Entegris, Inc. (ENTG) Earnings Call Transcript & Summary

Great. Good morning, everyone. I'm Stacy Rasgon. I'm Bernstein's senior analyst covering U.S. semiconductors and semiconductor capital equipment. And I can't express what an honor it is to have our guest here today, Bertrand Loy, the President and CEO of Entegris; and Bill Seymour, the Vice President of Investor Relations. Our session today is going to last about 50 minutes. Bertrand is going to give a few opening remarks to start us off just to talk a little bit about what Entegris is and what they're all about for people who may be a little less familiar with the company. From there, we'll move on to Q&A. [Operator Instructions] We'll have plenty of time for audience Q&A like that at the end. With that, it gives me a great pleasure to welcome Bertrand and Bill to our conference today. Thank you so much for being here, guys.

Thank you, Stacy. Thank you, and good morning, everyone. Let me maybe start with a very, very brief introduction on Entegris for those of you who are not familiar with the company, and then we'll open up the floor for questions. In 5 minutes, Entegris is a U.S. company headquartered in Massachusetts just outside of Boston. The company has been in existence for about 50 years. Our mission has not really changed all that much for all these years, and it is about developing process solutions to help our customers improve the product performance that they develop and help our customers improve their yields. To do this, the Entegris solutions can take different forms, and in broad terms, they align with the 3 divisions comprising our portfolio. The first type of solutions will be advanced materials that are part and parcel of the integrated circuits and defining -- so our materials really define the chip performance to a great extent. The second type of solutions would be filters and purifiers that are really essential to yield optimization. And the third type of solutions would be advanced materials handling, solutions that really help preserve the purity and the integrity of the materials and the chemistries during transportation. So we focus at Entegris, on applications and markets where the manufacturing processes are the most complex and the most challenges. In other words, this is really where we can create the most value for our customers and by extension, the most value for our shareholders as well. And when you have a focus like this, the semiconductor ecosystem is naturally becoming our primary area of focus, our primary market. And close to 90% of what we do is sold into the semiconductor ecosystem, broadly defined. Now when investors think about suppliers to the semiconductor industry, they usually think about large equipment companies. And they usually run for cover because they are afraid of the inherent cyclicality of the industry CapEx cycles. But I'm here to tell you that Entegris is very different. In fact, most of what we sell are consumable products, recurring revenues and while we have some exposure to the CapEx cycles, our unit-driven business model has proven very resilient across cycle for the last 10, 15 years. Another attribute that makes Entegris very unique is the growing importance of our value proposition to the semiconductor technology road map. At Entegris, we like to say that we operate at the crossroad of materials and materials purity, which really allows us to develop very unique co-optimized solutions faster than any one of our competitors. And it is important because when you think about the semiconductor road map, it is calling for further miniaturization. It is calling for more challenging device architectures. We've been hearing a lot about stacking more layers on the memory chips. And all of those architectures and all of those technology road maps are providing tremendous opportunities for Entegris to increase the content per wafer. And this is really what has been allowing us to outpace the industry. And those tailwinds are also what we expect will be up, putting us in a position to outpace the industry by about 3 to 4 points for the next several years. So you would say that though it sounds like ambitious goals and they are, but this is something that we've done. If you look at our historical top line performance, you will see that, for the past 5 years, our CAGR to our average growth has been 11%. If you look more recently at last year, 2020, our growth was about 17%, and our guidance for 2021 is 17% to 19%. So very focused on top line growth, obviously, but as a management team, we are equally focused on the bottom line expansion. We have a strong financial discipline. And if you look at our results over the last 5 years, you will see that our bottom line, our operating income expanded at about twice the rate of our top line growth. The final point I want to make is that, as a team, we have a very disciplined approach to capital allocation. We defined a framework about 10 years ago, and we have operated within that framework very consistently ever since. And the choices that we have made has helped us create what I would argue is really significant shareholder value over the recent 5 years -- 5 to 10 years. So with that, Stacy, I'll turn the call back to you for Q&A.

Fantastic. Thanks for that Bertrand. [Operator Instructions] Bertrand, in this type of format, I generally like to try to stay away from short-term questions if I can. Given this current environment, I don't think I can do that. So I am going to have to...

I was about to say it sounds like there's a but. But go ahead.

I don't want to spend a lot of time on it, but I do at least want to touch on because I do know -- we've already got some questions popping up on the pigeonhole. Most of them are short term. So I do want to just at least get to them. Just in general, I mean, like we're in a current situation -- at least in my dozen plus years of doing this, I've never seen in the industry in terms of this level of shortage in supply constraints, and you've been doing this a lot longer than I have. Can you just talk a little bit about what you're seeing in the industry regarding shortages and constraints, how some of those may be impacting you, both internally as well as externally? And anything that you're -- both the impact that's happening and anything that you're doing to try to mitigate it at this point?

Well, in our recent call, we talked about the fact that demand for our products is very, very strong. But we are facing indeed some supply challenges. Everything is relative, obviously, since we provided guidance for the year of 17% to 19%, which is much greater than any other electronics materials company, but it is true that we have supply chain challenges, which are all very actively managed, as you would expect. The first -- there are probably 3 types of challenges that we're facing. The first one is access to the freight capacity. I think this is well-known and not something that we can really do much about. But we expect that to, over time, become less of an issue when travel resumes and economies reopen a little bit more fully. The other 2 factors are things that we can manage and we are managing. So second bucket would be around our own capacity limitations. A lot of that has to do with the ability to fully staff 24/7 shifts, and labor shortages have been a challenge in the U.S. in particular. We've made a lot of progress over the last 2 months. And I would expect to see some benefits of those recent hires in Q2 and more of that in the back end of the year. And then the third bucket would be just managing our supply chains. There is some tension there, some of which come from the same labor shortages, some of which come from availability of raw materials. But again, we have a great team. They are doing the impossible to manage those pain points. And I think that, again, we will be in a position to deliver what I would expect would be a very, very strong performance in 2021.

Got it. Got it. And I guess, typically, the semi companies and the further back you go in the supply chain, the less read companies have on demand. Obviously, demand is extremely strong right now. Would you have any read at all on how sustainable the sustainability of that demand is in terms of true demand versus phantom, I suppose, right, in the wake of all of this? And I don't know how much of a read you guys would have on that. But like any color or any commentary you might have on that, that would be helpful. And then I think we'll move off the short term.

Yes. So there are many different ways I could choose to answer your question. I mean the first one probably would be thinking about our unit-driven business, which is really driven by the level of activity in the fabs. And everything we hear from our customers is that they are really running at full capacity right now. And their inventory levels are actually fairly low and in some cases, actually below their usual safety stock levels. So again, no reason for us to be concerned about an inventory build situation at this point. I think again, the fabs are operating full out. And that's good news because that impacts 70% of our business. The other component would be the industry CapEx. And I think that in the CapEx, I think there are still a lot of questions about the sustainability of the current trends over the next 2 or 3 years, and I don't have a perfect crystal ball. I am a little bit skeptical about the sustainability just simply because I think that there are many, many factors driving the elevated CapEx levels that we're seeing right now, one of which being nationalistic reasons by the U.S. and some European countries as well trying to bring back some semiconductor manufacturing capacity when decisions that have probably less to do with economic rationales and probably more to do with national security reasons, which, again, could put into question the sustainability of the current cycle. But again, for us, it doesn't matter all that much given the fact that we only have a modest exposure to CapEx, and as I was saying in my intro, I think we have demonstrated that we have a very resilient unit-driven business model. So something we keep an eye on, but I'm not losing sleep on it.

Got it. Maybe that's a good sort of segue into that whole like CapEx-driven versus unit-driven business. Can you maybe talk a little bit more about what that means? I mean, is units chips? Is it wafers? Is it layers? And then I guess on the CapEx side, like what sort of specific things are you doing that are CapEx-driven versus unit driven? And I think that, [ correct ], I think you said unit-driven 70% of the business, correct?

Correct. Yes. Sorry, it must be my French accent, usually, I try to...

No, I just wanted to make sure I...

7-0, yes, it's not 1-7. You're right. So let me give you some examples. So first of all, we use that nomenclature just because we are very unique as a supplier. You have actually very few independently traded materials companies, and we wanted to convey how different our business model was, how different the drivers to our business were. So chemistries, materials are almost -- have almost a 1:1 correlation with wafer starts. It's not exactly 1. I'm not going to go into that detail. I don't think it's for today, but it's a very close correlation. Then you have liquid filters, which the frequency of replacement is also very closely tied to the level of activity in the fabs. And some of those filters can be replaced every week, every month. But again, the cadence is fairly closely correlated to wafer starts. So all of that are examples of unit driven. And for us, unit driven, to be qualified as a consumable, you need to replace those products within 18 months. Everything that gets replaced less frequently is viewed as CapEx. So we have products such as wafer carriers, the FOUPs that you see in the automated systems in the fabs. Those products get replaced but get replaced every 4 years, every 5 years. So they -- we consider them CapEx. And then we sell systems. We sell gas purification systems because the way you purify gases are through those large systems, and you sell them once when the fabs are being built. And that's really the ratio. About 70%, 7-0 will be unit-driven business product lines and 30% would be more of a onetime sale or less frequent replacement products.

Got it. And you touched a little bit on this in your opening comments in terms of some of the secular drivers. You talked about things like more layers, more complexity. Can you talk a little bit maybe about -- and maybe in a little more detail about some of those more secular demand drivers. I get the idea that semi growth itself is strong, and I get the other -- the drivers. But how does that increased complexity like specifically affect you? Is it just a matter of like more layers? Or is it -- are you developing new materials, new technologies to attack this? What are you guys doing on the secular side there? How does that help you?

Right. So actually, let me provide a context maybe for your question. I mean the reason why we believe we will be able to continue to outpace the industry by 3 to 4 points is because we believe we can increase the Entegris content per wafer node after node after node. And the reason we believe we can do that is because on each wafer, you have, in fact, more layers. So you have -- if you think about what our customers are trying to do, they are trying to increase wafer density. So gate density in logic, bit density in memory, and they're trying to do that in 2 ways. One is miniaturization of the features, and then it's really stacking up layers, so the vertical scaling. And the combination of those 2 factors provide tremendous opportunities for us. As they do that, the architectures become more complex. They need to move away from the commodity materials that they've been using for decades. And they are migrating to more advanced materials that have better electrical properties, better structural properties to enable those high aspect ratio structures. And that presents significant challenges and significant opportunities for a company like Entegris. So we are very actively developing those new materials that can be deposited in thinner films, thereby reducing the aspect ratio challenges. We're developing new chemistries to do very conformer etching up and down those very narrow and deep structures. The list is long. I mean go back to the recent Analyst Day presentation. We'll see some examples of that in our CTO presentation. But again, it's a target-rich environment for our materials platform. And it's a target-rich environment for our liquid filtration platform as well because those architectures are increasingly susceptible to smaller and smaller contaminants. And because of that, our customers need to adopt more advanced filtration. They need to change the filters more frequently. And increasingly, they need to request higher level security for their chemical suppliers. So our SAM is expanding very rapidly as a result of all of those factors. So back to your question. Yes, the big driver is wafer starts, but it is also the increase in process steps to process the wafers. And it's also a function of the growing complexity of those architecture and the further miniaturization of the features on the wafer -- on the surface of the wafer.

Got it. Got it. And if I think about that growth algorithm, it feels to me like you've actually gotten more positive over the last year. So my understanding is the algorithm as you grow, you think you can grow 2x GDP plus 300 to 400 basis points above the market. I think when you were here last year, it was 200 to 300 basis points above the market. So you've taken it up in the last year. Is that just a matter of like the strong demand we're seeing post-COVID? Or did something else structurally changing your view of what Entegris is actually delivering?

So you're right in terms of the algorithm for the formula. One is strong conviction in the semiconductor industry itself, and that's the 2x.

That's the 2x GDP.

Right. And that is the tide that will lift all boats on the water, right? Now there's no glory in just going up and down with the tide, right? So we want to do better than that. And that's why we have the objective of outpacing the industry. And you're right that our goal today is more ambitious than last year. And the reason for that is that it's twofold. One is more wafers today are produced at the leading edge as opposed to a year ago. And that's something we do not control. I mean, our customers are really controlling the number of wafers produced at the leading edge. And with the acceleration driven by COVID-19 and the digitalization of our lives, we've seen a big push to those higher-performing servers and computers and everything else. And that has led more wafers to be produced at the leading edge where we have higher Entegris content. So that's one reason. The other reason is our teams have done a tremendous job at positioning Entegris on the newer architecture. So when I look at the pipeline of opportunity today, it is of greater quality than it was a year ago. And a year ago, it was of greater quality than it was 2 years ago. And I think it has to do with the broad statement I was making earlier, which is our value proposition is becoming increasingly important. We have more and more opportunities to jointly develop novel solutions with our customers. And again, the rate of migration to the new nodes is picking up. In fact, it's not slowing down. And if you combine all of those factors, that is what gives us the conviction that we will be in a position to outpace the industry by 3 to 4 points.

Got it. Your comments on leading versus trailing content, I find are interesting because we've got constraints right now. They're almost worse on the trailing edge right now than they are on the leading edge. I've actually been wondering, the industry historically has not added a lot of trailing capacity over time. Today's leading edge becomes tomorrow's trailing. And I've actually wondered if that needs to change, if we need to add more trailing edge capacity. Like is that like a positive or a negative or a neutral for you guys if more of the addition mix goes to trailing edge? It sounds like your content is a little less, but I don't have a good idea of the magnitude.

So I mean, firstly, short term is very good because it means that the mainstream fabs, the trailing edge fabs will be running at full capacity. And I think that's going to be the case for at least the next 1.5 years to 2 years.

It feels like that, yes.

Then -- but then I think what will likely happen is there will be certainly some capacity added at the trailing edge. But I think that what I would expect actually, in fact, is more of the trailing edge to transition to the near leading edge. So more of the applications that used 65, 45-nanometer architectures moving to 28 nanometer and maybe even more advanced. And that's great for us because it means that I think there will be an acceleration of the transition to near leading edge, where we have, in fact, greater content per wafer. Because if you look at the logic, if you go back to the Analyst Day presentation that we made in November of last year, you will see that the migration to newer nodes is actually much faster in memory typically than in logic. But I think that because of the shortage in trailing edge chips that we are seeing today, I could see actually an acceleration -- an accelerated transition to the near leading edge. And that would be actually very positive for us, obviously, if it happens. It's still a big if but still speculating.

Yes. There's a lot of big ifs out there right now. I wanted to ask you, you brought up memory versus logic. How does your content like look versus like between logic and memory? I mean, again, I could imagine -- my dumb hypothesis would be that maybe it's better for memory, especially NAND, just given the sheer number of layers that are there, but like maybe I'm thinking wrong. Like how do we think about the opportunity across the different end market buckets?

So the picture is evolving very, very quickly. We had very little opportunities in memory architectures 5 years ago. Things changed very quickly when NAND architectures migrated to 3D structures. And that accelerated even further when they crossed the 64-layer mark. So again, I think if you step back, at the end of the day, we sell solutions to very complex architectures. So the more complex, the more opportunity. Memory architectures were not that complex not too long ago. It's true for DRAM. It was true for NAND as well. Things have changed radically, both for DRAM and NAND. And that's why the opportunity today for us on a memory chip is becoming much closer to what we see in logic. So today, in logic, we still have the largest opportunities per wafer in terms of Entegris content, but memory is catching up really quickly. And that's good news because I think it's actually going to bring additional resilience to the model, and we will have obviously more opportunities for growth, which is a good combination.

Got it. And what about as logic now is increasingly finally migrating to mainstream EUV, is that a positive, a negative or neutral for your business?

So it is a positive, clearly. EUV will make it economically viable to further miniaturize the features on the surface of the chip, which means that it will require more advanced filters. I mean, you understand the virtuous cycle that goes with miniaturization. So all of that is really good. So clearly, highly favorable to our microcontamination platform, highly favorable to our AMH platforms. We had concerns early on about what it could mean for our cleaning business, our cleaning chemistries because EUV was associated with the reduction in process steps. But it was a theoretical risk. I mean the reality has proven otherwise. I mean, there's certainly some reduction in process steps, but it's short-lived, and we were seeing that. Without mentioning the customers, I mean we have now tangible actual data suggesting that it has a very, very small impact on our materials business, so overall, long term, very beneficial, obviously, enabling further miniaturization and in short term nothing really to worry about.

Yes. I mean I have to imagine like the tolerances on EUV or like in terms of what it can withstand are so much lower. Like for filtration and microcontamination, I mean it's got to be a positive, I would think.

Exactly. So it was a positive, obviously, within the fab environment, but increasingly, it drove new requirements for the resist manufacturers in particular. I mean they had to start adopting very advanced filtration requirements, adhere to much more frequent replacement of those filters. All of those choices, by the way, were dictated by the fab customers. So -- and then, of course, in order to preserve the purity and the stability of the resist during transportation, they have to migrate to more reliable and pure packages as well. So again, great opportunities for 2 of our 3 divisions.

Got it. Should we talk about the businesses a little bit, dive in?

Sure.

All right, good. Maybe we start with the specialty chemicals. Can you talk just a little bit like what that is, like where you see the growth prospects, the margin structures and how those are evolving? But maybe we'll start there and we'll go to the other ones.

Okay. So SCEM is a division that I expect to do extremely well over the next few years. We expect them to outpace the industry by 3 to 4 points. And a lot of that will come from some of the solutions I was describing at high level. The industry will be looking for alternative materials to replace, tungsten to replace copper, to replace a number of materials that have been used for decades by the industry. We are very focused on those new materials, whether it is molybdenum or ruthenium or other types of molecules that will become high-volume molecules in the next 5 to 10 years, so very exciting to develop those materials. And then the other area of focus will be the etching chemistries that will be required for the higher layer count architectures in memory. We talked about that, but also will be required in logic when the logic architecture migrates to gate all around technology for instance. It will have to -- we will have to enable very precise etching around those nano sheets. And I think that's going to be a great opportunity for Entegris as well. So those are just examples, but there is a long list of opportunities around this division. And the margin profile will continue to be at about the corporate average and expanding over time.