Qnity Electronics, Inc. (Q) Earnings Call Transcript & Summary

Maybe to round out your Interconnect Solutions business, and then we can get to the semi side of things. On the PCB side, it seems like there's 2 kind of diverging trends in that business. You've got high-end AI PCBs going to servers, whether it's CPU, GPU. But then you've also got some non-AI PCB business, some smartphones and PCs. Those businesses aren't doing fantastic. So can you summarize how you're thinking about the PCB market and what trends you're seeing there?

Yes. So if I started and I kind of combine the last 2 questions. So when we think about thermal, we talk about kind of our big 3 leading-edge technology positions in the interconnect space. Thermal management is the largest of the 3, then advanced packaging, and then the AI PCBs would be #3. Together, those 3 comprise about 30% of our interconnect portfolio. In the first half of this year, those 3 combined were all growing more than 50% year-over-year. So really strong growth from all 3. Specifically, what we're seeing on the circuit board side is back to some of the dynamics that I talked about around the high layer count boards and the high-density interconnects where you're trying to get smaller and smaller geometries so that you can maximize the circuit space on every board, right? And when you do that, fundamentally, that upgrades the material set that you have to use in order to get the electrical performance and the reliability that you need at the smaller geometry size. So we've seen some nice share gain opportunities there on top of the content gains as our customers more broadly adopt those technologies in there. But we've also seen what I would call a little bit of I call it a data center halo effect in that as customers put in the equipment to do these higher, more capable circuit-level boards, they're not just limiting the sale of those boards to their data center customers. They're making them available to their consumer device customers, they're making them available to their automotive customers. And what we're seeing is that several of those customers knowing that eventually, they'd like to see AI move from the cloud and the data center to the physical AI world of vehicles, devices and machines. And in preparation of that, they're adopting some of those architectures even early so they can get a sense of how that's going to fit in within their overall system dynamics. And so we're seeing them already adopt some of those architectures and some of those higher, more sophisticated boards into their application. So we're getting that effect -- the growth, not just in data center, but we're seeing really solid content growth and opportunities in automotive, in premium consumer electronics, in some of the broader industrial applications, even in environments where unit volumes may be a little bit flattish.

So let's switch over to the semi side. I think a good starting point would be as we think about the big segments of semiconductor foundry logic, leading-edge, you've got mainstream nodes and you've got memory. Can you summarize what you're seeing from a fab utilization perspective and where volumes are in each of those big buckets?

Yes. So certainly, the largest, most significant growth is coming from advanced logic and HBM, DRAM, right? We see utilization in advanced logic. Kind of in the end of the second quarter, we talked about it kind of being in the mid-80s, kind of probably going into the high 80s, maybe even into the low 90s as we moved into the second half of the year. DRAM and HBM is already there. They're already operating in the high 80s, maybe low 90s. I'm not sure how much more we can squeeze out of there. That's pretty much as sold out as you can get, although if there's a way to squeeze more out, they'll figure out how to do it. On mainstream logic or legacy logic, that's kind of -- both mainstream logic and NAND are kind of both sitting in the low 80s right now. I think we're having very constructive conversations with the mainstream logic side of the house. It's a little bit uneven. Those who maybe have more consumer exposure are trying to -- they're struggling a bit more, but many of them have figured out how to really exciting growth opportunities, whether that's in data center or some of these other industrial applications. And net-net, we think that's a positive for overall utilization. We expect kind of continued cadence of slow and steady improvement in legacy logic. And NAND continues to continues to inflect upward as well, a little bit more gradual. I know there's a lot of work going on to NAND to say, can you utilize anything in NAND technology to release some of the bottlenecks in DRAM or HBM. We're excited by those opportunities. We're working with customers on some of those. But ultimately, we're also trying to help them advance to the next node of scaling within NAND as well, whether that's getting from 300 layer counts up to 400 layer count NAND architectures.

And then on that -- on the mainstream side, higher memory pricing, are you seeing like a slowdown from higher memory pricing from -- impacting that mainstream business?

Yes, it's something that we're watching across our mainstream customers. I think the ones that have more consumer exposure, I think it'd be naive to the -- our customers have been pretty transparent in terms of they've made public comments around it's restraining their growth a little bit. What I've been encouraged about is how nimble and adaptive and agile many of them have been to be resilient even in a downturn with consumer -- with some of their consumer electronics to find other ways to find meaningful growth opportunities. And you see that in some of the power and sensing modules that they're commercializing within data centers. But some of them are -- many of them are doing quite well in automotive. And really in the premium consumer electronics, when I look at our business, even in a year where volumes are a little bit restrained in consumer electronics and in automotive, we're seeing high single-digit to double-digit growth in both of those segments because of the content uplift that we're seeing. And some of that is coming through mainstream logic. Others is coming from kind of the more back-end technologies.

On the leading edge side of things, we're at a very exciting point. We're starting to move in the 2-nanometer and sub-2-nanometer nodes coming into really high-volume manufacturing. Can you remind us, how does your content per wafer scale as you move from FinFET node to gate all around? And maybe as you go from like 2-nanometer, to the sub-2?

Yes, it's a great question. And I think this is -- I get really excited about these technology migrations and the opportunities that they create. So I think when we think about going from like FinFET to gate all around architecture, fundamentally, there's a layer scaling effect that benefits the material intensity. So if you go from like, say, a 14-nanometer construction to a 2-nanometer construction, you'll roughly double the number of layers that you're doing. So there's a nice -- particularly -- and where that benefits us the most, where we have the most exposure to these content uplift is really in the chemical mechanical planarization or polishing part of the semiconductor process. That's a little bit more than 50% of our Semiconductor segment. So anything that increases the CMP process intensity is going to have a disproportionate benefit. We're the market leader for both polishing pads as well as cleans. We've got a strong slurry offering as well that's been growing very nicely for us. What I'm really excited about is not only do you see a doubling of the layer count that you're using on the number of layers within those architectures, but several of those layers, especially on the advanced logic side, the process intensity is such that you're doing multiple polishing steps in several of those layers. So it's almost a double multiplier effect. If I take back, we don't disclose the specifics by customer because they get into some proprietary information. But there's some pretty good third-party research out there that talks about the material intensity increasing for each node migration of, say, somewhere between 20% to 30%. And for us, if you think about a migration that is a little bit more CMP-intensive, you'd be kind of on the upper end of that range, something that's maybe a little less CMP intensive. It's a little bit on the lower end. As we move forward and we look forward to the angstrom-era nodes, what I'm most excited about is the growth and the adoption of backside power. Right? And we're starting to see some of the early adoption of that and even with some of the 18A technology, but getting the backside power in some of the 16 and 14 architectures because that's a real step change in CMP process intensity. And our teams have been actively working on 16, 14 and even 10 technology more than a year now. And the number of wins that we've seen, kind of a significant number of wins in kind of both the 16 and the 14 node migrations, particularly as they adopt backside power architectures.

What is more important in CMP, pads, slurry or clean?

So look, there's probably others that have different opinions. I think my answer would always be pads. Right? I think that's where the engineering and the architecture of the process of the CMP process really starts with the pad. The pad controls the interface with the wafer. The pad also controls the interface and the mechanics of the flow for the slurry. And the efficiency of that process combined ultimately determines what -- how much of the cleaning solution is actually necessary. Because if you had a perfectly efficient polishing process, cleans wouldn't even be necessary. Now fortunately, for us, that's never the case. You always need a cleaning solution. But ultimately, we think that -- and for us, the pads business is our largest business within CMP. And we have a sizable cleans and slurry business as well. As we get to more and more advanced angstrom-era nodes, the degrees of flexibility in terms of how the technology profile and the window operates become smaller and smaller and more challenging. So there's more and more opportunities to leverage technology across those domains to help optimize yield and performance for the customer. Because ultimately, it's about yield in the fab, right? All the customers care about is if you can help them with even 1%, 1 point in an advanced node fab, it's worth $200 million. So if you can get even 0.1 point of yield through something that we're doing on the CMP step, it's worth more than the price of anything that we have because the bill of material on everything in that CMP process step is very small relative to the value of 0.1 point of yield.

Do you -- what is your capacity situation in CMP? Because it is growing very strongly. Are you fastly constrained anywhere?

So we've been steadily adding capacity in CMP really since the peak in 2022. So in 2022, our cleans and slurry business, for example, we're only manufacturing in the U.S. And consistent with our local-for-local model, we knew we needed to add capacity primarily in Asia. So since that time, we've added cleans and slurry capacity in Taiwan, Korea and China. And as a result, we've seen that business continue to grow significantly over the last several years. In addition to that, we've added capacity in the U.S. and in Taiwan and in Korea for pads. We announced 2 large projects, capital expansions in the first quarter, 1 in Delaware in the U.S., the other in Taiwan. The first line, both of those are oriented -- or the first thing in both of those facilities will be expansions to our pad capacity. The first line is up and running in Delaware. We're doing customer qualifications now, and we should have commercial sales maybe by the end of the year, knock on wood, but certainly will be commercial volumes next year. And then the next phase of investment is already going into that facility. We're doing investments in the Taiwan facility. We'll have the equipment installs done by the end of the year, and we'll be in qualification process. early next year. And depending on how that goes with customer qualifications, we'll have volumes out of that facility sometime next year.

Great. I want to talk about pricing. You guys are firing on a lot of cylinders. The cycle is turning up pretty strongly. Can you talk about how you guys are approaching pricing from a material side? That -- you're passing through cost increases, are you getting opportunistic? How are you guys approaching the pricing side?

Yes. It's a little bit of both. I would say, going back for several years now, whenever we see kind of market dislocations on raw material costs. Sometimes, that could have been tariff induced. Sometimes we saw some of that during COVID. We certainly have seen it with some of the Middle East disruptions. We pass those costs directly on to the customer. There might be a quarter-to-quarter, kind of maybe a 1 quarter lag on the timing of that. But we talked about earlier this year, in the first quarter, we talked about just what's going on in the Middle East having about a $20 million impact. Really not on our raw material buy, it's mostly on freight and logistics. And so we've already taken actions to pass those costs through to customers so that it's neutral from a margin perspective for us for the year. By far, the more impactful way that we think about pricing though, is on new product launches. So every year, we're launching kind of a steady stream of new products. I think there may have been even some announcements in the news flow today on some new products that were launched in our Interconnect business. But typically, the way we think about pricing is to price for value on the launch of the new products. And just mechanically, the way that shows up in the numbers is because there's no prior reference point, there's no reference point in the prior year for those new products. It shows up as part of the volume number. right? So 1 of the reasons why the volume outperformance has been so strong for this year is, yes, we're getting content gains, but we're also getting pricing power from the new products that we're launching. And then on the legacy products, there's always a little bit of a technology price paid on legacy products. That's kind of flattish this year. Historically, it's average kind of minus 1% a year. That's a little bit less of an issue this year. But where we're seeing significant benefit is on the launch of new products.

Interesting. Maybe in the last couple of minutes that we have, I want to talk on the margin line. So can you talk about with all this revenue growth, how do we balance this against your EBITDA expansion? You've got some nice EBITDA expansion. Some of your revenues do carry lower relative EBITDA margins, if I'm not mistaken. So can you talk about how you're viewing EBITDA margin expansion from here?

Yes. I think we've got -- we're really happy with kind of an industry-leading margin performance kind of where we sit today. But we're excited with opportunities to continue to see margin expansion and margin growth, I would say, fundamentally driven by kind of 3 core drivers that will contribute to opportunities for margin expansion over the next few years. The first one and maybe the most obvious 1 is just the volume growth that we're seeing and the natural operating leverage that you get from volume growth through our existing facilities, that creates a nice tailwind. The second one is really all the fastest-growing products that we've talked about, whether that's the advanced nodes within the semiconductor realm or the advanced packaging in interconnect and the thermal management on the interconnect side, all of those -- the fastest-growing parts of our portfolio are also the highest value parts of our portfolio. So you're getting a nice mix effect while we get the volume growth, the mix effect is also adding to that. And then the third component is really, we announced earlier this year, a transformation program that was really consistent with the spin. And when we spun, the way I would say it is we got pieces of a lot of things whether we wanted it or not. And so there was an opportunity to kind of clean up a little bit through a transformation program so that we were optimized for a semiconductor pure-play company. And we sized that at roughly $100 million of EBITDA benefit over the next couple of years. We're making great progress with that just to bring that to life a little bit. About 50% of that benefit is going to come through productivity efforts within our existing facilities. A lot of that may be things like putting in automation. We see lots of opportunities to do more with AI, automation and machine learning in our existing facilities. That will improve quality. It will also improve capacity, and it will be at lower cost. So the combination of more capacity with better quality at lower cost is a really very high return initiative. So we've kind of doubled down on that as part of this transformation initiative. The other 50% is really through a combination of commercial and innovation excellence, where the goal is how do we accelerate the product, the innovation engine so that we can go after more opportunities. We're seeing a significant increase in the number of opportunities to engage with customers and OEMs. We want to increase our ability to respond more effectively without having to add incrementally a lot more cost. How do we make the scientists and engineers that we have as efficient and effective as possible? Some of that may be through molecular modeling, simulation and machine learning technology applied in the R&D construct. Some of that may be engaging earlier with customers in the design process. So the earlier we engage with them, the more efficient the R&D development can be. And then the last theory is really on the local-for-local model and optimizing that. The example that we use is when we spun -- we have a very low percentage of our business that is in Europe. But as part of the separation, we got way too many distribution points in Europe than we can -- than we know what to do with. So part of that is maybe simply consolidating some of that footprint within the distribution network to something that's more fit for purpose, more intentional. Well, again, it will get -- customer service will go up, costs will go down. And it will be just more -- a much more efficient operating model. So when you take that and you work that through the entirety of the system, we expect -- we'll see some of that benefit even in the back half of this year. It's baked into our guidance for 2026 that we talked about at the last earnings call, but the bulk of that will really start to come in, in the back half of 2027 as we get those investments up and qualified and they start to flow through.

Is Ken on board with all those projects?

Absolutely. He's excited. He's actively led these kind of large-scale transformation initiatives. And -- and I'm pleased that Mike, who's been kind of in the interim seat, is also heading up our transformation office as well. So it will be a natural flow in and transition for him to be able to continue leading those to make sure that we don't lose any momentum with that transformation opportunity. So those 3 combined, the transformation, the mix benefit and the volume benefit all combine to give us kind of a pathway towards nice margin expansion kind of at the gross margin level, I would say, puts us into -- were kind of sitting in the high 40s today. It gives us an opportunity to maybe get at the company level, kind of up to or above into the low 50s, 50% or into the low 50s and nice margin progression over time on the EBITDA line as well.

And the last topic I want to touch on is capital allocation. And specifically, I want to talk about -- sorry to put you on the spot, the inorganic side of things. You've got a really nice portfolio, a lot of really great assets within it. You've talked about being open to doing maybe some bolt-on or tuck-in M&A. So I guess, where is your head? Where is Ken's head at inorganic opportunities that you guys are looking at?

Yes. I'm excited to allow Ken the opportunity to talk about this. I know he's passionate on this subject. He and I have had the opportunity to have many conversations about this over the last couple of months. But really, the way we think about M&A -- so our first capital allocation priority is really around organic reinvestment in capacity and in R&D to continue to drive the growth and the success of the company. But after that, this portfolio has taken shape over time through a combination of really smart acquisitions. And most of them have been the bolt-on, tuck-in variety versus the transformational, right? So we want to continue to be very open towards building out. We've got a nice pipeline and a disciplined process around kind of smaller bolt-ins or tuck-ins to continue to expand our portfolio of market-leading technologies. The areas we've talked about being of most interest are the ones that are also the fastest-growing: advanced packaging, thermal management. We've got a nice semi consumable business. It's more of an equipment consumable. We think there's lots of other areas within the equipment landscape that are kind of pseudo consumables that we think would be an interesting addition to the portfolio. Ultimately, those things will help us to kind of fill out the portfolio and continue to build and accelerate the growth trajectory and the momentum that we've already seen over the last 6 to 9 months.

Okay. Maybe last quick one on buybacks and dividends. How are you thinking about cash returns?

So we've got an active buyback program in place, a $500 million authorization. It's fairly open-ended. The way that we're thinking about that right now is primarily to offset normal annual dilution. And obviously, we're really happy with the cash flow potential of this business. It generates a lot of cash, kind of normally would be in that mid-teens percent of sales kind of cash flow potential. The balance sheet has plenty of flexibility. We finished the second quarter at kind of 2x net leverage, so some balance sheet flexibility of firepower to do deals that we wanted. We have a modest dividend in place. We've got kind of the share back program. So shareholder remuneration is out there. We've got flexibility there. But I think our priority and our focus is going to be how do we continue to accelerate growth.

Perfect. Well, we're excited to watch that continue to accelerate. Thank you so much, Jon, for joining us.

Excellent. Thank you, Melissa, and thank you all for being here today. I know it's been a full day for everybody.

All right. Thanks, everybody.