Applied Materials, Inc. (NASDAQ:AMAT) BofA Securities Global Technology Conference Call June 3, 2020 10:00 AM ET
Dan Durn - Senior Vice President and Chief Financial Officer
Conference Call Participants
Vivek Arya - Bank of America Merrill Lynch
Hello, everyone. This is Vivek Arya from Bank of America Securities Semiconductor team. Really honored and delighted to have Dan Durn, Senior Vice President and the Chief Financial Officer of Applied Materials join us this morning.
What I was hoping to do was really dive into some Q&A. And if along the way, if you have any questions, please use that Veracast window that you have in front of you to e-mail me any questions. But with that, welcome, Dan. Really glad to have you at our conference.
I appreciate. It’s great to be with you this morning.
Right. So, Dan, let’s just dive into it. If you could just give us a quick State of the Union. You guys reported very strong results recently. But just the whole, I think, supply and demand dynamic in the market has been very turbulent because of all the COVID headwinds.
So just give us a snapshot of how you’re thinking about the demand environment and the supply environment for you in the near-term? And then we will go through some of the longer-term trends and growth drivers for Applied Materials.
Sure. So just from a demand standpoint, demand continues to be strong. Company’s results in our fiscal Q1 and fiscal Q2 on a year-over-year basis in our semi business was really, really strong.
I think our semi-related businesses were up 18% in fiscal Q1, 13% in fiscal Q2. And so for the first-half of our fiscal year, we’re up 16% in our semi-related business. We exited Q2 with record backlog in those businesses. And in the quarter, fiscal Q2, we saw record orders.
So clearly, the demand environment continues to be strong. We have been calling for continued strength in foundry, logic throughout the year, broadening of the order book, multiple customers, multiple nodes, and that’s playing out as expected. We said that the swing factor in memory is going to be really what happens in the back-half of the year.
And what’s going to determine ultimately how the memory market shapes up is going to be a function of whether or not you get capacity ads in the back part of the year and what happens to the wafer starts per month as our customers pursue their roadmaps.
And if you look at our results in fiscal Q2, you’ll see that our memory business was up 25% sequentially. And so off of the Q1 levels, we’ll see strength versus Q1 we saw in Q2. We won’t grow sequentially every quarter like that, but we’ll see some follow-through on the strength we saw in fiscal Q2, at least, in our business.
From a supply chain standpoint, on March 16, concurrent shelter in place orders went out for the Bay Area and Malaysia, and a significant part of the supply chain, either directly or indirectly runs through those geographies. I don’t think we’re unique in that regard. But certainly, that’s how our business is impacted.
And we talked on our earnings call about the recovery of the supply chain and returning to health and strength. And we laid out a framework around this, that hopefully investors view as helpful to give a snapshot at least of what we’re seeing in the current environment.
So if we go back to the mid-March, supply chain was cranking pretty hard to support a very steep ramp. And overnight, when those shelter in place orders went out, you went from almost red line to no activity. And then we work with suppliers to get the semiconductor industry, as well as the ecosystem, broader ecosystem designated this critical infrastructure and get factories back opened.
Once our suppliers opened their factories, they’re going to get back to pre-COVID levels of staffing and output. And that pretty much, the vast majority of that happens by the end of our fiscal Q3. So the current quarter we’re in, we expect to have the vast majority of that in the rearview mirror, given everything we see in terms of the response of our suppliers.
Then, you’ll take the next two quarters, our fiscal Q4 and our fiscal Q1, which is the back-half of the calendar year to make up for the lost volumes that were incurred as part of this shutdown and unmet end market demand.
The one part of the health of the supply chain that will be a longer-term issue is the logistics channels. Clearly, capacity has come out of the system with the commercial airline industry. It’s a channel that was leveraged very heavily by the industry. A lot of things short of full systems, ride in the belly of commercial aircraft. And so we’ve had to pivot to alternative supply and logistics channels to get things from point A to point B. I don’t see recovery in the commercial airline industry anytime soon.
So that aspect will be with us. And the implication of that being with us for the foreseeable future on the logistics channels is, we can still get things from point A to point B. But it does take longer. What used to take three days now takes five. What used to take five, six days now takes eight, nine days. But we can still get it from point A to point B, but it costs more. In some cases, significantly more than it used to.
So that’ll be an economic headwind that’ll with us for the foreseeable future until we see recovery of the commercial airline industry.
Got it. That’s a good overview. And Dan, as you know, China and the U.S. trade tensions are on top of mind for a lot of investors. And last year, we saw the broader semiconductor industry being affected by it. This year, it seems like the semi-cap equipment industry is perhaps more right in that line of fire, if you will. So give us a sense for how investors should think about the impact of these trade restrictions?
On the one side, we have the new Department of Commerce restrictions and what they want to do in terms of the military end-use aspect. On the other side, we have additional restrictions on Huawei, them potentially canceling orders with TSMC and so forth. So give us a holistic view on what these new restrictions mean for Applied in the near to medium-term?
Sure. Let me start with what Gary said on the earnings call a week or two ago. And there’s no changes from what we said on our earnings call. We spent a couple of billion dollars a year on R&D. We’re a highly innovative company. And IP protection, free fair trade are super important to us.
And so let’s start there. Then, let’s talk about Department of Commerce. There is rules that have been enacted military end-use, civil-military fusion. And then there are rules that are contemplated being enacted and this is really Huawei, and I’ll deal with them each in order.
As we think about military end-use, civil-military fusion, we’ve spent a lot of time with government officials since those rules came out. We’ve got very senior level advisors, people who used to be at very senior levels of government prior administrations. We’ve got a really strong team inside the company.
We’ve got a point of view based on all of those discussions that there is a path forward to be in compliance by the time these rules are enacted at the end of June. We feel good about that perspective. And I would say, we feel stronger about that perspective over the last couple of weeks since our earnings call based on continued engagement on these different levels.
So we feel pretty good about that. We’ve also said on the earnings call, we’ve got a flexible operational footprint that gives us some degrees of freedom on the back-end as well. But the intent of the company is to understand and be in compliance by the time the rules are enacted. We feel pretty good about that path.
Now rules that are being contemplated, and this is specifically as it relates to Huawei, we’re still in the comment period, so it’s early in the process. Final rules are probably several months out. And so responding to what’s been written to date, again, engaging with government officials, advisors, our internal team, but also these rules are requirements for customers, not on us as a company.
So we’re also engaging with the broader ecosystem and leveraging their perspective. Based on everything that we understand and what’s been written today, we feel that there’s no need to change our expectations around the way we service our customers and the expectations we have for end markets, and what we see developing in those end markets.
We’ll continue to stay close to it, continue to monitor it, continuing to engage with government officials, advisors and the ecosystem. But given everything we see today, we don’t feel that there should need to change expectations around these requirements.
I see. Just a clarification on that, Dan. Historically, have you needed licenses to ship products to China? Because the one scenario is that, yes, the industry and I’m not being Applied specific. But let’s say, your industry is able to somehow isolate what was perhaps being done for military usage and that has stopped.
But what if that foundry in other parts of their business is supporting the military in some way, shape or form? Do you see any scenario in which customers in China are substantially restricted? I imagine that is not part of the baseline plan right now?
Yes. So there’s a couple of things. First of all, regulations aren’t new. This industry has been involved in regulations and licensing aspects, depending on things from a historical standpoint, and then there has been a relaxation of that environment for a number of years. And now, you see it potentially going in a different direction.
So this is not a new thing for us to be working our way through understand both intent and what’s written by the government, engage in a constructive way to make sure that we’re in compliance in all of the geographies that we operate in. So this is not an unfamiliar environment.
And as you think about military end-use, understanding, getting insight, gaining a perspective are not banned. And when you look at intent and again, engaging with government regulators, the way the language is written, it doesn’t look like the intent is an outright ban. There has been other situations where intent has been an outright ban and the language is written very differently.
And so, again, military end-use, it’s not – it’s a predominant aspect of the business as opposed to any aspect of the business. But we’ll go through our process and we’ll make sure we’re in compliance and doing the right thing.
Got it. And just the whole longer-term aspect of reshoring, right? And we have seen TSMC announced some plans to bring some production to new factories in Arizona. We have heard, Intel also step up and offer some suggestions on how it can have a foundry operation.
I understand that there is nothing near-term that back perhaps changes. But what is your longer-term view? How real is this reshoring of manufacturing capacity in the U.S.? And what implications does it have on Applied, right? You’re the largest semi-cap equipment company.
Yes. So a couple of things on this. And I would say as a result of the current environment and whether it’s pandemic-related, COVID-related, whether it’s geopolitical-driven, I do think there’s going to be a trend.
Broadly speaking, this is not a semiconductor comment, but it’s a broad comment. I do think there’s going to be a trend towards more localization of supply chains. We’ll see how it develops over time. And commensurate with that trend, I think, you’re going to see more intelligence and automation injected into supply chains than you currently see today.
So very broadly speaking, I think, that’s positive and bullish for semiconductors long-term. The TSMC announcement, you could consider maybe one step in this general trend, certainly, the first in our industry to make an announcement like this. I think, it’s helpful for our business for two reasons.
One is, when you think about WFE equipment, the most efficient version of capacity deployment is large integrated factories. So I do think that global supply will meet global demand. If the demand is for 100,000 wafer starts a month, the industry will build 100,000 wafer starts a month. I don’t think the industry will build 120,000 wafer starts a month to satisfy 100,000 wafer starts in demand.
So that’s not the inefficiency I’m getting at. What I’m saying is, is, one, 100,000 wafer start a month factory is a capital-efficient way to bring capacity online. It is more capital-inefficient to bring online two 50,000 wafer starts a month factories. And so I do think that this is a more capital-inefficient deployment of capacity, which is bullish for us in our industry, and I think that serves us well long-term.
The second aspect of this that I think is helpful to our business is, we’ve got a great services business. It’s enabling for our customers. And I do think the service entitlement on capacity built outside of Taiwan for TSMC and other geographies for other customers.
But capacity built outside of Taiwan, I think, has a higher service entitlement. TSMC has an incredible network in critical mass and strong nucleus of technical talent and an ecosystem that can be leveraged as capacity gets deployed inside of their home region.
Deploying capacity outside of that region, we will do everything in our power to help make them successful, leverage our network and our capabilities to make that as smooth of a process as possible. And that gets us to the conclusion that the service entitlement outside of Taiwan is probably greater. So I think that’s a nice adder for our business as well.
Got it. Let’s go to the longer-term drivers and we will talk about WFE and so forth. But the one very attractive aspect of Applied business is exactly as you mentioned, the services business. And we have seen services and recurring revenue businesses and other parts of technology get very high multiples. What do you think it will take for the semi-cap industry to start getting that right justified valuation for this very attractive part of your business?
I think we’re on the cusp of it today. I think time tends to solve these problems. I think there’s an increasing appreciation today versus where we were two, three years ago, certainly, on the importance and value of these businesses.
We’ve seen the resilience of our service business in the current memory downturn. I think investors have taken note of that. And the dampening effect that it has in terms of overall company performance. You don’t see the wild swings that we did see historically as part of the cyclical nature of the industry given the scale of our services business in the current environment.
And so I do think it’s going to be a matter of time. This was our first billion-dollar quarter. It’s fantastic. Our installed base is going to grow every year, downturn or not. Our installed base grows each and every year, which is the starting point for the growth model around our services business.
And then in the most recent environment, again, our fiscal Q2 was our first billion-dollar quarter for services business. 60% of our semiconductor services were what we call long-term service agreements. So it’s a strategy that was put in place back when our overall service business was below $2 billion.
Gary came into the company. He had a vision of what this business could be for us. He architected a new strategy. We got a new leadership team, and we’ve been executing pretty well. And now we’re on a run rate, where the business is over 2x what it was a handful of years ago. And now 60% of the revenues are from long-term service agreements, which are more subscription-like revenues.
And so I think, in terms of ascribing a higher multiple to that segment of the business, I think, time tends to solve these problems. And I think there’s a greater recognition today of just how valuable 40,000 systems and over 150,000 chambers as part of our installed base truly is over the lifetime of those tools, which in many cases is, 25, 30, 40 years in the field. It’s an incredibly valuable asset for us.
Absolutely. And do you think there will ever be a point where the industry would consider providing metrics such as billings or backlog, the kind of metrics that we often see software companies present to kind of show the strength of their recovered recurring revenue. Do you think it is possible to provide those metrics and have them actually be useful and be good representation of your services business on a longer-term basis?
Yes. I think that there’s maybe a potential opportunity longer-term. What I would say is, as I look at our level of disclosure around our services business today, and I would call it best-in-class. We’ve broken it out as a reporting segment. We give the profitability of it.
We talk about, in addition to the reporting segment, overall monetization of the installed base, which is how some people who – some peers, competitors, who don’t segment report like to talk about their service business. So we also look at it through that lens as well. 38% of our semiconductor revenue comes from monetization of the installed base.
So there’s a 5 point differential from us relative to the competition. And so I would consider what we have a best-in-class services business. Is there an opportunity for more disclosure? Potentially. But I think what we’re doing today is best-in-class and we’ll always look for opportunities to do more. But I think we set the bar pretty high in terms of transparency and openness around this business.
Got it. So let’s discuss some of the key drivers of the business. Last year was a tough year for the industry. The – your memory customers went through a downturn. And when this year started, I think, the expectation was for wafer fab equipment or WFE to grow 10%, 15% this year, and I think Applied was actually guiding towards the higher-end of that. How have your thoughts changed about WFE recovery in 2020 over the last few weeks, just given all the COVID headwinds and some of the other restrictions that you mentioned before?
Sure. Here’s what we would say about WFE today. First, our business is performing extremely strongly in the current environment. I talked about our semi-related businesses, systems and services, up 18% year-over-year in our fiscal Q1, 13% in fiscal Q2, 16% for the first-half of the year. Our memory-related businesses were up sequentially 25% in the most recent quarter.
So our business is performing really well. And we said that our systems business in our fiscal year would grow strong double digits. So we feel really good about how we’re performing. I do think there’s a debate going on in the market right now. Is the overall market going to be flat? Is it going to be down a little bit? Is it going to be up a little bit? I think it’s too early to tell ultimately, how the overall end market shapes up. But our business performance relative to that backdrop is incredibly strong, and we feel really good about how we’re performing.
I think the swing factor for WFE this year is what happens in the back-half of the calendar year. And specifically, as it relates to memory, we see foundry, logic playing out pretty much as anticipated. We talked about some weakness in specialty nodes, automotive, industrial, but continued strength on the leading edge, multiple customers, multiple nodes. And that’s hanging together really well. We like what we see in our Q3 and Q4.
Swing factor around memory gets back to, I think, what we see in 2020 is similar from a proportion of spend. $51.5 billion WFE in 2019; 60%, foundry, logic; 40%, memory; and balance between NAND and DRAM on the memory side. I think, you’ll see the proportion of spend similar, call it, 60-40 and balance between NAND and DRAM. I think that construct kind of holds.
The question is how much money will ultimately be spent? And I think it’s a function of how our memory customers dialing capacity adds, as they pursue technology roadmaps to modulate bit supply to meet bit demand in the market. They’ll be cognizant of inventory, cognizant of pricing.
But ultimately, they’re going to dial in a capacity statement, where wafer starts per month and NAND and DRAM fall for the second year in a row, will they add some incremental green fields to keep wafer starts per month on par with where we exited 2019? Or they dial in incrementally more wafer starts per month and raise 2020 above where 2019 is? I think it’s too early to tell.
So we haven’t come out with a new expectation around WFE. In this environment, we felt it was important to share how our business is performing and we want to see more data in terms of actions our customers take, return of health of the supply chain and how our competitors respond to that second dynamic as well before we provide more guidance on what we see for WFE, hopefully, that helps with our perspective.
Got it. So maybe just to say that in a different way, if we kind of break down first-half versus second-half industry, foundry and logic and memory. How do you think for the industry, foundry, logic first-half versus second-half and memory first-half versus second-half plays out from what we can see today?
Yes. So we don’t guide first-half by second-half. What I will come back to is a question that we were asked on our earnings call. That said, we’re exiting our fiscal Q2 with a record backlog, $650 million of unmet systems demand in our semi systems business. And if you take that out of the back-half of the year, what is the profile look like? And I made the comment that I wouldn’t necessarily say the second-half would be below the first-half.
So I’ll leave it at that, to give you a sense of what we see is underlying strength into the back-half of our fiscal year. We’ll monitor it quarter-by-quarter, and we’ll share on our earnings calls what we see. But I want to come back to an anchor on that question that was asked on our earnings call.
And the insight I provided in terms of true underlying demand and then resolution of unmet demand in the back-half of the year and into our fiscal Q1, and what that profile looks like. And I think that gives you a pretty good sense of what we see playing out.
Got it. Maybe the last one down on that. If we can reframe that discussion from an end market, what would cause memory customers to change their mind about spending in the back-half? Do you think is it just that cloud demand could decelerate? Is it the number of, I don’t know, 5G phones being fewer than what people are thinking?
So if you look back in history, what has really kind of swung memory demand, right? And how do you think about those end markets as we look into the second-half of the year?
Sure. I think in a traditional environment, you watch what happens to inventory. You work down inventory, both at our customers and customers’ customers, so the entire channel from an inventory standpoint getting to a more normalized level. And you look for recovery from pricing standpoint in the markets, and we’ve seen those traditional steps play out as we’ve progressed through the current memory cycle.
I think the COVID pandemic introduces another layer of complexity into the decision-making calculus of our customers. So you’ll have those traditional trigger points that will ultimately then lead to increased utilizations and then capacity adds to modulate bit supply.
I think the unknown introduced by the COVID pandemic is government containment actions impacting consumers around the globe. What does that do to overall global GDP global growth? And then what is that ultimately do to fit demand in the market? And I think our customers are trying to sort through and triangulate ultimately, how those containment actions ripple through consumer behavior and ultimately, global growth.
And then there is a debate going on in terms of shape of recovery, is it V shape? Is it U shape? Is it elongated U? Is it W? And I think there’s going to need to be more data to provide insight and a perspective from the customers that will ultimately lead to decision-making. So I do think the normal triggers still hold, but there’s an added degree of complexity in the current environment that bears watching.
Got it. I got one question on the panel, which is how do you think about any pull forward of demand, given the Huawei situation, because I think you mentioned record backlog, record orders. Are you able to kind of discern from that any pull forward of demand from a china customer perspective?
The short answer is we don’t see it. We don’t see it in the conversations. We don’t see it in the behaviors. Our expectations around domestic China spend are the same today as they were several quarters ago. We talked about an incremental $2 billion to $3 billion of spend, given what one foundry, logic customer – domestic foundry, logic customer in China said right before our earnings call, we’re now at the top-end of the range that was contemplated when the $2 billion to $3 billion of incremental spend range was created.
So now we’re just at the top-end of the range. There’s really no change in the expectations. When you break it down to the dialogue, when you break it down to the actions, customers will have multiple year roadmaps of how they add capacity around technology nodes. And we don’t see capacity that was scheduled for two years out being pulled into the current environment. And so we just don’t see it. It’s not what we see from a behavior standpoint from the customers.
Got it. Now, last year was a good one for Applied from a market share perspective. So you gained 50 basis points of WFE share. How are you thinking about the areas of potential share gain this year? And what will it take to get back to your peak market share?
Sure. So a couple of things. We talked about the first-half of the year, our semi business up 16%. We talked about Q3, our systems business being up high single-digit sequentially, up again into Q4, strong double digits for systems business in fiscal 2020. And that’s against the backdrop of an overall market that I know there’s a debate flat down a little, up a little, we’ll see.
Where we’re seeing signs of strength, you see it in our traditional leadership businesses CVD, PVD, implant, epi, thermals, just really strong performance across the Board. When you add to that some of the inroads we’re making on inspection and control, Gary, I think on the last call indicated that shortly, we’re going to be announcing a new optical inspection tool. It’s been out in the field and it’s gaining traction. So we really happy with the early performance.
Our e-beam business is performing well. We talk about creating integrated materials solutions under vacuum multiple processing steps. Kokusai is on the horizon for us. And so that’s not only going to add to the current revenue base, but we talked about an installed base of over 40,000 tools in our current business as we sit today. The Kokusai business is going to come with an incremental 10,000 tools. So a nice services opportunity for us over time.
So we feel really good about the company’s performance in the current environment. Again, we talked about in Q2, our – sequentially, our memory business was up 25%. So it’s across the Board, strength across the Board. Getting back to 21%, I don’t want to predict things. We’ll watch it this year. Does it happen this year? We’ll continue to monitor it. But what we do see is a lot of strength across the entire portfolio and how we’re positioned against the current environment. So we feel good.
Got it. 21% by 2021 is not a bad target to have. Maybe just a last one. Dan, as you’re getting towards the end of our time, your display business. The fiscal year guidance suggests a very strong growth for display in the back-half. Just how is your visibility there? And what is the right way to think about the longer-term growth opportunity for your display business?
Sure. So display this year is playing out how we generally thought. The slope and trajectory of the business is in line with how we would have used the business several quarters ago. We knew it was going to be a back-half loaded year, given what we were seeing from order patterns and plans of our customers to deploy capacity.
We talked about this year being a recovery year for mobile and a digestion year for TV, that seems to be playing out how we’ve thought about the business several quarters ago. And we feel pretty good about. We feel confident in terms of where we sit and that progression into our Q4. So we feel good about that.
Longer-term, this is a technology-driven industry. Where we compete, we’ve got 75%, 80% market share. It’s built on the back of critical enabling technology that allows us to differentiate ourselves. So we really love that position. The roadmaps going forward have key inflections, OLED into TV, foldable, flexible displays into the handset. Those new generations of technology are more capital-intensive than the current generation.
So we like the setup there. Displays are becoming increasingly important to the user experience around devices. So we like that long-term thematic trend of the importance of that technology to user experience differentiation.
And then as we think about the path forward, being exposed to more of the market that gives us another factor of growth to create deeper penetrations into adjacent spaces of the market that we don’t currently serve today, we continue to make progress on that and we’ll continue to drive that strategy going forward.
So we feel good about the business. We feel good about our performance in the current environment, and ultimately, long-term where the business is going.
Great. With that, Dan, we’re at the end of our time. Thank you so much for taking the time to share your insights with us, and thank you to the audience. And if you have any follow-up questions, please feel free to reach out to me. With that, let’s close the call. And thank you again, Dan. Really, I appreciate your time.
I appreciate the invitation to participate. So thank you. Take care.