MKS Inc. (MKSI) Earnings Call Transcript & Summary

All right. Good morning, everyone. This is Krish Sankar from Cowen. The next company presenting is MK Instruments, and we are fortunate enough to have John Lee, President and CEO, here. John, thank you very much for your time. I really appreciate you attending this conference. And maybe I'll just jump straight into Q&A.

It's very interesting that you guys last had an Analyst Day in December. And then in the 5 months or so after that, the industry has grown rapidly. And your last earnings call was a testament to that where you had very strong numbers. And what is interesting to me was that your guidance, based on your guidance for June, you're already scratching the surface of your 2025 longer-term model. So I'm just kind of curious are there any puts and takes we should think about on the margin front first, either -- that could either be biased to the upside or downside as we move along. Given that the industry has grown rapidly in such a short amount of time?

Yes, great. Thanks for the question, Krish, and also thanks for the invitation. Happy to be here. Our long-term model is still the same. It's intact. And that model is a 50% flow through on gross margin with incremental revenue and a 40% flow through on operating margin with incremental revenue. And you're right, we've done a lot better than that in the last 4, 5 months. Part of it is that typically, when we do a ramp, when there is a ramp, we tend to do a little better because cost -- take a while to catch up. And then on the way down, the cost -- takes a little while to shut the cost down. But overall, long term, we think that model of 50%, 40% still is intact. And I think as we drive revenue up higher, we'll certainly try to work on improving that margin, but right now, I'm very comfortable with that model.

Got it. Got it. All right. Makes sense. And I got to say the earnings power has been pretty impressive. Then switching on to the semiconductor side, kind of where are the inventory of your semi cap customers? I understand that the whole ecosystem is supply-constrained today but where do you think the inventory is? Are they building inventory at this point? And where do you think your customers' equipment lead times are today?

Yes. Well, we probably -- we don't typically comment on our inventory levels of our customers specifically. But I can say this from our daily operational meetings with all of our customers, we don't see any let up on the demand. And certainly, anything we can ship they are using, and it's going out within the lead times of their equipment builds. I think there could be areas where people would want to build that inventory, given the constraints over the last couple of years. With COVID, in Asia, [indiscernible], everybody wants to be a little careful, but -- and there may be a little bit of that, but I think if there is, it's in small pockets because we're constrained, frankly. And everything we can ship is needed for their shipments. And so I think we're in the part of the cycle where I'm not that concerned yet about any kind of inventory build.

Got it. Got it. And I mean, you probably don't have the visibility into this, but when you talk to your customers, are there any concerns that you guys discuss in terms of like double ordering or any such thing. I understand that happens more for the downstream. But I was just wondering, is that something -- a topic that comes up in conversation with your customers?

Yes, that's probably something that was -- should have come up as a conversation 5, 10 years ago when everybody thought we're going to win this year, but I think the industry has matured. And I think our customers know when they're going to win and know when they're designed and know when they're not. And so I feel like because the industry has matured, the customers we have, have pretty solid market shares in the markets they play in. I don't feel like there's any kind of double ordering. So I don't think there's -- well, if you could build it, maybe the end user will take it. I think processes are too complex for that anymore. So I think it's mostly not double ordering, I think it's ordering because they got to ship it and they have a PO on their side.

Got it. Got it. And then talking about share, it seems like you've spoken about gaining market share in the public supply business and some of your competitors also playing the same. So I'm just trying to figure out, can you just help us understand if I look at the power supply ecosystem, let's put it in 3 buckets: RF, DC and RF match, where do you compete? And who are your competitors in each of those 3 verticals? Then on a product basis, is it more on the debt side? Or is it more on a conductor directory catch? Can you just give some color in terms of 3 power supply verticals and the different end products?

Yes. No, happy to clarify that. So we do not provide DC power supplies. So we do not compete there. We supply RF power supplies, radio frequency power supplies and RF match. So those are our 2 segments. And some other customers or competitors supply all 3. But in those 2 segments that we compete in, the biggest share gain published by third-party in 2020 was the R Power segment. And that was that 1,100 points a share gain. And overall, it was 600 basis points. So Match growth grew a little bit, too, not just as much as R Power. A lot of it is driven by the work we've done over multiple years and investments we made 5, 7 years ago in R Power when we saw the inflection -- the market inflection of vertical scale. Obviously, when VNAND came across, I think, came out, it surprised everybody was like, wow, you can actually do that. And -- but we jumped on and we said, this is going to have implications for certainly R Power, certainly in etching of these highly -- high aspect ratio vertical features. And so we made those investments. So that's a big driver for us. That's a dielectric etch application. We're taking that kind of technology to connector etch as well. Now dielectric evolve -- is evolving faster because VNAND is evolving fast. Every year, you get another 18 layers, 32 layers, whatever you want to call it, onto the last node. Connector etch is evolving less quickly. So that will take more time as we gain share there because usually that's when you change is where a new application or new technology is required. So hopefully, that helps. It's R Power, big share gains, the culmination of multiple years of work. And the story is not over. That story is not written or finished yet. We will continue to invest there because we see continued opportunities for R power and Match. And then I think verticality, this inflection is really what's driving a lot of this great inflection for R Power. And by the way, by the way, verticality also drives ALD, atomic layer deposition, as many people know. And we also supply ozone generators to that market. We have always supplied that. It's just that now that market is becoming more and more important as well.

Got it. Now on that path, one of the things that some of the semi cap customers have started talking about although it's still further down the road is the verticalization of DRAM, like 3D DRAM down the road, kind of like you said, 3D NAND really helped make dielectric etch more complicated, more time-consuming and therefore, more power intensive, do you visualize something that if the industry moves to 3D DRAM to? Or is it too early to make a call?

Well, there's a couple of structures that are being talked about for 3D DRAM, not stacking, but 3D DRAM. By the way, in stacking, there's a lot of vertical structures there, too. So again, that also is good for R Power. But to your point, 3D DRAM, if you look at the maybe 1 or 2 options, options that are being talked about, it's going to look a lot like DRAM. It's going to be turn it sideways, connected from the side and stack. And when you do that, it is a vertical structure with lots of high aspect ratio features. And that, I think, would be great. Again, for leaders in dielectric etch and R Power to enable that dielectric etch, I think that's a great place to be. Now I don't know when that will happen, but people know that you can't continue to shrink in 2D. NAND was the first to go through that transition. I think DRAM has to eventually go through that transition as well.

Got it. Got it. And before we move to advanced markets, I still had a couple of questions on the semi cap side. We have seen a lot of demand come from China as we grow the semi industry. There's also a growing ecosystem of semi cap companies in China, although they're really behind the U.S., European and Japanese counterparts, do you see a potential risk of a similar kind of evolution in the subsystem component space of some domestic Chinese competition?

Yes. We always evaluate that. We're certainly always paranoid about competition from any region or any area. And I guess we haven't seen a lot of that, Krish, and that's not to say it can't happen, but I think when you look at the complexity of the OEM tools, even the OEMs in China have to match the same capabilities as the leaders today, or else, the fab won't buy those tools. And when you are making tools that are that complex and the processes are so tight, you need critical subsystems that are very precise. They've continually been innovated over many decades to get there. And to your point, Krish, it's not like every critical subsystem over 20 years has evolved, some have become commoditized. But we've gotten out of those, by definition, because they are commoditized. So the critical subsystem part of WFE, by the way, has actually grown over the decade as a percentage of WFE. So critical subsystems is a great place to be. WFE is a great place to be. And the competition to innovate continuously, that is only for a couple of companies that have been able to compete against us in every category that we have an almost -- and no company can compete against us in every category.

Got it. Got it. Very helpful, John. And then maybe switching gears to the advanced market side. It looks like that's mainly the laser business and the PCB business. So talking about on the laser side, last year was a tough year driven by COVID, macro, China, et cetera. This year, it looks like the business has recovered pretty nicely. How do you think of the advanced markets over the next couple of years? Do you think that once you head into the back half or later next year, the year-over-year comps start getting difficult, but do you think it is still a long runway left on the laser side.

Yes. Well, the comps will get more difficult probably because there was a bad dip in the year because of COVID and trade wars, but I think the way we look at Advanced Markets and our Surround the Workpiece portfolio, which includes lasers is that this is a long-term trend. And this long-term trend, the way we think about it is about Advanced Electronics. When you think about Advanced Electronics, the first question is do you like that? Do you think it's going to grow? Yes. And I think it's because it's useful, right? And everybody is going to use more of it. And going to generate more data, so we need more servers and, et cetera. When you think about Advanced Electronics, you and I do not buy a chip because it's cool, we buy a phone because it's cool and it's useful. And more and more of the ability for that phone to be better every year, prompting us to want to buy it, is not just the chip, it's the packaging that goes into it. And that packaging requires laser-based processing because it's miniaturized so much. And it requires Advanced Packaging techniques. So laser-based via drilling, but the flexible circuits are needed now to connect the 5 cameras or going to 6 or 7 cameras soon, the touch sensors, right? All those sensors have to be connected to the chip and to the motherboard, you have to make them connected in a smaller and smaller space and volume, and that's what Advanced Packaging is all about. And so that's why we think Advanced Markets has a long runway because it's about Advanced Electronics. And if you like that, you've got to like everything in that supply chain, that's a profit pool for making that Advance Electronics. And that's how we think about lasers and chips, semiconductor impact.

Got it. Got it. And I just want to talk about competition a little bit on the laser side. But before we get to like coherent or the western competition, I want to talk first about combination in China. The reason I'm asking it is because I understand you don't compete with IPG Photonics, but if you look at the fiber laser ecosystem, the raycus, Maxphotonics of the world started developing a fiber laser 10 years ago. And the last 3 years, it's become a real product and they're competing with IPG. And you folks actually have a pulse laser, which is very differentiated, but is there a risk that you worry that Areca or Max Photonics or enhanced laser could start working on a pulse laser product and kind of have the same thing that happened with fiber lasers, 2 years down the road?

Yes. Well, I would say this, some Chinese companies have already been doing pulse laser sort of decades, actually decades. But as of today, they're really addressing the lower end, the easier applications where we actually don't compete. So it's similar to any kind of critical subsystem that we have, which is at some point, it's either going to continue to require continuous innovation, in which case, we love that, and we think we can win or become commoditized. And it becomes commoditized given enough time, everybody can catch up. But in lasers and pulse lasers today, our view seems to be in our experience is that there are some easier applications, lower power, less critical pulsing less stringent pulsing specifications where there are Chinese competitors and other competitors that can do that. And -- but at the most advanced, most critical pulsing energies and pulse rates, at wavelengths, that is where we compete, but only seems to be with the western companies that have been at it as long as we have.

Got it. And then turning about the western competition, obviously, beyond -- so what happened at Coherent. And so now that they're part of II-VI, do you feel like the pico and femtosecond laser competition is going to heat up now, now that Coherent is part of a more aggressive organization.

I don't know if it's -- that they would be more competitive or less because they were always a good competitor, Coherent by themselves. I don't know if II-VI will add more to that or not, but that remains to be seen, but we always have had 2 plans to grow our product lines and businesses where we think we want to grow. And one is organic and one is inorganic. We tried inorganic, we didn't get it. But even before that, we had an organic effort in picosecond lasers that we talked about. And so now we go to plan B, right, which is organic. And we'll take the same playbook out. It's maybe not apples-to-apples, but very similar to our core. We believe that this is going to really enable growth in the future because of some inflection, then we will double down on the R&D there so that we can take that share. And so we talked about multiple double-digit design wins in picosecond lasers already even before the whole coherent debenture occurred. And so we'll just have to go to plan B and grow organically.

Got it. Makes sense. Makes sense. Then switching a little bit to the PCB side. Is the demand there still primarily coming from smartphones? And the reason I'm asking it is we have seen in the past, whenever there was like tightness in MLCC, some of the business slowed down, and now in the last few quarters has been very strong because of smartphone, how much of it is smartphone driven? The reason I'm asking is there are some concerns that you may see some slowdown in smartphones in the back half, mainly because of increasing COVID cases in India and Southeast Asia?

Yes. Yes. And I see that, too. And I think that's unfortunate, but -- of what's happening in India for sure. A lot of the via drilling for flexible circuits is driven by smartphones, that's true, but not every smartphone is equal in terms of how much flex content they have. And so the most advanced smartphones, think of the guys in Cupertino, they have the most flex. And so therefore, if you're levered to that supply chain, you may not see as much of an issue as the lower-end smartphones. They're very capable, but don't use as much flex, may have some kind of headwind. So right now, we seem to be okay there. The demand seems very strong for the flexible via drilling type of tools we have as well as MLCCs. But we also have kind of a tailwind of wearables. More and more wearables are -- they have to have flex. Airbuds, rings, watches, all that stuff that is tiny requires flexible PCBAs circuits as well. So there's also this kind of tailwind of wearables that helps as well.