GlobalFoundries Inc. (GFS) Earnings Call Transcript & Summary

All right, everybody. Let's get started with the second presentation. I'm Ross Seymore, the U.S. semiconductor analyst here at Deutsche Bank. Our first presentation from CoreWeave was all about AI, and we're going to get into that well with our next company up on stage GlobalFoundries. But GlobalFoundries is especially well suited to the second most important theme in semis, which is geopolitics where semis have touched all kinds of different angles of tariffs and subsidies and grants and equity investments, all kinds of timely issues and opportunities. So we're very, very happy to have John Hollister here, the CFO of GlobalFoundries. So John, welcome to Dana Point.

Thank you, Ross. It's great to be here. Appreciate the invitation to come out.

So why don't we start off with a little bit of that geopolitical side of things and we'll start with the tariff side. GlobalFoundries is very unique in that it has a global footprint, but a substantial one on the manufacturing side here in the U.S. So talk a little bit about what tariffs mean to GlobalFoundries.

You bet, Ross. So for one thing, there's the input cost dynamic, which is fairly limited. We frame that as roughly a $20 million impact in the second half of this year, which is less than 1% of our cost of goods sold. The more important point is the point you're getting at, which is what does it mean around customers' interest in utilizing GlobalFoundries as a source of wafer fabrication. And this really plays into our strategy very well as a company. So our strategy is really built on 3 pillars. We provide differentiated technology solutions and essential nodes of semiconductor fabrication and that means 12 and 14-nanometer and above. We built very deep strategic partnerships with our customers and our ecosystem partners. And then we have a unique geographic footprint in the world where we are the only major foundry that has the kind of footprint we have with the presence in North America, in Europe and in Singapore and Southeast Asia. We have a new China-for-China partner where we can work locally to provide semiconductors to the Chinese OEMs. We'll talk more about that, I'm sure, in this conversation. But really, we provide a very good solution for customers who are interested in having robust supply of semiconductors outside of China and Taiwan and given the geopolitical changes and state of play that is offering us a lot of interest from customers. You see it. More and more press releases are coming out from the largest smartphone maker of the world, Continental, recently in the automotive side. So we're excited about that opportunity and working closely with our customers to continue to leverage that unique geographic footprint in the world.

Some of those press releases that you mentioned are interesting because they could be with other chip companies. So potentially direct customers, could be somebody who does a little bit of both, like you said, with the biggest handset vendor in the U.S. or somebody like Continental closer to the OEM side and then years ago, you even had them with automotive OEMs. Talk a little bit about what's bringing them to you. Is it just the fact that they need to build in the U.S.? Is it the unique processes? What's really the decision-making variable?

Yes. It's really a combination of those factors. Again, we can work very closely with our customers and be nimble and capable of fine-tuning what we do to specifically address what they want, what they need for their solutions. This allows us to provide a differentiated, more analog mixed signal centric approach to semiconductor fabrication that can be very much oriented to the customers' needs in areas like low-power and high-power delivery, RF performance and we also offer numerous fabrication techniques outside of CMOS in SiGe and in GaN and in silicon photonics that are highly tuned to exactly what the customer needs. That's one factor. And then we think about the possibility of tariffs, coming all the way back to your question, impacting where folks want to source semiconductors, we have the right footprint to drive that. We have a 300-millimeter fab in Malta New York. We also have a 200-millimeter fab in Burlington, Vermont and are in the process of adding more and more of our technologies into the Malta fab and in Burlington, but focusing on the 300-millimeter side with 22FDX, 40-nanometer, 55-nanometer technologies moving beyond the traditional Malta footprint, which was originally more oriented on the FinFET side. So this is garnering a lot of interest from customers who are looking at domestic sourcing as being a more important part of their manufacturing strategy going forward.

Some of those press releases with a variety of customers, have they lengthened the engagement duration that you have with them? A few years ago, when we had LTSAs all over the play shortages and semis, that was kind of -- that created duration that didn't end up being quite as long for some companies as we hoped and created a problem down the road. I get the sense these are a little bit different. So what sort of visibility improvement did these provide?

Yes, for sure, is the short answer to your question. In fact, in the second quarter, we entered into an extension and the lengthening of our relationship with the major customer in the smartphone space right along the lines that you're talking about. So yes, we're seeing that interest to develop and build, and it is leading to longer life cycles and customer relationships.

So why don't we pivot to another side of the geopolitical side, which is some of the CHIPS Act and government investments or those sorts of things. Of late, it's gotten a little more popular to talk about equity investments as a possible avenue. Does that apply at all to GlobalFoundries?

No. Our CHIPS Act -- chips funding framework is well intact. We are progressing toward our milestones. We're beginning to receive chips funding according to our milestone completion. This is a long-term program. We offered a refreshed announcement if you will, of $16 billion, which combines a number of prior announcements that we had and also expanded that to a certain extent of $16 billion in North America, that's over the next decade-plus period of time, Ross. So this is more of a long-term play that gets into, as I mentioned, offering greater technology diversity to the Malta fab and eventually expanding the footprint of the Malta fab. We have the ability to nearly double the footprint of our Malta fab and add that capacity and also expand and modernize our Burlington fab as well. So that's all part of our chips trading, but it does not involve equity funding.

And it's seems that we were talking a little bit earlier that it seems to be the one big beautiful bill going from the 25% to the 35% investment tax credit is the much more important dynamic of late.

Yes, that's right. Near term, that's a more substantial impact, and that's very helpful to us, both in terms of the CapEx portion of our ITC allowance and also the OpEx portion that goes into our SG&A spending in support of those CapEx investments.

So we talked a lot about the global footprint that you have in the U.S. centricity and how helpful that is. But you mentioned briefly about the China-for-China side of things. It's good to be on one side of this ledger, but you can't ignore the other. What are you guys doing in China?

Definitely. The first item to note is that we have existing business in China, with China automotive customers and other IoT customers, et cetera, and China smartphone customers. And we represent, once again, a unique opportunity for the Chinese customer base where we can provide both platforms. There are a number of domestic Chinese fabless semiconductor companies who are interested in having a global source for the global market and that's more existing. And I would say there, the interest is even growing in that cohort of customers, if you will, or those legs of their strategy where they're looking for a global source for their ex-China global business. More recently, we've established a framework with a Chinese foundry provider to provide a localized source within China for Chinese OEM customers who also want to have localized supply, similar to North American customers and European customers, et cetera. So that's now moving on to implementation and look forward to having that be additive to our overall business in China. And I'll just also quickly highlight, we will control the PDKs, the masks, the front end, the customer relationship. That's all under the GlobalFoundries, footprint and that relationship look forward to working closely with our supplier to provide that localized source of wafer fabrication.

Does the definition of a localized source to the Chinese government change over time? Is it something that's fluid? Or do you think what you're doing right now sufficiently checks that box for the foreseeable future?

I think it's the latter. I think we are providing what is sought for in that dynamic.

Got you. Why don't we pivot over a little bit to the supply side? You just mentioned about expansion and what you're doing in Malta. How do you see supply demand right now? It looks like we're coming out of a cyclical bottom. We'll get into some of the end markets later, but just from a pure supply point of view. How are you seeing things?

Yes. I think it's starting to get there. We had this enormous surge in the COVID era, and that led to some accumulation of inventories, particularly in the smartphone market. Also, the IoT market, in particular, had fairly high inventory levels. Some of that remains somewhat persistent, but we are starting to see those inventory levels come down and see more of a normalization there. Our factories are running in the low 80s of utilization on a global basis. So we have some room to go to further take down inventory and get the utilization factors up, but we're making progress.

On the pricing side of the equation, I know you did some strategic things in one of your end markets. But generally speaking, pricing has been pretty darn stable.

That's right. So you look at our the amount of sole-source business we have over the last 4 quarters, 90% of our design wins are sole sourced. That relates to all the factors of our strategy, but clearly looking at the differentiated technology leg of our strategy. It very much relates to that. So on a like-for-like basis, for the most part, we're seeing quite stable pricing.

And that's more from a cyclical perspective that I asked initially. From a longer-term perspective, is GF's strategy to have pricing be a tailwind, a headwind or more of a unit play and just remain neutral?

Yes, more neutral across. I mean there's exceptions. We can see allowances at certain times. Sometimes, we raise prices given on the situation, but overall, more just stable pricing.

So again, going back to the Malta example you gave of potentially increasing capacity over time. And you just said you're at about 80% utilized right now. When do you need to really put the pedal down and start investing on the CapEx side again?

That's right. So GlobalFoundries has been able to benefit from a significant amount of investment in the essential node capacity that we have globally, we invested about $7 billion over the past several years in our footprint, and that's allowed us to be very capital efficient. Last year and this year moving into the next year as we fill up that capacity and leverage our improvements in utilization to drive growth. I would say, Ross, as we get into mid-90s of utilization as when we really need to start thinking about adding on additional capacity. And we have plans in place for all 3 of our 300-millimeter fabs in Singapore, Dresden and in Malta. And in fact, we have the ability to expand our capacity within our 4 walls yet before we even need to get into adding on a new clean room space.

Well, that's exactly what I was going to ask is -- if I remember right, in the past, typical new fabs will be -- they'll have a $1 billion or $2 billion handle, at least from your CapEx point of view. Is that the sort of step-up we should expect or for at least the next year or 2, it would be more incremental because you do still have room within the existing 4 walls?

Yes, it's more of the latter. We're running in the neighborhood of 10% of our top line. I can see that coming up as we move into the second half of next year, but not super dramatically. We need to add equipment within the 4 walls and then ultimately get into layering on additional cleaning space.

Got you. Let's talk a little bit about the unique technologies that you have. And I guess, before we get into some of those, you mentioned the lithography, this is as small as you would go. Do you have a risk of kind of noting out as customers want to go to smaller nodes and you just really don't do that? And how do we think about the revenue headwinds that could create before we get into the good side of the revenue tailwinds that the technology would offer?

Sure. Yes. I mean let's back up and think about the service available market that we're addressing, which is roughly $80 billion. And we see that growing to $120 billion over the next several years. This remains a very robust market size, market growth opportunity for GlobalFoundries at the run rate we're at, we're high single-digit market within that market. So we have plenty of room to grow here, and it's a sustainable market that will exist for a very long time. And then we think about how we differentiate our technologies and offer new features to customers within that SAM that are otherwise not available to them. So thinking of both of those factors, I think we've got plenty of room to grow without their risk.

So let's get into some of the unique technologies, things like silicon photonics, some of the power side of things actually allow you to have an ancillary touch to the AI dynamic. So talk a little bit about how differentiated those are, how fast they're growing, the end markets that we can address those sorts of dynamics.

Definitely. So looking at the silicon photonics business -- let me look at the comms infrastructure data center end market, more broadly first. Within this, we have silicon photonics and satellite communications on the RF side are both performing very well this year. Silicon photonics is sure is on track to be a $200 million business for us, not a lot of communications is on track to be a $100 million business for us. And these are growing at very high rates. Mid- to high double-digit kind of growth rates year-on-year as we see in silicon photonics area, the pluggables market performing well, and this is more a rack-to-rack communication within data centers. We also have the opportunity over time here to also offer copackage optics, which will get into more scaling up in data centers and provide with much more concentration of optical connections within a rack. Over time, we also see the opportunity to offer advanced power delivery solutions for data center applications as well. So yes, these are very much tied to the AI market and have a growth corollary with the ongoing growth in AI.

You guys have given statistics about the percentage of your design wins that are sole sourced, usually 90% or something like that. So very, very high. What percentage of your revenue comes from those sole-sourced design wins?

Roughly 2/3.

Is that number going -- obviously, they're going to -- they're going to converge at some point.

That's right, fairly stable, and that should increase over time given the design win traction that we're seeing.

So let's get into one of the other technologies that you didn't mention, at least not directly, but the GaN side of things. And I think there's some interest right now, obviously, it addresses the -- can address the sweet spot in some of the AI side of things and even the core. But you've also had one of the bigger foundries, if not the biggest, get out of that business. Talk a little bit about why you're committed to the business and the opportunity it offers.

Sure. So we see gallium nitride, GaN, is offering advanced power delivery solutions for 100-volt and 650-volt applications in automotive and industrial, as you mentioned, in data center, and we're excited about that. We acquired a company last year, Tagore, to offer more of a system-level approach to how we're addressing the GaN power market, and this is under development and being productized now, and we look forward to the growth opportunities that, that's going to afford us.

A little bit of a segue or a little ancillary question. You mentioned about copackage optics, the packaging side of the equation. How does GlobalFoundries play on that side given that it's importance is rising across all these different form factors and applications?

Definitely. So we're developing advanced packaging technologies, both in our Malta fab as well as our Singapore fab and this can involve different forms of post wafer fabrication manufacturing services, whether that's die-to-die bonding or use of interposer layers, these types of technologies that are not necessarily what you would see in a traditional OSAT environment but our advanced wafer processing beyond the basic wafer fabrication. So this is important for some of these applications which you can actually have heterogeneous wafers being combined in these form factors to deliver the performance that's required and we foresee being able to offer that in our 2 advanced packaging sites.

So just as a wrap-up question on these different technology topics that we've discussed. Who do you see to be your primary competitor? You have the unique technologies. You only operate in a certain node window. You have the packaging we just talked about, the global footprint. So all kinds of different questions and answers on that depending upon what investors want to talk about when I have discussions about GlobalFoundries. So how do you guys competitive landscape?

Yes. I mean if you think about the top foundries in the world, it's TSMC, UMC, to a certain extent, Samsung. These are our traditional competitors, TowerJazz as well as out there. So -- but if you really think of once again about what is unique about us in terms of our technology differentiation, our strong customer partnerships and our unique geographical footprint, we really don't see any one of them that has exactly a precise mix of those factors with us.

A lot of times, I actually think if some of the fab light analog and mixed signal companies is both your customers and your competitors. How do you go about convincing them to do more with you instead of doing it more internally?

Yes. It's a good question. It really just depends on the technology. And I've seen these companies over 25 years have no problem finding areas where they can compete, but also areas where they can cooperate as long as it's good business, and we're driving the market forward that generally is not an issue.

So let's talk a little bit about the end markets that you serve. And just do the biggest one first on the smart mobile device side of things. I think it's about 40% of your sales or so. Can you just talk a little bit about some of the recent performance in that end market and then the longer-term growth rate do you think that deliver?

Definitely. So we're bullish in this market over time. It's been an area where there have been some inventory headwinds and also some consumer sentiment headwinds this year that we work through. But if you look at how we're diversifying our footprint here across haptics and audio and imaging, in addition to the already very strong position we have in the RF front end and in the wireless transceiver space, we have a strong position here and are bullish about our future here. I don't necessarily look at this one as being a very high growth rate for the company. I think that's present in all 3 of the other end markets. But this is a good stable piece of for us and we're bullish about our position going forward.

And you mentioned on your last earnings call about some price competition, some strategic decisions you made to either maintain or grow market share. Talk a little bit about what goes into that. That's the first time I can remember you guys calling out some like that in quite some time.

Yes. In some of these applications, we may have a dual source dynamic with the customer and have the opportunity proactively gain market share by offering some price concession to increase our share of wallet significantly that's the dynamic that we're talking about here. That helps our utilization, helps the customer, helps us longer time and is a win for us because it provides a greater longer-term revenue opportunity for GlobalFoundries.

Got you. And then automotive is one -- it's about 20% of your sales. That's been the other end of the spectrum as far as growth. That's been a home run for you guys. Somewhat ironic because the last year, 1.5 years has not been the greatest of times for most automotive semiconductor suppliers. Talk about what's allowed you guys to grow.

Yes, it's been phenomenal. I mean, if you go back 5 or 6 years, our automotive business was around $100 million as we've taken it from that to more than $1 billion and are anticipating mid-teens growth in automotive this year again. It's really started on the back of our 40-nanometer automotive-grade microcontroller, which is -- has been leading in the market and allowing us to gain significant share there. As we move forward, we've also begun to diversify into advanced radar solutions using our 22FDX platform and now also moving in to power delivery with some of the power technologies like 55-nanometer BCD. So it's really been growth in share and growth in silicon content in vehicles that have gone from, call it, $500 a vehicle to now $1,000 globally, and we see that continuing to expand as well. So while the SAR has been stable, as you mentioned, the silicon content continues to come up, and we continue to gain share. So it's kind of a multiplication of those factors.

And you have deals within this vertical with chip suppliers, with Tier 1 and then with the OEMs themselves. That's a little unique that you have the entire ecosystem. Let's just talk about the Continental one as one example of that. How does that create business for you? Do they get design wins and then bring in the chip guys? Which level in those -- that 3-tier cake, for lack of a better word, do you focus on, if not all of them?

Yes, we really do focus on all of them. And as we look at the Continental opportunity, they are beginning to form a semiconductor vertical within their platform, and we've been identified as the exclusive supplier for that. So we're very excited about that. That's a relationship that plays very well into our technical capabilities as well as our footprint. Going back to the recent phenomena in our industry as we faced shortages some years ago, companies began to really understand, we need robust semiconductor supply. It's strategically important for even national security and economic security. So we need to have robust frameworks around the world to drive our semiconductor supply. This is an example of that.

Perfect. Two more end markets we'll hit on quickly then. The home and IoT side of things a little bit more of an inventory digestion still to go there? Are we nearing the end?

I think we're near the end. We do have some more inventory digestion to go on to happen there. Good news there is on design wins. We see very good design win traction in IoT. Looking at the company overall, we had 200 design wins in the second quarter, which is nearly double the number of design wins we achieved in the same period of the prior year. And if you look at what's driving IoT, it's around some of these technical factors again, of low-power superior RF performance. We've seen success in design wins with our 22FDX platform as well as our 12LP+ platform for WiFi applications. This is an area of diverse -- diversity in terms of the end markets that we can achieve and the overriding phenomenon, ultimately of this will be Edge AI driving proliferation of AI capability, not from the core out to the edge as processing capability continues to pick up.

And should we -- we or investors think of this market as mainly cyclical? Or do you think there's a strong secular dynamic underneath the --

I believe there's just very strong secular with this market as you see more and more wirelessly connected, even battery-operated applications throughout the economy, proliferating over time.

And then the last one and definitely not the least, only about 10% of sales, but growing really fast is the comm infrastructure and data center segment. This is where silicon photonics fits. It doubled or nearly doubled year-over-year this year versus last. Talk a little bit about where you're exposed there. I know we talked a little bit earlier about it, but summarizing it into this end market, what's the growth driver there? It doesn't seem like there's any cyclicality per se right now.

Yes. No, I don't think so. So this is an end market where the fastest-growing pieces of this are now approaching roughly half of the total end market exposure that we have, and I'm expecting that to continue to develop over the coming years here. So you've got more and more need for communications and data centers. And what can be achieved using copper connectivity is reaching its limit. Companies need to shift to optical communication that brings in silicon photonics, and we have a leading solution initially for the pluggables market, and over time, moving into copackage optics, as we talked about earlier. And that's not the only application. We also have satellite communications with our unique capabilities, providing communication inside satellites themselves and also in the ground terminals that are the receivers for satellite communications, and that's with multiple leading satellite providers in the world.

So wrapping up the end market discussion, you guys guided what you did in the third quarter. Some areas were good. Some areas had some inventory to burn, but you still were pretty confident in your full year goal, which seems to imply a pretty good fourth quarter. What's going on in the fourth quarter? I know you didn't guide overtly for the fourth quarter, but with your full year guidance, what's going on in 4Q versus 3Q that gives you that optimism?

Yes. We see continued momentum. And importantly, we do have a one particular example of a customer repositioning some of their deliveries from Q3 to Q4. We spoke about that on the earnings call. So that's out there for a pickup in the fourth quarter. And we see the ability in the fourth quarter to expand our gross margins as we look at both the mix improving, depreciation continuing to roll off, some pickup in utilization and also continued management of costs for the company.

That's a perfect segue into gross margins, which is going to be the next thing I was going to ask about anyway. You guys did a great job, the LTSAs, which some people think that's a bad 4-letter word. Actually did exactly what they were supposed to do for you guys. Year-over-year, that might look like there's some difficult comps right now, but you've held in pretty well even excluding those from last year. Talk about the march from where you are now in the mid- to upper 20s to 30 and then eventually getting to 40. What does it take to get to your target?

Yes, definitely, Ross. It really takes a continuation of the factors that we spoke about picking up our factory utilization, enriching our mix, having a depreciation platform that's stable and even rolling off a bit more as we head into 2026. Those are the main drivers improvement in gross margin to 30% ultimately to our goal of 40%.

What -- is pricing still -- that's just kind of a neutral in your equation? Or as you move from the 2/3 of revenues being sole-sourced to closer to the 90% being sole-sourced where your design wins are, does that have a natural uplift for you?

Yes. So that's a good question. And I would say like-for-like pricing, yes, is more of a neutral effect. But as we can leverage more greater profitable product lines and applications. This can drive mix improvement in the overall dynamic of pricing and more importantly, gross margin. We focus a lot on price, but the ultimate test is the gross margin delivery. And we do see the mix getting richer in terms of gross margin delivery over time now.

How much variability -- you guys have done a great job of holding your price pretty steady. But in the next upturn when utilization does go from, say, 80% to 90% or 95% and things get a little tighter. How much variability is there on the pricing side? Or do you try to keep that relatively stable go for market share, create longer-term relationships, those sorts of things?

Yes. I mean it really just depends on the -- but it is a competitive market. We have to be responsive on price. We do have the ability to raise price in some cases, but also want to deliver to our customers' needs on where they need to be economically. So it just depends on the situation. Overall, I'd say more stable pricing is the right expectation.

Let's wrap up in the last few minutes we have here talking about cash, what it all kind of comes down to. We talked a little bit about CapEx earlier, but we'll put a little finer point on it here. You talked about -- I think this year's guidance is about $700 million in CapEx or so or capacity capital intensity. How do you think about that going forward? Do we go up back up to the 15% to 20% range? Can you stay in kind of the low double digits? Is it really dependent upon how much demand booms anytime?

Yes, it really is. As we talked about earlier, I think staying roughly where we are for the next year or so is a reasonable expectation. I think as we continue to see top line growth, the utilization pick up, see the need for more CapEx out there to expand our capacity. We -- I can't see that moving into the mid-teens to 20%. Long-term model would be 20% and 20% free cash flow. So that's -- those are synergistic with each other as we continue to grow the top line and throw off cash.

I mean you guys run out of the internal capacity within the walls that you already have, how do we think about the stairstep up? Let's talk in dollars, not percentage of revenue.

Sure. Yes. I mean you can see going from $700 million to $1 billion and beyond of CapEx to drive that. And we'll be thoughtful about it, Ross, and think about how to phase that over and as we move forward. But once again, we have robust support from governments around the world, helping us with these investments in this critically important strategic technology that we offer. So that helps a lot.

And you guys are throwing up, I think, still $1 billion in free cash flow this year, adjusted free cash flow. It seems like at this point in the cycle, that's a pretty good base to be able to look at. So hopefully, that can grow going forward. How do you think about capital allocation? And we can get it a little into your MIPS acquisition, if you want to talk about M&A.

Yes. Sure. Yes. So I think we will continue to look for opportunity like that to add interesting technology and businesses that can help us further differentiate what we offer. MIPS is a great example of that. We will continue to evaluate opportunities there. And over time, we remain a fairly recent public company, and we're continuing to form our thoughts on whether it's dividends or share repurchases, more ahead for that. But all 3 of those pillars are potential applications of capital allocation for us.

And there's no limitations on that per CHIPS Act or any other government incentives that you have in any of the various regions?

Nothing that we can't manage.

Got you. So what's the -- wrapping up then, cyclically, would you agree we're kind of coming off the bottom across the majority of your more cyclical segments and now the secular is starting to shine through?

Yes, I would. I think that's a fair assessment. As we look ahead, we'll have to be mindful of the macroeconomic conditions and what's happening more broadly, but it does look like we're starting to clear out the inventory generally speaking.

And then secularly, it seems like post the pandemic get rid of as but have significantly longer relationships that you've signed with all these press release that we have. The secular outlook for the company seems as positive as ever.

Yes, I believe so. I think we've got a good stable position in the smart mobile devices market, where we have growing diversity in what we're offering. And strong growth expectation in the other 3 end markets that we're serving for the reasons that we talked about.

Why don't we wrap up with one question, and I'll give credit to my predecessor, Brad, who asked this, what's -- but I'll ask it in 2 ways. What's the thing that you're most excited about that we haven't discussed here today or investors don't appreciate? And then also what keeps you up at night?

Yes. I think I'm very excited about our strategy. I think it's coming together. We've got the 3 pillars of the strategy that I talked about earlier. I think that's really come into fruition now. It is exciting and offers us a great platform to grow from for what we do. We look at the policy angle and what's going on in the world with respect to broad macroeconomic conditions related to that. We just have to be mindful of that and be reactive and responsive to that.

Perfect. Well, we are pretty much right on time, John. So thank you so much for joining us here.

All righty. Appreciate it very much.