resolution, 100x faster to map out those fingerprints, no one has the same portfolio of these integrated technologies like what we showed in our Investor Day in the wiring. So it just creates an enormous opportunity, both for our systems business, but also for our service business. If we can drive that 2x faster with 30% bigger windows to get better yield, it's just an enormous opportunity. And we're just, I think, scratching the surface, C.J, of what's possible driving this through technology.

I guess maybe to drill down a little bit deeper into that, perhaps, we could focus on foundry and kind of 7-, 5-, 3-nanometer transitioning. So I think for Applied in the 7 to 5 opportunity for you all, I think, only grew low single digits. But curious in the move to 3-nanometer, it sounds like new etch [ steps ], different contacts, interconnect. And then also gate all around it at 2-nanometer with even more benefits. How are you thinking about your SAM in the move to -- from 5 down to 3 and 2?

So C.J., I have a very strong perspective. We work deeper and further ahead with customers than ever enabling the 3-nanometer, 2-nanometer, beyond 2-nanometer types of structures. And again, you go to our largest customers, when they show the road map to double energy-efficient computing in their presentations, they show the 5 things I talked about, the new materials, the new structures, connecting chips together in new ways, new application-specific architectures, and the fifth is the new ways to shrink. So this 50% improvement in wiring resistance is an incredibly important innovation like you talked about also, the FINs are growing taller for the FinFET, there are a lot of different technologies that we're driving to enable those structures and also those interfaces so you have the right electrical properties. And then you go to gate all around. So our -- if we were sitting here 2 years from now, I have a very strong perspective that the relative contribution to PPACt, power, performance and cost in what we do in materials, structures, enabling that road map, that relative contribution will increase at 3-nanometer and it will increase significantly more at 2-nanometer. And C.J, we have a Logic Master Class that is coming on June 16. I would really encourage you and other listeners to attend that Logic Master Class, and we'll talk a lot more about those specific opportunities. But there's 0 question in my mind that the relative contribution to power, performance and cost is going to increase significantly going forward relative to what we do.

That's great. I guess hitting on a couple of other growth drivers for [indiscernible] ICAP, I think, well over $3 billion this year. Just curious, from a structural perspective, where are we on the lagging edge supply side? And is this something where you have visibility to multiyear investments? Or is this kind of a one-and-done type of spend?

Yes. So the trailing edge, a little over 2 years ago, C.J., we formed a group called ICAPs, IoT, communication, auto, power, sensors. And I'm personally spending a lot of time. I've been spending actually right after this meeting, I've got a meeting with one of those CEOs, a lot of time focused on those markets. So it's maybe 1/3 of the total foundry logic business today, we believe it's going to remain very strong. When you look at this explosion of data, 0.5 trillion or 1 trillion connected devices, machines generating 99% versus 1% of the data, the content growing in automotive, the electronics content growing in automotive, there's no question. That's why we made that change in our organization more than 2 years ago. In Applied -- C.J., I know you know that this is not driven by 2D shrinking. This part of the market is really driven by innovations in materials and structures and packaging in all of those different areas. So there is 0 question in my mind that, that segment of the market, when you think about automotive or sensors that are going into many, many, many applications, that's going to be very, very strong going forward in the future in Applied with our breadth of technologies, being at the foundation of enabling that is also very, very strong in our overall margins we've communicated there are accretive to the overall company margins.

That's great. How about advanced packaging? Obviously, with desegregation in triplet-type strategies, advanced packaging led by TSM and Intel is clearly a very interesting area. And it certainly sounds like you're bringing your full portfolio to bear to this market. I think you've quantified that at roughly $750 million business this year, I think it was $500 million last year. Where could this be in 5 years? And what would be the key underlying product drivers?

So I think we've talked about even over $800 million this year, C.J. So no question, significant growth. I believe this is just a tremendous opportunity. When you think about the 5 drivers of power, performance and costs going forward, you can see, as you said, TSMC, Intel, NVIDIA, AMD, many companies talking about the importance of packaging, enabling power and performance. NVIDIA came out with a new GPU packaging design surrounded by high-bandwidth memory that gave them 3x higher speed, 50% less power, there is going to be tremendous innovation in packaging. And applied is in a great position -- think about taking this wiring technology that -- where we have clear leadership in the front end into packaging. So we have strong position in PVD, CVD, CMP, plating. We have the new Sym3 via etch, where I talked about on one of the earnings calls recently, we won the business at one of our largest customers. We announced a new technology called hybrid bonding, where you can bond 2 chips together to shorten the length of that wire and improve -- increase your IO density and improve power and performance. We're the only company that has a full flow packaging lab, and we have some of our leading customers placing engineers in our packaging lab. Of course, we're segmenting IP relative to what we're working on with some of those customers. But it's just a tremendous opportunity. And relative to the fundamental technologies that enable power performance through connecting chips, chiplets, IP blocks together, that's where Applied has tremendous strength. We are expanding, like I said, this one technology around hybrid bonding is working in alliance with another company. And there are other innovations there, C.J, that will give us tremendous opportunities. I don't want to quantify exactly what that's going to look like. But I think that will be my own personal view, higher growth than what we're communicating relative to the overall growth for the company. I think it's a tremendous opportunity for Applied and for the entire ecosystem.

Very helpful. So we have about 5 minutes left. I guess would love to spend just a little bit of time on the service side. As you think about adding more value to your customers and knowing what the ASPs are, the various tool sets that you're offering. How does that impact your vision for growth for service revenues? And as part of your thinking there, is there a razor blade model here that -- where you'd be willing to maybe take a slightly lower price to get the installed base up to drive that annuity? Would love to hear your thoughts there.

Yes. So VLSI Research publishes the overall system market share, wafer fab equipment and the service share. And it's really, C.J., the only place you can go and where everything has to add up and is fully transparent. So if you look at that data, what that shows is that Applied has the largest percentage of our for pure service business, not combining other equipment into that overall number. Our growth, our strategy has really changed significantly over the last several years. We really focused on these longer-term agreements, the subscription revenues, and that percentage of our total business has grown from around 30% to 40%, to over 60%, in the latest quarter 70% of our orders are these long-term agreements, multiyear agreements with customers so the quality of our service business has grown. The renewal rates are in the range of 90%, very strong pull with our customers. The focus there has been, to a large extent, optimizing the cost, yield and output for our customers in high-volume manufacturing. So they're optimizing their value capture over the life of a technology node. I think, C.J., the opportunity is with technology, we touched on this some with this applied actionable insight acceleration, AIX, it's just an enormous opportunity to accelerate R&D to deliver bigger process windows through combinations of technologies, that really, I believe, is in the early phases. We showed that in the investor meeting, the opportunities there. As you can imagine, complexity is increasing. When we're delivering these tools that have ALD, PVD, CVD, all of these different things combined in a single platform, optimizing that through technology. And that was one of your first questions early on is enormously important. Creating a bigger process window across all the tools in high-volume manufacturing gets higher yield. That is enormously important. So again, we have seen a big shift in growth in our subscription revenue, those long-term agreements. That also -- every one of those tools under those long-term agreements, the revenue per tool increases something like 3x. So that growth also has a multiplying effect relative to the overall compound annual growth rate and the biggest opportunity from my standpoint really is this the T in the PPACt, accelerating time, better and faster is worth a tremendous amount to our customers.

Well, I think we've exhausted our time together. Thank you.

That's too short.

Always enjoy speaking with you and thank you for taking the time to be with us today.

Thank you so much, C.J. Really great seeing you.

Thanks, Gary, be well.

You, too. Bye-bye.