All right, cool. Good morning, everyone. We'd like to get started. Thank you so much for joining. My name is Toshiya Hari. I cover the semiconductor and semi-cap equipment space for Goldman Sachs. Very honored, very excited to have Gary Dickerson from Applied Materials. So just a brief intro of Gary for those who do not know him, I'm sure everyone does. He's one of the longest-tenured CEOs in the semiconductor industry. He began his career over 35 years ago after studying engineering management and business in Missouri. His career has given him deep exposure to all of the key chipmaking technologies. He was a lithography section manager at AT&T.
He also spent 18 years at KLA, where he and his teams developed process control technologies that helped increase KLA's revenue from $80 million to $2 billion a year. He became president and COO of KLA, and then he served as the CEO of Varian Semiconductor from 2004 through 2011, during which he helped the company's market cap grow by about 7x. Just this week, Gary celebrated his 10th anniversary as CEO of Applied Materials, the largest provider and the broadest provider of semi-cap equipment. And over the past decade, I believe your market cap has expanded by more than 10x. So congratulations for all that.
Oh, thank you. Thank you.
That's amazing.
Really happy to be here.
Yeah, thank you for the time. So Gary, I definitely want to spend most of our time on long-term strategic elements of your business, but I did want to kick off with a near-term update. You reported very solid results for the recent quarter. You provided guidance that well exceeded market expectations. At a high level, what are you seeing in your business today?
Well, we're in a really strong position in the areas of the market that are growing the fastest. So, in the last quarter, of course, ICAPS has been a really strong business for us. That's IoT, Communication, Auto, Power, Sensors. That business, that market, is doing extremely well, and our position within the market is also very strong. The leading-edge Foundry Logic is also an area that more and more of the dollars are moving to materials innovation, and Applied is extremely well-positioned there. DRAM is a market where we've gained 10 points of share in the last 10 years, so that one is another one that is fueling our outperformance in the near term. And then packaging is another area that's very, very strong for Applied. And we also delivered record services growth, even though the overall market is weak.
Got it. Gary, as sort of we talked about earlier, you know, you've been in the industry for a very long time, and you've experienced many cycles. I know you're not in the business of predicting cycles, but as you compare and contrast this current cycle with past cycles that you've lived through, what are some of the similarities? What are the fundamental differences that you can point to?
Yeah, I think, you know, if I look back through my career, if you look at the transition from 200 mm to 300 mm, before that transition, the wafer fab equipment capital intensity was around 17%, and then you had this transition where you were able to achieve 2.3 times the number of chips per wafer in the transition to 300 mm. So in that timeframe, a little over a decade, wafer fab equipment was really no-growth cyclical. So that was one big factor. Another factor was the pervasiveness of chip demand. So you had PCs, the computer on your desktop, and people were waiting for the Microsoft upgrades, and so that was, again, the 200 mm to 300 mm. You had the PC market, then you went to mobile social media.
So everybody has a computer or camera in their pocket, and every year around the holiday season, you know, you had everyone focused on hitting that timeframe, and the pervasiveness went up dramatically. And then now you have everything around us. I think everybody reads about chips in every industry, transforming every market, 7,000 chips in electric vehicle. If you look at an AI server versus a industry standard server, enterprise or cloud, you have 8x the amount of Foundry Logic chips in that server and about 8x DRAM. So those markets, and we're still in the early innings of that transition to those AI servers. So, you know, if I look at, you know, our position, you know, Applied is outperforming this year.
This will be the fifth year in a row we've outperformed overall wafer fab equipment spending, but we're in a very strong position in the areas of the market that are the strongest. The other thing I would say, maybe different than the past, one of the things we did in our service business is we transitioned from more transactional service, break-fix for parts and labor, to certain subscription agreements with our customers, and that's also giving us strength in this time period where the market is weaker, and that's how we were able to deliver the record service results.
Right. Right. Gary, you talked a little bit about the capital intensity of the semiconductor industry. It was very high prior to the 300 mm transition. It came down, and it's coming back up again. I think, you know, when industry folks talk about the semiconductor market and its size, there seems to be a consensus around $1 trillion in 2030. I think there's less of a consensus on where capital intensity is headed. Can you share, you know, some of your thoughts on that?
Yeah. I mean, We're deeply, deeply connected to every single customer, deeper than we've ever been. Five years ago, we formed the Integrated Materials Solutions Group, and for leading-edge Foundry Logic and memory, and then also about four years ago for our ICAPS business, the trailing-edge business. So we're really deeply connected with all those customers, many technology nodes out, not just with the equipment engineers, but also, again, the device integration engineers. So we have a lot of visibility in, you know, gate-all-around, backside power delivery, CFET technology, all of those big inflections that will happen, over the next several years. And one thing that is absolutely happening is more and more of the roadmap is enabled by materials innovations.
If you look at all those things I just talked about, you know, all of those are areas where Applied has 50% share, or in some cases, even more than 50% share of those inflections. So, you know, what we see is customers are really all competing on power, performance, and cost. We call it PPACt, power, performance, area, cost, and time to market. That's the race they're all running, and we're just in a really, really great position as they're driving all those materials innovations in the structures. Capital intensity, you know, it's kind of mid-teens right now, and, you know, I would say that, you know, going forward, we see, you know, probably high teens in terms of capital intensity. As customers are really, and more, I would even say more important than that, is that more of that is moving to materials innovations.
It's very clear in leading Foundry Logic. Of course, ICAPS is already all materials, packaging is already all materials, and you'll see memory moving in that direction, you know, further out in time, towards much more of this incremental spending coming from materials innovation.
Gary, you and your team, you know, you were extremely early in focusing on the notion of PPACt.
Yes.
I feel like your customers are increasingly talking about PPACt as well.
Yeah.
Is that resonating across your entire customer base?
Oh, absolutely. So if you look at Mark Liu, Mark Liu talked about his roadmap for energy-efficient computing for TSMC. He talked about five, five drivers for that, and those are all the same things that we've been talking about for the last many years. You know, they have new materials, new ways to design technology, co-optimization, 3D building of structures, all of these things that, that we've been talking about for years, and are real strengths of Applied Materials. We have a very broad portfolio, so when you go into some of these inflections, gate-all-around, backside power, again, we are delivering the most critical technologies for those inflections.
So you, whether it's TSMC or AMD or NVIDIA or, or any of those companies, when they talk about what enables their roadmap going forward, you know, they're very much aligned with what, with what we've been talking about for the last several years.
Got it. Maybe double-clicking on gate-all-around and backside power delivery, from your vantage point, where are your customers in those transitions? And more importantly, what do you enable and what's sort of the opportunity set for Applied Materials?
Yeah, so on the transistor, which is Gate-All-Around, going from FinFET to Gate-All-Around, in FinFET, we're close to 50% of the spend, and what we've said is that we'll be over 50% of the spend with Gate-All-Around. Again, this is all materials innovation with technologies like Epi. There is selective removal and shaping those structures. There are double-digit number of steps in those material shaping technologies. We have the majority of those steps at all of the three leading customers. There are material modification steps in that flow, and then another area where we have tremendous strength is E-Beam. So we have around 50% share, maybe a little over 50% share of all of the different E-Beam market segments, and those technologies are really, really important as you're developing all of these new structures.
It really, it gets back to how fast you can, t he learning rate, the learning cycles in E-Beam technology is really, really important. You know, you have these Nanosheets that you're building in a Gate-All-Around structure. The width of that nanosheet is absolutely critical to device performance, and, you know, E-Beam is one way where you can see that at all of the different levels. You know, the billions of transistors you're building across the chip is very, very difficult. But that combination of technologies is really what makes Applied special. And so Gate-All-Around, about a $1 billion incremental opportunity for Applied, about a 15% increase in terms of the spending for 100,000 wafer starts for our customers. Backside power is another one that's really a great opportunity for Applied. So if you, if you think about, everyone has a smartphone.
The smartphone has a processor chip in it that has 60 mi of wiring. 60 mi of wiring and 15 billion transistors in that tiny little chip that's, you know, as thin as your fingernail, as thin as a sheet of paper. And so the backside power, we have tremendously high share in enabling all of that wiring, and as the wires become thinner and longer, then getting the data through there fast with low resistance is probably the biggest area of focus for all of our customers. And again, we have very way more than 50% share of all of those wiring steps. So now you have signal lines and power lines on the front of the wafer, you're taking the power lines to the back of the wafer.
And again, as that transition happens, we'll have more than 50% share of that transition to backside power delivery. The great thing about this, now, this is really along the lines of what I talked about in terms of materials innovation, enabling more of the roadmap going forward. With backside power delivery, you take those power lines out, you can shrink the chip up to 30% with no feature size change, and you have better power and better performance. And that, backside power delivery is the same zip code as gate-all-around, around a $1 billion incremental opportunity for Applied. So both of those areas are areas that are incredibly important to our customers for their roadmaps going forward for PPACt, but really, really strong positions for Applied Materials.
Got it. That's super helpful. Advanced Packaging is another inflection that you've talked about. It was pretty topical at Semiconductor. Your team discussed it in depth. What are you seeing there, and again, what sort of opportunities are for.
It's a great, great, great opportunity. So we had a panel discussion there with Intel, AMD, Qualcomm, and then some of the leaders in bonding technology. And really, what companies are doing, if you look at NVIDIA with their GPU, you know, they've moved to some advanced packaging technologies that give them tremendous improvement, something like three times faster speed in their systems and tremendous power reductions. So, and, and these chips are getting bigger and bigger. People are packing as much functionality, AI technologies, as much functionality as they can on those chips, and they and all of them are reaching the reticle limit. So to go beyond that reticle limit, they're looking at technologies like advanced packaging, and that becomes very, very important for power, performance, and, of course, cost is also important.
So for us, again, here's another area where we have over 50% share of our served markets. We have the broadest portfolio of products, PVD, CVD, the plating technologies, etch technologies, CMP for planarization. We're working with BESI on hybrid bonding, and there's other technologies that we have in the pipeline for packaging. But that's a billion-dollar business for us today, and we've said that we have an opportunity to more than double that over the next few years going forward. And we have a full flow packaging pilot line in Singapore, so we have our leading customers working with us as they're developing those new packaging architectures. So really, really big inflection, and the opportunity for Applied is really significant.
Got it. So earlier this year, you announced the EPIC Center, which is a multi-billion dollar R&D platform, I guess, in the state.
Yes.
45 minutes away, perhaps.
Silicon Valley, yeah.
Silicon Valley. So I guess for context of what catalyzed this investment, you've already had sort of the Maydan Technology Center. How do they differ in how you collaborate with customers and partners?
Yeah, I think that, again, there's always a race to deliver innovation in the industry. So we've, we're so deeply connected with the R&D leaders from all of our different customers. You know, one of the top R&D leaders, I've seen face to face ten times already this year, and our teams meet with these people every single week. But we, we really saw an opportunity to accelerate time to market, maybe as much as 30%. And if you think about how much money you spend, per R&D node, you know, that type of acceleration, certainly for value creation, is important, but the R&D savings, the R&D efficiency, is also, tremendous.
So really taking this process that's a kind of a serial process of evaluating innovations and then running evaluation wafers, back and forth, you know, from one site to the next, bringing those people that. We already have our R&D customers in the Maydan Technology Center in Silicon Valley, but this is having full-time teams there in Silicon Valley. And, you know, really innovating the way we innovate, just really tremendous pull. And we. As we were formulating the concept, we were working with some of our top customers on these concepts, and just phenomenal pull. But bringing those people into this new EPIC Center to innovate the way we innovate, having customer space, bringing more partners, so we can really even build stronger innovation networks.
We have the broadest set of technologies to enable these roadmaps, but bringing some of these other partners even closer to work faster and accelerate R&D, tremendous opportunity. And really, our largest customers are tremendously enthusiastic about the concept.
Does this concept allow you to benefit as sort of.
Well, you know, there, it gets, we have an opportunity to be designed in even more so with our systems and with our services. As we have all the innovation in our service technologies, having those customers there, where we're co-developing all of those future technologies, the beyond Gate-All-Around, their CFET technology or 3D DRAM, or any of those different technologies, I think what we do in this industry is like magic, y ou know, I, the, the wiring technology that gives you the 60 mi of wiring, there's one platform that we have that has seven technologies under vacuum to enable that low-resistance wiring. So we're constantly innovating at an incredible speed with these customers. So it's all the learning rate. Who's gonna learn the fastest is going to win. Those are the companies that are winning today, the companies that collaborate better and faster.
EPIC just takes it to a whole new level.
Got it. No, that's great. You mentioned ICAPS earlier. I think you led the formation of the group. Again, to level set the audience, what drove the formation of ICAPS? This was several years ago, I believe.
Yeah, more than four years ago. So ICAPS, IoT Communication, Auto, Power, Sensors. And we could see that that part of the market, i f you think about everything around us getting smarter, in our homes, entertainment, we work with system companies that have tiny, tiny eyes, tiny ears, healthcare, everything, you know, the 7,000 chips in your advanced electric vehicle. So we could see that that market was going to grow significantly, and there's a lot of innovation that takes place. If you look at sensor technologies or power electronics or analog technologies, there is a lot of innovation there. So more than four years ago, we saw that that market was going to grow a significant amount. We pulled all of the different aspects of the technologies across Applied together, 200 mm, 300 mm.
We have a device integration team. They only focus on those ICAPS technologies. So we could see, again, more than four years ago, that was going to be a significant market. And then going forward, people talk about a trillion connected edge devices, and content is growing. There's no question, all of those types of ICAPS chips, that market is growing. But again, bringing that together more than four years ago helped us create much deeper strategic relationships with all of those customers.
Did you foresee the business growing to this extent four years ago when you formed the group?
We did, actually.
You did? Okay. That, that's impressive.
That's why we formed that group.
Right. Right. Makes sense. And then I guess the one, one of the most common questions I get from investors is, how sustainable is this, right? And again, Gary, you're very close to CEOs and R&D leaders.
Yeah.
You know, five years out, 10 years out, I think most people agree that it's gonna be significantly larger.
Yeah.
Slightly shorter term, what are your expectations?
Well, I think that, for 2024, we see ICAPS stable. You know, certainly we've seen, a significant amount of growth, so in 2024, you know, we see that, that market stable. But going forward, you know, we see overall semis, you talked about the trillion-dollar market. You know, we think ICAPS is gonna grow along with semis at pretty much at that same rate, kind of, high single digits, mid to high single digits relative to the compound annual growth rate. And again, I think people underestimate content growth in so many different types of applications, for this ICAPS type of a, market.
Got it. Shifting gears a little bit, just want to talk about the geopolitical backdrop. I know, Gary, you spend a bunch of time in D.C. and other parts of the world. You know, what is the debate currently, and how has sort of the evolution of the geopolitical backdrop impacted how you think about your strategy, if at all?
Well, I think what we've communicated is that in with the export controls, that impact was between $1.5 billion and $2 billion, and we've communicated that that's toward the low end of that range. Certainly what has happened here recently is alignment with other countries. So Japan and Europe, we've seen alignment around those different regulations, which is important to have a level playing field going forward. Now, you know, with all of that, ICAPS, for us, in China, is still a very strong market. And if you think about all of those industries, electric vehicles and power electronics or sensor technologies or all of those, all of those areas, there are still tremendous opportunities there. And then for us, relative to the ICAPS market, we formed that group, as we talked about more than four years ago.
We've introduced, in the last four years, 20 major new products in that ICAPS market. We have very significant new products that will come to market over the next couple of years that will further strengthen our positions in ICAPS overall. So again, I'm still optimistic about that part of the market.
Got it. Gary, related to that, the other question that I often get from investors is: Could the U.S. government contemplate restricting, you know, a toolset that would be embedded in your ICAPS business? I personally don't have an edge in addressing that question, but how would you go about?
Well, you know, I'm not gonna speculate on, you know, what potentially could happen. But what I would say is that as you go to those ICAPS businesses, those technologies have been around for a fair period of time.
Yeah.
You know, really, the only way you would have anything that would be effective would be alignment with all of those other countries. Yeah, beyond that, I'm not gonna speculate.
Got it. Makes sense. And then on the competitive front, you've done a really good job competing with incumbents. We, again, get a lot of questions around competition in local China.
Yeah.
I know those companies have been around for decades, and they have yet to make it to the global scene, if you will.
Yeah.
Personally, I'm not too concerned, but what are your thoughts? Have you seen any improvement, evolution in their competitive footing?
Well, I think, you know, like, we're always mindful of competition, even in these non-critical types of applications. All these companies, normally, they'll start with more simple types of applications, like on etching, you know, etching bond pads that are, you know, very, very simple types of applications. But I think the only really way that you win in this market is to innovate faster than anyone else. And so for Applied, the real strength for us is the breadth of our technologies. About one-third of our business are these integrated solutions. I talked about, you know, transistor or wiring or even in power electronics or sensors or any of those different types of markets. We have real breadth in terms of understanding and working with our customers to bring those innovative technologies to market. So that focus has helped us.
We're innovating not just in these leading-edge technologies, but in semi-critical types of applications, and especially in ICAPS. As I mentioned, 20 products over the last four years, and a really strong pipeline of new products going forward.
Got it. You mentioned how you're really happy with the services business.
Yeah.
It's performed really well. It's grown year over year for, I think, 16 consecutive quarters or somewhere around there.
That's correct, yeah. 16 quarters in a row, yep.
You talked about the transition from, you know, transactional business to more recurring business. Where are we in that evolution? Which inning are we in, and how should we think about the growth profile of your services business going forward?
What we've communicated is that we can grow our services at a double-digit growth rate, and even this year, especially where memory is incredibly weak, so utilization in some of those factories are down, a fair amount. But even in that, we delivered record service growth last quarter, as you said, 16 quarters of year-over-year growth, within AGS. The opportunity there is tremendous. As complexity is increasing or you have those integrated platforms like I talked about, where you have seven technologies combined under vacuum, those are very complex systems. So for our customers, there's tremendous value in how fast they qualify those systems, so accelerating R&D. Another one, then once we talk about making a chamber golden, then you transfer into high-volume manufacturing, you want to make them all golden.
So for us, that is hundreds of thousands of pairs of chambers per year. So we have technologies there that we've developed that accelerate that time to ramp, and then also the process window optimization. So that's, I would say, maybe first inning. I mean, there's absolutely have proof points that we can accelerate the time, and we can accelerate better results through these technologies. And then in high-volume manufacturing, it's about yield, output, and cost, and we have a lot of innovations there also. So I think, you know, both on digital services that we're developing that, you know, are really, really extremely valuable for our customers, parts innovations, tremendous focus there also. So we have, I have very, very high confidence in double-digit growth in AGS going forward.
If you think about that business, it's running around $6 billion a year. Double-digit growth in AGS will add significant growth to Applied going forward.
Makes sense. Maybe I'll pause here and see if we have any questions from the audience. Maybe not? I'll keep going. Gross margins.
Yes.
So, given everything that you've just said, you're essentially developing sort of rocket science-like tools and services. Your gross margins are in the mid- to high-40% range. I always wonder, could you do 50%? Could you do 60%? And obviously, it's not that easy. I understand that.
That's a good question.
What is the conversation like with your customers as you collaborate more and closely with, you know, with your customers? Is there an opportunity to improve gross margins, not by 50 basis points, but by a couple of percentage points?
Yeah, I think, you know, we've talked about, we have implemented price increases with our customers, and we're certainly focused on also cost reduction. But the supply chain challenges that we encountered through COVID really delayed the margin improvement path that we were on. So again, we've talked about it before, we were not able to deliver to demand. So the focus for us was really maintaining customer trust as the number one priority. That meant we had to expedite parts, we had to pay more for logistics. There were lots of cost headwinds that we'd been encountering. I would say the good thing about what we've faced over the last couple of years is that we've put way more focus into supply chain, reengineering our supply chain, strengthening our supply chain.
So I have high confidence that going forward, as we double the size of the industry in Applied Materials, that we're going to perform at a much higher level than we did in this recent time period. So again, all of that reengineering of supply chain, I think will also pay a lot of dividends for us relative to cost going forward, and that optimization, and also in our aftermarket with our service business. And then as we're delivering these high-value solutions, I think we're in a really great position. As I said, if you look at the nodal spending, a higher percentage is going to move to materials innovation, and again, those are opportunities for us from a margin perspective.
Gary, you just mentioned you've sort of made changes to the supply chain. What were some of the learnings, if you will, from the pandemic? And maybe expand on that. What kind of changes have you introduced up and down the supply chain?
Well, you know, I think one of the things that we really never had to worry about before was this multi-tier visibility. So, you know, believe it or not, I was, I almost felt like I was our chief supply chain officer through that timeframe. We had many cases, and this really gets back to the ICAPS business, where we didn't have chips in the components, and we never had to worry about those chips. And in a lot of cases, what happened was, the chips would come from our customers to a distributor, and then the distributor would sell them to our suppliers, and the suppliers would ship components to us. So there were lots of cases where chips were committed, and then, you know, at the last day or last week, they're pushed out, in some cases, a year.
So this was incredibly difficult, incredibly disruptive. Of course, you know, for us, you know, I could call up our customer CEOs and say, "Look, I can't ship these tools unless you call this company and tell them to get us the chips." So, you know, I had many and many of those phone calls. But I would say that for me personally, I became very deeply connected into the supply chain. Even last week, I was visiting two of our suppliers, meeting with the CEOs and their teams. So I would say that, again, for us, we have hundreds of engineers now focused in supply chain. We will make Applied a much more efficient and effective company. You know, we, I think that there's so many lessons from this.
I think you never waste a crisis and, you know, certainly, there will be tremendous improvements in cost, inventories, and the most important is customer trust. Improving customer trust, tremendous, tremendous, opportunities there.
Got it.
I think agility, again, as we go forward, you know, again, I have high confidence we're going to double the size of Applied going forward. So, you know, we have to be better to be able to meet the demand going forward.
Got it. Gary, in the last 60 seconds or so, anything about the Applied story or anything about the markets that you play in, that we as investors and analysts collectively overlook or underappreciate? I know you spend not so much time with investors, but you still do, so any sort of, you know, disconnects, if you will, between how you think about your business versus how the market perceives it?
Yeah. I think that, you know, I've talked about strength of our service business, the double-digit growth there, and you can see the resilience that we have in a market where utilization, especially in memory, is fairly low in 2023. Our position in the big markets going forward is really tremendous, and more of the spending per node is going to these materials innovations. So Foundry Logic, we're in a great position there with Gate-All-Around, Backside Power Delivery, CFET technology in the future. In DRAM, we've gained 10 points of overall WFE share over the last 10 years. That share of the DRAM market for us is as strong as our leading Foundry Logic. I think that's maybe something people don't understand, and we are a fair amount larger than our process equipment peers in DRAM.
DRAM, they have moved logic processes into the periphery for high-speed DRAM. That's all of that logic leadership capability that we have. That has moved into DRAM. Capacitor scaling, patterning, there's so many areas there in DRAM where we have unique and tremendous strength. Packaging is another one I've talked about, that we're, you know, on a path to double packaging, much, much stronger than any of our peers. So if I look at the markets that are going to grow the fastest, you have about one-third leading Foundry Logic, real strength there. One-third ICAPS, very, very strong position for Applied in ICAPS going forward.
And then in memory, if you look at DRAM and NAND, NAND is an area where we're not as strong, but DRAM, if you think about, again, 8x DRAM content on an AI server and compute memory versus storage memory, we're much more bullish on compute memory growth. So again, all of those areas are tremendously strong for Applied Materials.
That's great. Really enjoyed the conversation. Thank you so much.
All right. Thank you, everybody.
Thank you.