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

Greetings, and thank you for attending JPMorgan's 12th Annual U.S. All Stars Conference. My name is Harlan Sur, Semiconductor and Semiconductor Capital Equipment Analyst here at the firm. And very pleased to have Dan Durn, Chief Financial Officer of Applied Materials here with us today. As most of you know, Applied Materials is the global leader in the semiconductor capital equipment solutions with diversification into the advanced display manufacturing market as well as a pretty strong position in the fast-growing advanced packaging market as well. And the digital transformation that we're seeing in virtually all end markets is driving the need for what Applied calls their PPACt playbook, right, or helping semiconductor companies drive lower power consumption, higher performance, smaller area, lower cost faster time to market in their next-generation chip solution. So power performance area of cost, time to market, PPACt. And PPACt applies not only to advance manufacturing and leading-edge deep submicron technologies but also to a more mature and specialty manufacturing processes as well and next-generation heterogeneous advanced packaging capabilities as well. And we'll get into a lot more of that. But I'll go ahead and kick it off with the first few questions, and we can open it up for investors. For those of you that want to ask a question, please type in your questions into the digital conference book using the Q&A button at the bottom of the digital cockpit. So Dan, thank you for joining us today.

Yes, it's great to be here, Harlan. I appreciate the opportunity to spend time with you.

Yes. And let me start off with the long-term question, and I believe this does encompass the acceleration of the digital transformation we are seeing in all end markets and again, driving the need for your PPACt playbook. If I look at wafer equipment intensity, wafer equipment spending over industry revenues, that bottomed in 2013 at 9%. And starting in 2014, the first year of 3D NAND production, WFE as a percent of revenues has monotonically increased and will hit about 15% this year. Another way to look at it is that over the past 8 years, wafer equipment spending has grown at a 14% annualized rate versus the semiconductor industry that has grown at about a 7% annualized rate. Applied Materials has grown its semi revenues, equipment revenues at a 16% annualized rate over that period of time. So great outperformance by the team. So given the manufacturing complexity challenges, the PPACt challenges, does the team anticipate WFE continuing to outgrow industry revenues? And what are going to be some of the major drivers?

Yes. So thanks, Harlan. Our view is WFE will continue to outgrow overall semiconductor industry. Now if we take a step back and think about what's driving semiconductor demand because I think it's instructive as how it shapes our industry. So as I look at semiconductor industry growing to $1 trillion over the next decade, the predominant drivers of that are going to be the digital transformation of the global economy. Virtually every industry sector is going to transform digitally. And it's going to be spend that's driven by the largest companies exposed to these large industries. And increasingly, these investments are going to change the basis of how these companies compete, how they serve their customers, how they drive their road maps, how they drive efficiency into their supply chain. So the incremental return on these investments, I think, is going to be greater than what we've seen historically in the industry. And as we think about the building blocks of the semiconductor industry and those fundamental capabilities, I think we're transitioning from what was predominantly a 2D shrink-driven world historically to something that's far more diverse and more complex. When the industry for quite some time, every 2, 3 years, would shrink feature sizes, they get the simultaneous benefit of power performance area cost. We put a lot of capacity behind it in every 2, 3 years, the same thing would happen. And what we've seen increasingly here over the last several years is away from that. You no longer get simultaneous benefit of power performance, area, cost, when you shrink in 2 into dimensions. And shrinking into dimensions is increasingly difficult. So the industry is going to a new playbook. More diverse drivers of the power performance road map, new materials, new structures, new types of architectures, new types of packaging, different ways to compress the area of a chip. And that's leading to higher complexity in the industry. That higher complexity really serves our business well. And when you no longer have wafer size transitions -- we're on 300-millimeter wafers today. We're not going to 450-millimeter wafers as an industry. All of these factors are coming together shape and industry where we're seeing an upward bias in capital intensity over time. And so we think it's really, really well set up and in our favor. Structurally larger industry, no wafer size transitions, consolidated customer base is no longer flooding the market with used equipment and an ability to drive that power performance road map differently going forward. I do think we'll see an upward bias and we'll benefit nicely as those trends play out over time.

Let's focus maybe on more of the sort of near to midterm. During our July earnings call, the team raised their expectation for 2021 wafer equipment spending to greater than $80 billion from the high $70 billion on the last -- the prior earnings call, that's about 30% growth this year. In what segments of the market are you seeing the incremental strength versus your expectations at the beginning of the year? And on your 2022 outlook for continued WFE growth, although you didn't quantify it, did the confidence level on that expand relative to your view back in May?

Yes. So the way I would talk about this year, I would say the shape of WFE throughout the year, we've seen very little change over time. We talked about foundry logic being a strong growth driver this year. We talked about DRAM doing better than the average market. And we talked about NAND while growing significantly undergoes the overall market, that construct still holds. And then within foundry, logic and DRAM, again, the shape throughout the year, we talked about both of those subsegments being second half weighted to second half being greater than first half. NAND profiles differently. NAND, we said it's flat to down slightly second half over first half. Still grows, but the shape throughout the year is very different than foundry logic and DRAM. Very little in that setup has changed. And while we've seen an overall strengthening in the size of the market, I would say that, that's distributed across all 3 device types. If I were to rank order them from 6, 9 months ago, I think we've seen the most strength in foundry logic than DRAM than NAND, in terms of incremental increases in overall spend. And as we window into next year, I think if we were to go back 6 months in time, we would have said, based on underlying demand conversations with customers, the way we triangulate factors in the market, we saw WFE in 2022 being up over 2021. And given where we sit today, as the world gets back to working, we see certain pockets of the supply chain showing nonlinear profiles in terms of availability of supply. I think what's increasingly important to us as we see more of the year unfolding, that there's underlying end market demand from our customers that's unmet this year and slides into next year. So while we had a point of view for quite some time that 2022 was an upward sloping trend line versus where we are this year, based on this unmet demand that will likely be satisfied next year, I would say our confidence interval around an upward trend into next year has gone up over time in this environment.

And I believe that the investment community is also getting more confidence on the positive demand trends in the broader semiconductor industry continuing to next year, especially given, as you mentioned, industry-wide chip shortages. We've got strong demand from cloud and hyperscale, continued strong growth in 5G smartphones and infrastructure buildups to name a few. But there's been a lot of concern around memory fundamentals, like both DRAM and NAND into the back half of this year and into the first half of next year. Given your team's scale and market leadership, you have great visibility on memory WFE spend as well as on obviously, the supply side dynamics. So how does the Applied team see the supply-demand dynamics in memory exiting this year and moving into next year?

Yes. I think probably the best part -- best place to start this conversation is talk about what happened in 2020, use that as context for what we're seeing in 2021, and then how that shapes a perspective around 2022. So if I were to go back to 2020, what we saw is an industry that was growing overall WFE industry growing sort of mid- to high teens. Against that backdrop, we saw a NAND industry that probably outgrew the overall market 2:1 in 2020. We saw a DRAM market that was roughly in line with the overall market. What that shaped from a supply-demand standpoint is exiting the year, I think, you had a view that NAND bit supply was more in line with end market demand. So a very balanced supply-demand market entering 2021. DRAM was different. There was an undersupply of true end market demand from a DRAM standpoint exiting 2020 and at the start of 2021. In 2021, what we see as a market that's up mid-30s, plus or minus, we see foundry/logic being the strong grower. We see DRAM growing slightly above the market, and NAND growing, but undergrowing the market. So the rank order of end markets has actually reversed this year versus last year. We see DRAM outgrowing NAND. And if I were to take a snapshot of the supply-demand balance in each of those respective markets, in NAND, I think the supply-demand balance continues to be in line. I think it's a healthy market. I think supply is in line with demand. And I think that sets us up for another year. This is a healthy year spend in NAND. I think it's going to be another strong year of spend in NAND next year. In DRAM, I still think supply needs to catch up with demand. I don't think we're there. And I think that's why you get a second half-weighted end market from an equipment standpoint, second half-weighted market versus first half. I still think more bit supply needs to come online to get those markets in balance. Now if I were to take a snapshot of the spot market in DRAM, I think there's some near-term softness, which is leading to a competing narrative in the market. Is that softness fundamentally driven by weaker true end market demand? Or is it supply chain instability of the things that sit around the DRAM component that are preventing the end customer from producing a full system and shipping? The fact that our customers have probably strengthened their outlook from a demand standpoint and equipment perspective over the last 6 months, I think, give you a sense of how our customers are viewing that competing narrative. We don't see any waning strength. Our customers are pulling very strong for the technologies to get the capacity in place to bring DRAM bit supply more in line with true end market demand. And as we window into 2022 in the DRAM market, we would expect another strong year of investment but a market that's more in balance than, say, we saw exiting 2020. And for most of 2021, DRAM has been playing catch up, and we feel really good about the second half profile from a spend perspective as a result.

No, I appreciate that. Going back to the industry-wide supply shortage, chip shortages. How has COVID-19 in the industry-wide chip type has impacted the team's manufacturing operations, your lead times? And more importantly, how is the team increasing its systems manufacturing output to support the continued growth in the business that you expect for next year?

Yes. So if I were to go back to the start of the pandemic and just to refocus on -- I believe it was March of 2020. On the exact same day, the Bay Area in California and Malaysia went into simultaneous lockdown as a result of the start of the pandemic. And it created a significant disruption from a supply chain standpoint for the broad industry. I mean, still remember the conversation Gary and I had in his office, and we took a bit of a counterintuitive approach at that time. We said the pandemic is going to lead to supply chain disruptions. And so rather than pulling back on our hiring plans or cutting costs, we actually made a decision to lean into the process. We used our balance sheet to sweep inventory. We increased the headcount levels inside of the company because nonlinearity of supply means that the things we do, we're going to be dealing with different peaks and valleys intra-quarter, and we need more capability, not less as a company. And I think that positive bias to lean into a problem and make investments on behalf of our customers and do everything we can to put order into the supply chain has really served this company well. It was a more aggressive posture than I think the average company took, not only in this industry but in general. And I think it serves us well. So I think we've done a good job in this environment not only outperforming from a top line standpoint, but you also see the execution ability from a margin standpoint in this environment, both from an absolute standpoint and a relative standpoint, the company is performing well. So we'll continue to manage this the way we have up until now. Unfortunately, as the world tries to put the pandemic behind it and get back to global economic commerce, economic flows, there's still going to be some friction spots in the global supply chain. We'll just have to work aggressively to manage those the best of our ability the same way we have over the last 18 months, continue to do a good job, stay focused on serving our customers, make incremental investments where we need to. We've taken a point of view now for several years that our industry is going to grow structurally larger. Not only do we have that point of view, but we have conviction around that point of view. And you see the investments that we've made from a CapEx standpoint to walk our capacity in physical infrastructure up in line with the industry so that it's not the rate limiter on our ability to serve customers. And I would say we have a multiyear road map of incremental investments that we need to make as a company to continue to do that. So we'll stay biased towards growth. We do have a point of view. We have conviction around it. We're going to stay biased towards growth and be disciplined about the investments we make, make sure we make those investments. So it's not the rate limiter of our ability to serve customers.

Perfect. I appreciate that. We've got the incoming question from one of our investor clients. And the question is, how quickly do you see DRAM shifting to 3D structures? And how is the team positioned for the shift?

Yes. So from a timing standpoint -- let me handle it this way because I don't want to get out in front of our customers on announcements of road map and timing. What I would say is there's 2 structure inflections for the industry. One is in foundry/logic going from a FinFET transistor to a gate-all-around. And if you think about our Analyst Day invest forecast horizon out to 2024, you can see gate-all-around beginning to impact spending around that sort of outside window of our midrange model. But I won't be more specific and let our customers talk about that inflection. I see the 3D DRAM being on the outside of that inflection, transistor inflection in foundry/logic. In terms of how well we're positioned to these new inflections, I would say that both what we see from a gate-all-around transistor perspective as well as 3D DRAM, that it's going to play to the traditional sources and strengths of us as a company, our leadership product portfolio. Some places where we have 60%, 70%, 80% market share, it's going to be at the sweet spot of where we compete, the sweet spot of key industry-leading, key enabling technology for our customer and increasingly combining these capabilities. The other thing I would say as these inflections play out, particularly in foundry/logic, you're going to see new schemes with respect to wiring, both from a wiring resistance standpoint, but also how you power the transistors and where those wires that come into the transistor to power the transistor, you're going to come at it from maybe under the transistor instead of on top and will allow you to shrink the area. You're not changing the feature sizes but you're going to be shrinking the area of the chip because you're optimizing vertically how you arrange those building blocks of the chip on the wafer. And so we feel really good about the inflections on the horizon how they play to our strengths and what that's going to mean from an absolute and relative momentum standpoint as those inflections play out over time.

And then you gave us a good articulation around Applied's positioning on some of these next-generation sort of leading-edge technologies. But -- at the same time, we also continue to see very strong demand pool for analog, for microcontrollers, sensors, power semiconductors that utilize specialty and mature semiconductor manufacturing processes. And in fact, we estimate that about 50% to 70% of the chips in all electronic applications are made up of these types of chip solutions. And on the earnings call, you said that your mature and specialty business, what you guys call your ICAP business, was about half of your foundry and logic business. If we assume it's been and will continue to be about half of your foundry and logic business, that's going to be a $4.5 billion business for your team this year. The team just held a master class in this as well. But help us and help investors understand how we should think about the drivers of the ICAP business into next year and over the next few years?

Sure. So I think the first place to start is definitionally. When we talk about leading edge, and we talk about ICAP, different companies will describe it differently. For us, we view leading edge as the 3 most current nodes in the foundry/logic space. So basically 3 nanometers, 5 nanometers and 7 nanometers. ICAP is everything else. So everything 10 nanometers and above for us is ICAPS. And I think if we were to take a step back and think about the evolution of the foundry/logic space over the last decade, I think what you saw a decade ago was 90% of the spend on the leading edge and a modest 10% of the spend on these more mature specialty nodes. And then over time, what you saw was an increasing diversification. 90-10 became 80-20, 70-30, 60-40. Last year was balanced. This year, we see balance in that market. I would expect balance again to be on the table for next year. And as we think about the profile into 2024, what we said at our Analyst Day is not only do we see balance, but we see that balance continuing going forward. So I'm encouraged by the diversification of the foundry/logic space. The spend is still strong at the leading edge. The innovation is still strong at the leading edge. But now we see a diversification of spend on the ICAP nodes. And in the ICAP's acronym is IoT, communication, auto, power, sensors, it's a market-focused acronym. And that market-focused acronym says that innovation is alive and well in those industry verticals. And so as the world transitions from a mobile compute-centric model to increasingly what is edge cloud model and you push intelligence to the edge. As you push that intelligence to the edge, they need specialized solutions across a broad node profile and innovation, that is alive and well, and it's going to serve our business well. We've observed the trend playing out over a number of years. Several years ago, we reorganized our company around this leading-edge and ICAPS model. And we've got a lot of great engineers now focused on innovating across that entire node profile. It's leading to strong growth for us. We're well positioned, and we're deepening our relationships, our broad set of relationships across the customer base around this renewed focus of innovation across a much larger distribution of the nodes than I think we've traditionally seen in foundry/logic. I think that sets this business up well over the long run. I would expect both of those businesses to grow in line with each other over the next handful of years.

And then as a complementary part of the business, also you talked about in the last earnings call as well as on the recent masterclass, you discussed advanced packaging revenues are on track to exceed $800 million this year, nearly doubling since 2019. You own 60% share of your served market. You've got your big end-to-end customers like AMD that's driving very advanced 2.5D, 3D chiplet strategy. Intel is going after the tile strategy with their EMIB and Foveros technologies. All of this because Moore's Law continues to slow, and packaging is all of a sudden becoming one of the key drivers to help enable this PPACt. So how should we think about the packaging market growth over the next few years? What does apply have a leadership position? And what gives you confidence supply will continue to grow faster than the market here?

Yes. So I think you're exactly right, Harlan. And you're picking up on something that I think is going to be increasingly important for the industry and the power performance road map, and you're seeing multiple customers drive down this vector. Packaging, I think, today is an underappreciated aspect and contributor to the future power performance road map. Gary and the team, to their credit, saw this a long time ago. Back when Gary first came into the company, we put a renewed emphasis and focus on 3D integration, advanced packaging. And we made -- consistently made lots of investments around what we saw on the horizon was a really significant inflection that was going to take hold and then accelerate. And now we're in the early stages of that playing out. So to Gary and the team's credit, they saw this and we've got significant investments in place. We've got a full-flow packaging lab set up in Singapore. We've got a number of key-enabling technologies that we've brought together. We've co-optimized a number of these adjacent steps. And if you take one simple example through silicon via, it's difficult technology, but 5 key enabling steps. We've co-optimized those steps. And so it's basically a drop-in module today for our customers to do something that historically has been very difficult. So we think we're well set up. We're also partnering with other companies and industry leaders. Hybrid bonding is something that a couple of the customers are increasingly talking about. So we've partnered with an industry leader, where we're bringing 6, 7 technologies together under vacuum. Partner is bringing one of the technologies. We're combining it with 5, 6 of our own technologies together onto a tool to solve a problem for a customer in a highly integrated drop-in fashion. And so not only have we assembled a set of capabilities inside of the company, but we're also partnering with industry leaders to enable solving our customers' highest-value problems. What it's led to is a business that's growing over 60% this year, like you said, $800 million of revenue. And as we go forward, I see this being a multibillion dollar opportunity for the company, we won't be more specific than that for competitive reasons. And where we compete based on early insights to this inflection and consistent investment over time, not only in our technology but also full flow capabilities. You're seeing us win over 60% of what we compete for. So we think we're really well set up. Not only to capitalize as this inflection unfolds but help our customers drive this inflection into the industry. So we're excited about what we see, and we're excited about how well we're positioned. We'll continue to stay focused, continue to execute, focus on solving our customers high-value problems, but I would expect us to do well as this inflection unfolds in the industry.

No, the Applied team has always been known for strong leadership position in deposition in etch in things like CMP, epi, implant, thermal processes. And one of the areas where we've always felt that the team has just a great SAM expansion opportunity is in the area of process control. This is a $7.5 billion market. And to you credit, I mean, Applied actually has a very good #2 market share position, but there's still, from my perspective, a big market opportunity ahead of you. And I think that your prior view before the last earnings call was that you were going to grow your process control business 50% this year. Now you're saying that it's going to be up 60% this calendar year. There's clearly a strong focus on new process debug, production yield improvement. As time to market for these new processes are so critical, what areas is the team going to see the strongest growth or potential share gains this year in process control? And how does the growth prospects look for process control -- look for the apply team over the next few years?

Yes. So we do think this is a really great market, and you've touched on something. We see a lot of momentum in the business. And -- but when we think about our process control business, we think about it in 2 pieces. You've got your e-beam capabilities and you've got your optical wafer inspection business. And if we take a look at the growth rate of this business for us, greater than 60% this year, I would say our e-beam capabilities outgrow that overall segment growth rate. Optical still grows nicely, but undergrows the overall segment, combined greater than 60% growth. And I think there's an underappreciation today on the importance of e-beam technology and e-beam metrology in the industry. Feature sizes are getting so small. We've got sub 1-nanometer resolution of our e-beam technology. So you can see with specificity what's happening on the wafer. And the ability to get at those very fine feature sizes with industry-leading resolution allows our customers to dial-in their road maps faster than they otherwise would. We've got a great optical business. It's performing well in the market. And so the combination of these 2 capabilities are coming together in a differentiated way to solve problems for customers. And again, this year, greater than 60%, it's a business that we feel really good about and very focused on.

What's interesting is that Applied Materials, you help end-to-end customers develop some of the most advanced computational and computational acceleration chips in the world, right? You guys are at the heart of Invidia's leadership and AI accelerators, the Google TPU, AMD's, GPU platforms and their compute platforms. And so you're helping to enable all of the advances in these pretty complex compute workloads, many of which are AI, deep learning and analytics-driven. And at the same time, the Applied team is taking advantage of all of these AI capabilities in your own process equipment. And so can you give us some examples of that? I know that it's pretty prevalent and part of the success in process control, but I also know that you use it across many of your other product lines as well.

Yes. So I think you're exactly right. Not only we're a key enabler of these broad trends shaping how companies compete and people live, but we also consume these technologies with respect to our own business. There's an internal perspective, and then there's how we package these capabilities for the external world to add value for our customers. From an internal standpoint, we're strong adopters of these techniques to help drive our R&D in our road map. So I won't go into more specificity for competitive reasons, but we do adopt them from an internal standpoint. What we've been more open about is sharing these capabilities from an external standpoint. And when you think about capabilities, things like process recipe optimizer. If you were to go back a decade, let's say, illustratively a piece of equipment would have 5 knobs. A human being can get their arms around the interdependencies of those 5 knobs, how it affects on-wafer performance and be able to manage a process recipe through a series of experiments to try to dial it in. Today, when you have, say, illustratively 20 knobs on a system and millions of combinations and set points on that equipment, it's beyond the capabilities of the human being to simultaneously process the interdependencies from those variables. So what we've done is we've adopted algorithms and capability, sensors, metrology, helps our customers dial in recipes 50% faster and actually increased process windows by 30%. It's an enormously powerful capability to speed time to. Getting at the PPACt road map and focusing on the t element of the equation is highly enabling and highly valuable for our customers. Packaging capabilities like that, that, again, can bring those process recipes to market to 50% faster, increase those process windows to increase yield and make the process in high-volume manufacturing more robust, extremely capable and powerful. The other example I'd point to that we spoke about is our extract AI capability. When you image a wafer with an optical system. Based on the resolution of optical systems, you have uncorrelated set of potential defects on the wafer. What you do is you process that wafer, you pass the wafer to an e-beam system, e-beam review system, along with the data. And now in a massive way, you go into a fine level detail where you can start to correlate an optical signal with actually what's happening on the wafer based on that very fine resolution. And you train the optical system of which defects are meaningful and which ones aren't. You start to correlate those signals in a way that allows our customers to take action with how they process wafers in the factory to get better outcomes and do it very quickly. So there's an embedded learning capability in that system where you start to train the system in line and then it gives our customers a road map on how to change process variables to get better outcomes out of the wafer, enormously powerful set of capabilities, and it's being used real time today with our customers. And there's more investments for us to be making over time. But I think those are a couple of real solid examples of how we're taking these capabilities and focusing on solving some of the highest value problems our customers have.

Yes, it's pretty incredible. I mean, you're manufacturing systems and your process control systems are embedding like mini supercomputers into these platforms. I don't think a lot of investors understand how much compute capability in analytics power is integrated into our platforms, but those are some great examples, and I appreciate that. Let's focus on the services business. AGS, Applied Global Services, that was up 24% year-over-year in the July quarter, first time above 20% in 12 quarters. Was it primarily utilizations and that is driving the strong growth as your customers are driving strong manufacturing activity? Or is it more value-added services offerings or a combination of all of the above? And with near record 31% operating margins, I mean, you're already at your target model that you had expected to reach in fiscal '24. So is the performance here just operating leverage? Or are there other structural drivers of the op margin expansion as well?

Yes. So with respect to the target margins, I think we're well set up and making nice progress. If we were here in 2024, I would expect to exceed those targets that we set. So we feel good about the execution against those expectations and how well we're set up. I would say, from a growth standpoint, everything you mentioned, I would agree with, but here's how I would frame the growth engine around our services business. I think you first start with installed base. These are systems that are out in the field, some of them 3, 4 decades. We've got the industry's largest installed base, 40,000 systems, over 160,000 chambers. And like I said, the longevity of these systems in the field is very, very long. That creates an enormous lifetime annuity stream for the company to penetrate against. And right now, we're only about 20% penetrated of that overall available market that services our tools. So there's lots of headroom off of where we sit today to continue that to grow against that installed base. And the installed base is going to grow every single year. Good year in WFE, it grows faster. Less good year in WFE, it grows slower, but it still grows each and every year. So it's a nice structural tailwind around the services business. The other thing I'd come to -- the second element of growth that would come to is complexity. What we ship today is far more complex than what we did 5 or 10 years ago. And the service entitlement on the system is much greater than what it used to be 5, 10 years ago. So that embedded complexity of the technology that's being delivered to market is a nice tailwind for the business. And as I look at the road maps going forward and the increasing complexity around how our customers are going to solve those problems, I think that embedded escalator of lifetime service value for those complex tools serves this business well. The third element I'd point to would be long-term service agreements. So when you think about it, our customers can choose to spend some money with us. They can have their own capabilities, they can spend with third parties. When we sign customers up to a subscription agreement over multiple years, they spend the same amount of money they just spend more of it with us. We get a higher share of wallet and the customer gets a better outcome. And so as we've taken this business from 40% of our revenues to 50% of our revenues. Today, it's 60% of our revenues, growing that to 70% of our revenues is an important element of the growth strategy. At our Analyst Day, we talked about increasing the revenue per tool of the entire installed base by 20% from 2020 to 2024. And I would observe that we're over halfway there given where we sit today. So again, making good progress against this opportunity. The last element I'd put on the table, as governments play a role in distributing capacity geographically around the globe. I think you're going to get a larger number of smaller scale facilities and capacity that gets built outside of home geographies and home ecosystems, have a higher service entitlement attached to it. So as that trend plays out and governments distribute capacity around the globe, I think that's going to be a nice incremental add or an upward bias to the service business as it plays out over time. So we feel really good about the strategy, the execution and the opportunity.

Yes. And as it relates to the execution and the outperformance relative to the fiscal '24 model on the services side, I think the team has also said that just from an overall corporate level that you're set up well to exceed your fiscal '24 long-term model. In fact, we were pretty much there as it relates to the last reported quarter, you are 50 basis points away from your gross margin target. You're already exceeding the midpoint of your operating margin target range for fiscal '24. You talked about the sustainability and expansion of gross margins going forward on the call. But when we think about operating margins, there's another driver, which is your scale, right? So how should we think about your OpEx growth relative to revenue growth to sort of get an idea on operating margin expansion on continued growth in the business for the team?

Sure. So as we start the conversation about the expectations and where they were set, we talk about a base case model. We talk about an upside. At our Analyst Day, we talked about if I were to put a probability distribution between base case and upside, I think you skew towards the upside. So I do feel good about our ability to hit our commitments embedded in the base case, and I'd like to push through each of those commitments and push it towards the upside model. So I think we've got good execution and good momentum around that. The other thing I would say is company is performing well in this environment. And there are headwinds in this environment, as you point out, the pandemic, incremental costs. And I would say absent those near-term environmental headwinds from an expense standpoint, I think we'd be on top of where we set expectations from a gross margin standpoint. So I feel good about the execution there. We'll never be satisfied with where we sit. We're always going to want to push more and drive value for customers, but I think we're executing well in a tough environment. Then if we think about OpEx and we think about leverage built into the system, what I would say in our long-term model, we put an expectation of about 16.1% OpEx to sales. And if I were to go back 7, 8 years, I think the company was at 28% OpEx to sales. And we really leaned out the company, and we focused the vast majority of that spend, 70%-ish of that OpEx, focused on growing the company through R&D, fueling the innovation engine of the company. So longer term, I would expect to settle out closer to where that target number is 16.1%. We've grown very quickly in the near-term environment. We want to be disciplined about the rate and pace at which we add expense. And so we'll do it when it makes sense, but I would expect us to grow into that expense ratio over time.

And we're running out of time here, but my last question is, and we get this question many times from investors too, is as we see the greater push for localization of fabs and manufacturing capabilities, geographical diversification. There continues to be a push by governments for subsidy like programs. Like, for example, the U.S. Chips Act, right, which is yet to be approved by the house, but I think the expectation is that sometime in the second half of this year. How do you see this as a benefit? Or is this a net benefit for Applied Materials? Does this expand the wafer equipment, SAM opportunity? And does it also potentially expand your growth trajectory on a go-forward basis?

Sure. So I think it's important to break it down. First is the why, then the timing and then what is the impact. The why, I think semiconductors are at the intersection of economic growth in national security and governments around the globe are beginning to wake up to that dynamic. So we feel good about where we are on that strategic landscape in terms of importance as an industry. I think from a timing standpoint, we hear a lot of initiatives announced. I don't think they start to influence and impact the industry until 2023 and beyond and will run for multiple years. And if it happens sooner than that, then there will be upside to where we set expectations. But right now, it looks like 2023 and beyond and run for multiple years. I think it affects the industry in 2 ways or affects our business in 2 ways. You're going to end up with more larger number of smaller-scale facilities distributed geographically around the globe. We know that is more capital inefficient than how factories are built today. So I would expect there to be an upward bias in WFE intensity, capital intensity over time, and that will benefit us as a supplier. Governments will come in and normalize that economic equation with our customers between how they deploy capacity today and how it will be deployed in the future model. Our customers will be made indifferent, WFE capital intensity goes up, and I think that benefits us directly. The second way is what I talked about with our services business. As those factories get built around the globe and away from home ecosystems, where there's scale component, critical mass of engineers, the service entitlement on that capacity is higher than if it was built where the customer currently locates capacity. And so that's going to be a nice upward bias from a services business as well. So I like how well we're set up against these pending trends of economic growth and national security and how that benefits our industry.

Great insights, Dan. Thank you very much for your participation today. The Applied Materials team is clearly set up for a strong year, both this year and for next year, and we look forward to covering the team and the progress of the team as that growth strategy unfolds. And I also want to thank all of our clients for tuning in today, and hope everyone has a great week. Thank you very much.

Thanks, Harlan. Always good spending time. Take care.