GLOBALFOUNDRIES Inc. (GFS) Earnings Call Transcript & Summary

Hi, everybody. Welcome back. I'm Joe Moore, Morgan Stanley Semiconductor. Happy to have with us the CEO of GlobalFoundries. So we'll talk through it. Just quick on the disclosure side. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley rep. I apologize for having to say that so many times. So GlobalFoundries, I guess, 1.5 years since the IPO-ish, company has done really, really well and I think has really resonated with a lot of the themes that we're hearing in semiconductors now and the idea of focusing on the nodes that you do was probably maybe a controversial decision you made, but it really looks good in retrospect. So maybe we'll just kind of go right into some of the questions coming out on the near-term environment. But again, just really impressed with the progress thus far. Can you talk, Tom, about the General Motors agreement? Because I think that was actually something of a watershed for me where we're seeing something kind of unique. So can you describe the agreement there, what's happening and what it means for GlobalFoundries?

Well, first, I was disappointed you didn't read the whole safe harbors, and thanks for having me. You're right because of the uniqueness of what this deal does and how it intercepts where the automotive market is going. So we have a market that, well, unit growth will be relatively flat. We're talking double-digit semiconductor content growth. You have an industry that's coming out of a period where for the want of a $3 or $5 semiconductor, they weren't shipping a $50,000 car. So needless to say, these companies seeing how much technology is going to play in their future, being constrained in semiconductor, became what I call, semiconductor aware. And so what I see them doing and not speaking for their -- for them directly, is the -- first is to understand their semiconductor content where it's going and what I call harmonization or rationalization of their architecture. Some of them learned that they're using 75 or 80 different microcontroller partner with SKUs when they could have done it in 4 or 5. And so the first element for these car companies is, hey, we need to get our arms around the architecture of our car, harmonize it, rationalize it and make sure that we understand the different features we want to create the car and what's the best way to go do that from a semiconductor content. The second thing is the realization that -- while this market grows over the next 8 years, and for them, their need will grow, but the investment required to make sure that they're covered from a supply chain needs to happen. It's not -- the capacity doesn't already exist, so it's going to have to be put on. And it gives them an opportunity to decide where that capacity gets put on, how it gets put on with technology road maps or technology platforms they wanted on and how did they get at the best economics. And so this deal kind of covers all of that. It's a direct relationship between GM and GF for one part of the value chain and then a direct relationship with their design companies that will use the capacity they've created. So first and foremost, it's on a platform that they see a good 10 to 15 years horizon. For them, it was important that this capacity was created in the U.S. So it's technology and supply where they want it. But I think the most important part of the deal is this next one, the best economics. If capacity needs to be added, the inherent price to create or cost to create that wafer now goes up because you need to get a return on that new investment. It's not a fully depreciated. Now that investment and that return should only go to whoever is making that investment. If the foundry like GF is making it, that's the return, we're investing our capital, we should get a return on it. If an auto company decides they actually want to use their money to create that capacity, then they should get the return on that. The problem in a complex supply chain is if a foundry like GF makes that investment, creates a certain markup to get that return on it, passes that new cost through the supply chain, it doesn't go through its cost. It goes through as margin cost and gets marked up. And that cost at the end of the day, that incremental cost for that investment can mean the car company pays 20% to 25% more for that same partner. So how do you build the capacity. You want it with the best economics work directly with the foundry, secure that supply, create the economic model that eliminates that stack margin and then do business accordingly with a 3-way relationship.

Yes. I mean it's an impressive deal. And I think to me, I guess I look back and I would say you did a deal with Ford earlier last year as well. Did people at GM and Ford at the executive level know who GlobalFoundries was 2 years ago, like the awareness of the supply chains of the global interdependencies. You're probably the third or fourth person who said companies are doing a review and finding out, hey, we buy x amount of stuff from this company, we didn't even know. And so there's a lot more thinking about the supply chain in a way that should be really good for what you guys are trying to build.

Absolutely. I think back to the first half of 2021, I think I met either the Chief Procurement Officer, the Chief Operating Officer or the Chief -- CEO of every major car company because all of us were in a position where there was a lot more demand and we didn't even know where it was going ourselves, right, for the auto company. And if that goes [indiscernible] what I talked about, these companies had to quickly become, what I call, silicon aware. And not only is it to make the present launch of products they may go out but to comprehend what the silicon content is going to be when they go from an internal combustion engine to what we call ACE, autonomous, connected, electrification of a car and all the content that, that requires. And so we became very close to these companies. We participated in helping them understand where their supply chains or how we can help them. And the culmination of that relationship is some of the deals we're talking about today. And I think what we really like about the GM deal, it's a first of a kind because it's that one element where they're dealing directly with the foundry. But they could have supplied or locked up their supply chain given a long-term agreement with Tier 1 who then locks it with the fabless and then comes to a foundry, they chose to go right to where the source is. And so it's a first of a kind in that way, but I really believe it's going to become a model for many of a kind because everybody is going to want to have capacity to grow, the industry needs to grow, investment needs to be made, how do we do it in the most economic and efficient way. And it has to be these types of more complex but thoughtful relationships.

And then bringing that back to your business more tactically, I mean you've talked about more than doubling the automotive business this year. I know it's still relatively small relative to where you'd like it to be, but can you just talk generally about your automotive trajectory.

Yes. It's a little bit $100 million there [indiscernible]. We go back to 2020, our automotive business was under $100 million; last year, $375 million. For the longest time, last year, we were talking about a fourth quarter exit rate run rate of 2023 being $1 billion. Well, it looks like now we're going to be bumping ahead -- our heads on $1 billion this year compared to $375 million, and that will continue to grow. And I think that's a little bit back to the pivot of the company where we focused. A lot of the design wins that we're growing our revenue on this year, were design wins we won 2, 3 years ago, and we're continuing to win more content. We talked about other agreements, Joe. This isn't the first one. We announced an agreement with Bosch a little over a year ago to do their next-generation radar. Our precision battery management with other customers for electrification of cars. We do a lot of the content in cabin lighting. And we've done long-term relationships with those types of auto manufacturers. So lot of content. We're growing in a car, a lot of long-term agreements. This type of an agreement, I think, is going to be the way of the future.

Great. So maybe we could talk about some of the history that's led you here. When you took over as CEO, the company was trying to compete with TSMC at the cutting edge on single nanometer nodes. You were investing in a lot of different areas. You were buying [indiscernible] at one point. You really brought focus to the businesses where you can bring leadership and scale. Can you talk about what that's like from the standpoint of your customer relationships. You have a book of business, some of which was won because you were cheap 4 years ago. Now this real sense we got during the IPO process was people really want to do more with GlobalFoundries going forward. It's more of a partnership model.

Yes. And I think it ties back and maybe we'll touch on it later on a bit about the investment thesis you mentioned, our IPO a little over a year, close to 1.5 years ago. Look, we were a company that spent the first 9 years back in 2018, creating scale. We've spun out of AMD. You spin out a company that doesn't have scale, doesn't have scale by just spinning out. We bought Chartered Semiconductor. We acquired the assets of IBM. We made organic investments in Upstate New York. And then in 2018, we were a $6 billion, $6.5 billion company, but we were a company that was burning cash in a non-sustainable way. So the right strategy was to build scale, but how do you go take that scale and create into profitability. Over that same 9 years that GF was coming of age and building scale, the industry made a big shift. It made a shift from kind of a Moore's Law compute-centric industry to the pervasive deployment of semiconductors. And it turned out that the market dynamics were such that where we play 12-nanometer and above turned out to be 70% of the market opportunity. So when we looked at our company and took a step back, we're burning cash, it's not sustainable, 90% of our investments were going into "leading-edge" single-digit nanometer technology. But most of our revenue was in these other places, how do we go and pivot the company and focus our investments were targeted towards that pervasive deployment, 70% opportunity in the marketplace, $50 billion growing to $100 billion, be differentiated and create true partnerships with our customers. And it took 5 years. And this Saturday, I think is my 5-year [indiscernible]. And you first have to understand the end market to a degree where it's not just you want to be in automotive, you want to go down to the actual device type. If you want to be the very best in radar, you understand all the elements you'll need to provide to create solutions that have real value to your customers, value that they can capture in their products, value you can capture and you're selling. And so that doesn't happen overnight. And so what we really did was we focused the company on end market [indiscernible] technology-based, understanding their needs, take our $600 million a year in R&D and focus to win in those end markets where we have the biggest opportunity to create value for our customers, capture value for ourselves. And what you see, as we've evolved in that now, is the thesis that we played in a big SAM and we can compete and do well for ourselves, that SAM continues to grow. Our single-source business now is on a design win basis. Last year, 80% to 90% of our design wins were single source. Our revenue is over 2/3 single-source business on differentiation. And when customers are [indiscernible] single source, it requires a partnership. They've got to believe you're going to be able to deliver to them when they need it, you got to invest collectively to create that capacity and that you're not going to stumble in being a great supplier to them. And that journey takes time. If you go from a second source supplier to a primary source to a single source and that's by demonstrating your ability to execute, differentiate and win their trust.

And then maybe if we could bring that into the context of this near-term environment. So you have long-term agreements covering 90% of your revenue, I think, for the next few years. But you also have an environment where there are other people that want to work with you. You have an environment where some of those base businesses are struggling a little bit. You've seen weakness in some of your smart phone-related customers. How do you put all that together, how strict are you going to be in terms of [indiscernible] and does that open up opportunity because you're getting visibility on freeing up capacity that might be higher margin for you down the road?

So part of the remaking of GlobalFoundries was to not -- build the facility and hope people come, build capacity on a forecast. It was to use these partnerships and relationships with the customers to create for both of us durability, certainty and profitability. That collectively, we would pick market segments that differentiation mattered and that we can partner to create that technology, that there was certainty for them in supply and certainty for us in demand. And then that these solutions would provide the means for both of us to be profitable. And that was the framework we created these long-term agreements. Now they get really tested not when the industry is on fire. It's when it's soft like this. To date, I will tell you, they've done exactly everything we said they would during our roadshow and our presentations at the Capital Markets Day last August. They've created a framework where the same partnership relationship that entered into those agreements now get to sit down and figure out how do we deal with it? We could remix some of that capacity to areas where they have demand. And we've created a manufacturing capability that has that fungibility between corridors. We have the agility where if we're not building something on 40-nanometer, we can build something for 155. We've allowed them to extend the length of some of these contracts, spread some of that demand out in time. And then for some of them, it makes sense for them to use some of the contract and the cash they put upfront to offset underloading with an underutilization fee. But all of them in the context of that these contracts are the framework by which we work together and the economic intent for both parties always has high integrity. And to this date, you could see that's been working out that way.

Yes. I mean -- and then when General Motors asked for capacity in the U.S., you might not have had any [indiscernible].

Well and actually that's where we're adding capacity, yes, because that's where they would like it.

Yes. And I guess on that note, can you maybe talk to the CHIPS Act and other forms of government incentives around the world, how that plays into your strategy?

Yes. I think the whole idea of the governments around the world starting to make investments in this space is because they want to create global competitiveness so that businesses find it not a detractor to add capacity somewhere where they have in the past. In fact, what they're doing is creating the same types of situation that made Taiwan as a powerhouse in semiconductor, government involvement, industrial policy to create that industry that's so powerful in Taiwan. And so the premise of all of this is how do you create co-investments. I hate to call them incentives because incentives don't sound like business model, it sounds like free money. They're really co-investments because any government that invests in semiconductor manufacturing will get a return. Economic return in jobs, ecosystems, the amount of velocity of money that goes through an economy because manufacturing takes place there. And so what we look to do is create these investments that offset the extra cost to do work and to do manufacturing in some of these other regions of the world. We've leveraged that in building out our facility in Dresden. We have an arrangement with STMicro where we're going to help build on their campus to create scale and the French government is very -- and the European Union is very excited to bring GF into France. We're going to do that with the CHIPS Bill in the U.S. We'll talk about that in a moment. And we've done that in Singapore over the last 15 years, with partnership with EDB to go create capacity at the right economics. The CHIPS Bill for us is an opportunity to grow our U.S. footprint. And we see 3 areas of ability to do that. One is we have a 200-millimeter facility in Burlington, Vermont. We see that as a way to use that money to kind of modernize, bring in wide band gap technology capability beyond just CMOS. We're going to take a Phase 1 approach to our existing facility, Fab 8 in Upstate New York and use the floor space that needs tooling to go build out the rest of that capacity. In fact, that's where the first GF capacity will go. And then longer range, with the right customer committed demand and economics, go build a modular expansion at the site and double its capacity. That's the one thing about GF. For the better part of the first 10 years of our company, we broke our pick, building a global footprint, very expensive to go greenfield to create capacity. And once you have that footprint, you can add modular capacity, you get tremendous capital efficiency and economies of scale. And I think you're going to see in this industry, a lot of the greenfield build-outs are going to be painful because it's just the nature of the amount of capital needs to be deployed before you get your first dollar revenue up. That first wafer is really expensive [indiscernible] greenfield.

And I think the tax part of it is relatively clear from the CHIPS Act, but there's also this additional level of grants and partnerships that the government is forming. And it seems to me that there's a real opportunity in the nodes that you serve because the media focus is all on 2-nanometer and TSMC. But if you talk to the biggest customers of -- for those wafers, they've invested in Samsung. They have a multisource situation. They feel like they have geographic diversification. The people who are really worried about to have capacity in the U.S. from U.S. suppliers is your core customers. And I think -- so it seems like as the government does its research, I think they're going to hear a lot of people that are suggesting, hey, this is actually where we need to have domestic capacity.

Yes, there is a certain romanticism about so called leading edge and single-digit nanometer and it's a necessary but not sufficient part of the value chain and making sure that high-speed digital applications have domestic supplies is important, and that's happening. In fact, that's the first up in the CHIPS Bill funding. But the rest of us will get -- we need out of this to build what the rest of the supply chain needs, which is, what we call, feature-rich silicon. And one of the things about these grants, the capital intensity to add the types of capacity we do versus single-digit nanometer could be 1/3, 1/3 of the capital intensity. So you can understand why more of that money would go to a project that has higher capital intensity, but it doesn't mean money doesn't come to projects like ours. And then the last point is, you're right, there are 2 things. There's CHIPS funding, which is going to cover 10% to 15% of the project cost. And then there's the investment tax credit, which is 25%. And it's a combination of those 2 incentive programs or partnership investment programs, that come together to create the industrial policy that the U.S. is putting forward to create U.S. capacity.

Great. And last one question and then we can open it to the audience in case there's some. I guess one of the challenges that we had narratively around the IPO was today's cutting-edge become tomorrow's kind of feature-rich silicon, does it become more -- is there still growth not approaching the single-nanometer nodes. Can you just talk to some of the shifts? And it seems to me it's quite a bit different than it's been historically in this regard that the markets you serve are going to stay with you. But can you just talk about that .

So I'll answer that part, which is really the second part of your question. The first part was a little bit on the investment thesis for the company. I'll come back to that in a second. So it all comes down to what you believe and what's factual. The reason why the 70% of the market is not in single-digit nanometer today is it's all about what Moore's Law is and is not. Moore's Law was an economic model. I had 3 elements to it. Every generation of technology, the transistor costs would be less. Every generation technology, transition performance would be better. And every generation technology, the power per transistor would go down. #1 ended at 28-nanometer, I'd argue 14-nanometer is at parity. Every technology node beyond 14-nanometer transistor cost was up. So first and foremost, if you don't need that transistor performance, you're not going to spend more money for transistors. So many applications became not relevant in single-digit nanometer because the cost was too high. And that's not going to change. The performance per device diminishing return, you could still get some performance. But the real reason people still invest and then got the amount of money into single-digit nanometer is power per device. When you have a $100 billion to $200 billion transistor designed in a single chip and you're building up data centers, power is a big part of that equation. And so that's the value proposition. But it becomes very limited where that technology gets used. And that was part of the investment thesis for GF. Are you playing in markets that are here today and growing. Remember, we say is that we play in a total SAM, $50 billion last year, growing to $100 billion in 2030, big blue ocean for us. But the investment thesis was, you're playing in a SAM that's growing, you're playing with differentiated technology, single-source business. You have plenty of opportunity to pick and choose with your priority, market segments and technology that you want to apply to create and capture value. And that -- your long-term agreement with your customers give you that certainty, durability and profitability in your business. And that -- the fundamentals of the business would be [indiscernible] grow our business with the industry or a little bit better. But more importantly, we would grow our profitability multiples of that. And the last part of the thesis, and I'll tell you how we did the proof point was, hey, it would be easy to do it when it's an upmarket. How are you going to do it in an inventory correction when there's a cycle. So take a look at '22. We grew revenue 23% year-over-year, about $1.5 billion. Pick your favorite profitability metric. You like gross profit, you like net income, you like EBITDA. Every one of those $1.2 billion, 80% flow-through or multiples of our revenue growth. That's the delivery of the investment thesis. Here we are in an industry correction. We're talking about pricing holding strong, even going up as we remix our business. We're talking about our long-term agreements, having the ability to weather the storm with our customers and not being -- not seeing our business as impacted as the whole industry coming down. And so the test of our company is not what we did in 2022, it's what we'll continue to do in 2023 as we weather the storm until the eventual turnaround of our industry.

Great. Let me pause there if you have questions from the audience.

Thank you. Tom, you used the downturn to reduce costs. Can you talk about that a little bit and your confidence in hitting the target model that you outlined at the Analyst Meeting?

Yes. Thank you. Yes. So in our fourth quarter, we announced that we were -- we're not going to be smarter than anybody else in this industry. Everybody you've seen that -- we were going to go into a tough period of inventory correction, and we committed to take $200 million out of our structural costs for 2023. We entered the year, achieving about $115 million of that, and we're on target to take the rest of that out. It's a little bit interesting when you're in an industry that was as hot pace as we were in there were opportunities where or maybe we were spending to protect supply and we were able to cut back on that and then come out of this downturn with that fundamental $200 million of less operating costs as we turn on a company. Look, we're not going to be -- we have to be prepared for a second half recovery. We have to be equally prepared that this industry correction takes longer that we continue to perform. That's our intention, and that's our commitment.

Can you talk a little bit about your M&A strategy? And within the next 2 to 3 years, can we see you guys being aggressive on M&A? And is there certain end markets that you specifically want to target?

Yes. So we've been pretty [indiscernible]. We recently announced buying a resistive RAM technology team from Renesas. A couple of years back, we bought a design team in Bulgaria, very benign. We should never rule that out. But for us, it's about making sure we're creating real value for our business long term and not doing anything in a significant way to undermine our near-term financials. And sometimes it's a little bit hard to pick and choose that but we shouldn't rule any of that out. It has to make fundamental business sense and be consistent with who we are as a company.

Great. Well, actually, go ahead, Charlie. We've got to wrap up it.

Thanks, Joe. Tom [indiscernible]. So my question is about -- are you talking about that SAM is going to grow [indiscernible], but you're seeing that the China will be building a lot of capacity [indiscernible] So do you think that's going to [indiscernible] your market share in the long term?

Yes, that's the question on everybody's mind. How is the capacity being built out compared to the demand we see coming. And you have to first start with where we are today versus demand post this inventory correction and then what's coming on. As we see things, there's plenty of opportunity to us to grow our revenue consistent with customer needs and not chase putting capacity without, what we call, CCD, not charge [indiscernible] device, customer committed demand. And I think what we offer to the world is an alternative to Greater China capacity. And that's what we'll leverage our global footprint, and we will build that out on the differentiation that we create for our customers. And that's another thing, just because you add 28-nanometer capacity somewhere in the world or 40-nanometer, doesn't mean it comes with the special features that we had to create that differentiation for our customers.

Thanks. Tom, so we'll wrap it up there. Thank you so much.

Thank you, Joe. Thank you, everybody.