Advanced Energy Industries, Inc.

Perfect. Go ahead and get started. I'm Joe Petrocchi, the semi-cap analyst here at Wells Fargo. Excited to have the Advanced Energy team here, Steve Kelly, President and CEO, as well as Paul Oldham, the EVP and CFO. So thanks for joining us, guys. Good to be here. Good to be here. You know, maybe first just Oh, I think, actually, Paul, you want to go through Reg FD?

Yeah, just a reminder, everyone, that we may have some forward-looking statements today. Those are subject to a number of risks. You'll find those risks detailed in our SEC filings. Also note that we had our last earnings call on October 30th, which isn't that long ago, so we won't be providing any updates to guidance today.

Perfect. Okay. Maybe just to start, A little bit of news this week, right, with the China export restrictions. Is there anything you can kind of provide or help us just kind of any guide points there just as we're thinking about all the different ramifications?

Yeah, the export control regulations that were issued yesterday should have a minimal impact on our revenues. More significant impact happened back in October of 22. And so we took our hit then. but we don't expect much of a issue this time around. Most of our exposure to China right now in semiconductor is indirect to our major OEM customers.

Okay. Okay. That's hopeful. Maybe shift gears a little bit. So, you know, maybe first, can you run through, you know, the 3Q earnings, kind of where we are? You know, cycles are maybe a little bit different across the different end markets, but maybe let's start there.

Yeah, so I'll cover the top line, and Paul will cover the bottom line. But in Q3, it was a solid quarter for us. We beat midpoint guidance on revenue and EPS. We saw strength in semiconductor, and we also saw a lot of strength in data center. And so that helped us achieve our targets in Q3.

Yeah, I think as important as having a solid top line is we improved gross margins by 100 basis points sequentially. A lot of that came from the initial results of the factory consolidations and expense cutting that we're doing. It's good to see that because we've been working on this for a while and frankly the cost to do the transitions was equal or more than the actual savings. So we started to see that tip. We continue to expect to see some benefits there and we saw good leverage on the operating line too. So we're up about 120 basis points sequentially on operating income and as Steve said we're able to beat the earnings guidance.

Okay. And then I guess as we think about 4Q and we look into 2025, I mean, how do you – can you kind of walk through at least how you're thinking about some of the different cycles that are playing out right now? I know Semi, we're still kind of trying to see a little bit of recovery there.

Yeah. Yeah, so we guided to a higher Q4 sequentially, and we think Semiconductor will improve again over our Q3 revenue. And our Q3 revenue in semiconductor was our best quarter since Q4 of 2022. So it's notable. We also think data center will continue to be very strong. And that's going to move into 2025 as well, that strength. We talked about industrial medical on our call. And we think we're close to a bottom on industrial medical. We started correcting there in Q4 of 2023. So we're in the fifth quarter of correction. And our analysis of the distributor data shows that we're pretty close to going from a negative slope back to a positive slope on revenue in industrial medical.

Okay. You know, one of the things, more news, right, that's kind of been happening over the last several weeks, I mean, you know, President, incoming President Trump has talked about, you know, putting tariffs on Mexico, Canada, China. You guys have a fairly large, you know, manufacturing footprint in Mexico. Is there anything we should think about there in terms of is there a potential impact and margins or just, you know, revenue? I mean, walk us through, like, how to think about that, at least from what we know, right?

Yeah, maybe I'll just take a step back and go through our manufacturing footprint. Perfect. So most of what we make today is built in two countries, Malaysia and the Philippines. We also have a factory in Mexico. So the Mexico factory, if you look at that, most of the product we build there does not go into the U.S. It goes into other countries. And for the product that comes to the U.S., we have the ability to build that product in other factories in Southeast Asia. So I don't see Mexico as a huge issue. The other issue is China. And three years ago, we had three factories in China. Today we have one. And that one factory will close by Q2 of next year. So that closure process is underway. The last part of our strategy is to build a new factory in Thailand. So that shell will be complete March of next year. And I see that more as a 2026 opportunity for us. But again, I think Thailand is also a safe country when it comes to tariffs.

That's helpful. I guess unless it's every single country.

Then we're all in trouble.

We'll see what happens. You guys just had an analyst day not that long ago, a few weeks ago. The event you guys outlined new revenue targets, new EPS targets. I think you basically outlined the opportunity to double revenue in about six years. targeting like $3 billion in 2030. Talk about the confidence in being able to lay out those targets, the underlying drivers of that. Maybe let's walk through that.

Yeah, yeah. So I would say first our confidence is pretty high in that number. Of the doubling in sales, roughly a third of that will come from M&A. And so as we look at the organic growth and we march to the markets, we start with semiconductor. And what encourages us there is the tremendous uptake of our new technology by customers. So the customer pool has been tremendous since we launched these Everest and EVOS products in July of 2023. And we think we'll start to see that ramp as soon as the second half of next year. And we think that gives us really solid runway for share gain in semiconductor industry. If you take a look at the advanced processes and you add our share gain with increased capital intensity for etch and deposition processes, I think it allows us to outgrow WFE by a factor of 30%. So we said 1.3 times WFE is our expectation. Data center. Data center obviously is very strong right now, and we've been able to leverage our our feature set, which is particularly applicable for AI data centers. So we've typically designed our products to maximize reliability, to maximize the power density, and to maximize power efficiency. And as you spend more for a data center, and you consume more power, those three features become much more important. So I think we're well positioned. And what we said in the analyst day was we could add another $100 million annually to that business. Our focus on data center is not market share growth. It's to maintain solid margins, to focus on high-end opportunities. And if growth happens naturally, then we're happy with it. The third market is industrial medical. Now, industrial medical is very different from the first two markets I mentioned because it's a market where you have thousands of customers. And it's analogized into baseball. It's a lot of singles and doubles. whereas the semiconductor and data center is a lot of home runs. And so in industrial medical, we started this refocus effort about three years ago, where we moved more engineers into industrial medical so that we can increase our new product output. We took our sales force and basically split them 50-50. Half of the sales and FAE force call on semiconductor and data center and enterprise customers. The other half are focused on industrial medical. That's created a lot more opportunity. And the third big step we took was to invest in our new website. And so now customers can figure out what we make very easily. And if they like it, they could order right from our website or from one of our distributor partners. And so that's created a tremendous pipeline of opportunities for us as well. So we manage that much differently than we do the big accounts. That's really important. The last part is telecom and networking, and what we said was we don't expect too much from telecom and networking. We have some decent design wins there, but we don't expect much growth in that market.

Okay. Maybe let's kind of back up and we'll go through some of the more pointed questions within those end markets. You know, you talked about EVOS and Everest. talk about maybe can you just kind of explain maybe for those that are less kind of in the weeds like what are those what do those products do in the applications and then why are you winning against your competition yeah yeah so so the need for those products is is because the the processes are becoming um very complex and as you get to two nanometer and below

the current techniques for delivering power into plasma chambers are not good enough, essentially. And so if you're operating a fab, usually the two most important things are throughput, how many wafers you can get per hour, and yield. Those are two big parameters for fab operators. And what EVOS and Everest offer is a technology which allows the fab operators to to maintain or improve throughput and yield even below 2 nanometer. And so that's why there's such a customer pull, because the customers need something different to achieve that need of their end customers.

And I guess you talked about some of the targets in dielectric edge. Can you share, like, some of the win rates that you're seeing against, like, existing vendors that, I mean, you know, I think a market that you're kind of trying to gain some share, right?

Yeah, we haven't been specific about wins and dielectric. What we have said is that, you know, multiple customers are evaluating our technology for conductor etch, dielectric etch, and deposition applications. So we think we'll be in a position to make more definitive statements next year. But we're very encouraged with the uptake. If I haven't mentioned it, we're going to ship 250 units of Everest, EVOS, and NAVX combined. And those are units that we're selling to the customers for evaluation. And many of those units are in end customer FAPs today for evaluation.

OK. That's helpful. Maybe shift gears a little bit here. Can we talk about the data center opportunity? I think that's been kind of a big surprise this year that it's grown very nicely from AI. So talk about just the position there. Why is this all of a sudden driving growth and been a market that we kind of previously for you guys was maybe a little bit kind of an afterthought for investors and now it's important.

Yeah, when I joined the company in early 2021, I took a look at some of these data center opportunities, and they didn't come with very attractive gross margins. And so we changed the focus of our data center engineering team away from second and third source opportunities to going after sole source opportunities, solving tough technical problems where the customer is willing to pay attention pay us for that solution. And we were successful. And so that's what's driving our business today. In addition to the rise of the AI data center, our solutions are becoming more valued by our customers for the reasons I mentioned earlier. So I think our design efforts paid off. And we have a business now which I think is making very good margin. It doesn't average up the company margin, but it's a lot closer than it used to be, and we're pretty happy with the business.

Okay. You know, some of the kind of secular trends, right, like GPU power requirements obviously increasing significantly. You know, I think back at the InnoStay a few weeks ago, you talked about, you know, currently we're at like 100 kilowatts, but where does that go? I think you talked about like a 10x increase. Yeah. Yeah.

Yeah, when we talk to our customers, you know, we think it could get to one megawatt. That's what we talked about at the analyst day. So it really changes the complexion of our solutions. A lot of technical challenges given the limited amount of space we have to work with in a data center to deliver power. The need to make sure it's a highly reliable solution and at the same time efficient. So there's a lot of good challenges for us and what we see is the development cadence in the data center has really sped up. And it's all tied to the GPUs that are being launched by NVIDIA. And so they've been launching these GPUs at a fairly rapid rate, and every time they launch a new one, the power requirements change, and our customers have to make adjustments. So what it requires is for our customer to work closely with us earlier in the cycle, so that we have the power solution ready and they don't have to wait for it. So it's actually good news for us. It draws us closer to the customer and allows us to work on solutions, not just M plus one, but also M plus two.

That's helpful. I mean, your visibility in that market, I think in the past it's been something you've talked about. I want to say you said maybe six quarters, seven quarters a cycle kind of is a fair way to think about it. But maybe this is a little bit different.

Yeah, you know, I think if you look historically, typically this industry will go through four to six quarters of high demand, followed by a few quarters of digestion. It just keeps going back and forth. You know, there's a case to be made that with AI, with that build-out, with the shortage of GPUs, that that cycle could be elongated. So it might not be four to six quarters, it might be longer. That's what we hear. But we don't know for certain, right? But we think with the current dynamics, it's certainly a possibility.

Okay. And I think, like, you know, there's also kind of a dynamic of maybe, like, a content store in terms of, like, customers transitioning to, like, 48-volt. And, like, can you walk us through, like, just that dynamic as well? Because I think that can be positive for you guys. You had talked about that for a long time and how it seems like we're finally... seeing that?

Yeah, I have some customers who've moved to 48 volts, and for the others, there's a roadmap, a near-term roadmap to 48 volts, so I think it plays to our strengths. It makes a lot of sense, you know, as far as data setter economics go, so it's logical. And, you know, I think we're just in a position now where we're closer to the customer than we've ever been before, and we're working on the right things that will generate not just revenue, but also profits for the company.

Okay. In those kind of like being a little bit maybe more drawn into the design and optimization kind of R&D discussions, like is there room for margin improvement in that business or is it just kind of... I think we've seen margin improvement, substantial margin improvement over the last three years.

I don't think we could squeeze out too much more.

Okay.

But I think there's always improvement, particularly with the bill of materials.

Yeah.

These... Machines are very complex, use a lot of ICs, a lot of MOSFETs and so forth. So we continue to work with our suppliers, you know, to reduce pricing over time.

Yeah, okay. Anything we should be thinking about in terms of like, you know, that could limit like the upside of that business in terms of like manufacturing or, you know, access to components? Like, is there anything that we should be thinking about from that front?

You know, I think right now the biggest constraint will be the supply of GPUs. Okay. Yeah. You know, I think we learned a few lessons during the COVID supply constraints about which suppliers were supporting us and which weren't. And so we tried to design out the suppliers that weren't supporting us. And where we couldn't do that, we put in place a strategic inventory. Yep. But certainly, as we move forward, New designs, we're trying to use only components from suppliers that we that support us for the COVID supply chain constraints.

Let's shift gears a little bit to industrial medical, you know, the very, very large kind of fragmented market. I mean, can you help us kind of understand just like what are the key applications or in markets within that that we should be thinking about that are the real growth drivers?

Yeah, you know, it's a lot of applications. I think some of the prominent applications would be horticulture. There's a lot of plant growing happening today around the world. Test and measurement is a big area for us. Robotics, industrial automation, EV charging, drone charging. I can go on and on. But the common thread is... They're all high-value applications, and they typically need something special. And so what we do in industrial medical is we come up with standard products. They're technology leaders. And some customers buy those standard products, but many want something different. And so we have a group of people who create what we call modified standards. They take the standard products, change the software, change the inputs, change the outputs, change whatever the customer needs changed. And that creates a unique product for the customer and a sole source product. And the nice thing about industrial medical is for most of these applications, they're very long lifecycle. They just keep buying the product for many years, often until we obsolete it. And so it requires an additional few steps to customize it, but it pays off over the long term for the company. OK.

You know, I think right now, right, the demand is kind of difficult in that market. I mean, just given inventory dynamics and things, like, can you talk about maybe, you know, your visibility? Like, how have you or have you changed, like, the way you think about, like, channel inventory management and things like that? Because I think the prior cycle, right, like, this is a much smaller business for you guys, given it's kind of, you know, gotten larger in part from M&A. So, like... What have you learned, I guess, from this cycle that maybe you can take forward?

Yeah. In the industrial medical business, roughly half our business goes through distributors. Yeah. And so that's actually pretty helpful because we get inventory and resale reports every month from these distributors. And we track them very closely. And what we've seen over the course of the last four quarters is the inventory has trended down pretty nicely over time. Resales dipped in Q1, but they recovered in Q2 and Q3, and we think Q4 is going to be equally strong. So the underlying demand's not bad, right? The inventory was too high because of the supply chain disruptions. And so what I said during the last earnings call is we think we get to a point in distribution either this quarter or next quarter where we've kind of hit the bottom. They need to start replenishing their stocks. And so we expect to see a turn back up in industrial medical in the first half of next year. And it won't be a violent turn up. It'll be a gradual turn up, we believe. But, you know, we'll see that market recover. And I think one of our advantages is we took care of our industrial medical delinquencies in Q3 of 23. So that's why we started correcting in Q4 of 23. Others that play in the market, it took them much longer to catch up to their delinquent backlog, and I think it will take them a little bit longer to recover as well.

Okay. You know, maybe go back to the analyst day. I think you guys talked about, you know, $125 million of incremental annual revenue in industrial medical. Talk about, like, the drivers of that. You know, there's specific, like, programs or new product introductions that we should track, like, in terms of trying to think about, like, the acceleration of growth.

Yeah, so what we've talked about is our design pipeline, and it's two to three acts what it was before. you know, two years ago. And so those design wins will turn into revenue, you know, at different times. But we just layer it. You know, we continue to win new designs every quarter. And over time, you're going to see that translate into growth. We think 2x the market growth in I&M. And I think it's reasonable, given the investments we've made in the channel, in our sales force, and the development teams. We also have a manufacturing infrastructure, which is quite good, and it's geared towards high-mix, low-volume, which is our business in I&M.

Okay, okay. Awesome. Maybe just quickly on telecom networking, we talked about it a little bit a few minutes ago, but what's the outlook in that space? How do you think about the plans for that space going forward? being part of the overall broad portfolio, like, just kind of help us there?

Yeah, I think telecom and networking is one of those markets that we're glad we're in it in the context that we have products that can serve it. We have some marquee customers that we'll want to continue to serve. But for the most part, we've done a lot of portfolio rationalization there of lower margin products. And so it's a market that we don't see growing a lot. We'll serve it because it's complimentary. and we expect it to sort of stay in this $20 million, $25 million per quarter run rate without a lot of growth going forward.

Okay. Maybe shifting to some of the targets from a margin perspective that you outlined at the end of the day, I think one of the big things, right, is you guys have been kind of on this path to try to get above 40% gross margin. You talked about at the end of the day getting to 43%. You know, 40% to 43%, depending on the timing and the cycle. Help us understand, what's underpinning that?

That's a good question, because it's been a goal we've had for a while, and I think we've had a number of headwinds as we went through COVID, supply chain crisis, et cetera. But we've kind of turned the corner, I think, on that. And there's really three things that drive the margin improvement, two of which we largely control and can drive. The first is our factory consolidations. We're essentially going from 15 factories to five. We're able to reduce a lot of overhead associated with that and end up with a fairly modern, scalable factories as we go forward. We've already begun to see some of that benefit. That was the primary thing that contributed to the margin improvement in Q3. And we'll expect to continue to see some improvement over the course of the next few quarters with our next big factory closure being our China factory, which happens in the middle of next year. So we should realize between 200 and 250 basis points in total for margin improvement from just taking cost out of the structure. That should be structural improvement that stays with us going forward. And largely, we should be seeing that by the second half of next year for most of that.

And the baseline of that 200 to 250 is?

The beginning of this year, which is sort of 35%. Yeah, right. That's exactly right. It's hard to measure every quarter of the exchange. We have to take a baseline to measure from. So that's where we measure from. The second thing is what we call product mix. That comes from two things. The first of those is we have some significant new products coming. We've talked about the new products in semiconductor, which adds significant value. We have new platforms in industrial and medical. And even in data center, our margins are better than they've historically been because of how we focus. That's one element of it. The second is, as we continue to focus on more sole source designs, then those tend to carry heavier margins with them. Obviously, semiconductor and INM both margin up the company. And so we'll see a better mix of those products. And we attribute another 200 to 300 points of margin improvement to those activities. So volumes just stayed where they were for the next two years. We think that's enough to get us over 40% gross margin because those are structural changes we're making in the business that are in our control. Obviously, we've got to deliver the products. They've got to take up in the market, but those are things we're doing. The last thing is just volume. We should get some leverage from volume as the business grows. The way we've characterized that is for every $50 million in quarterly revenue, it's about 100 basis points of improvement over the near term. And if you look at between now and our aspirational goal of $2.5 billion of organic growth, then that's between 100 and 350 basis points of margin improvement. I think if you put those together, you get somewhere between 42% and 45%, and we've used 43% as kind of a target. The importance about that is our ultimate goal is to stay above 40%, even in a down market. So getting to 43% allows us to absorb a fairly significant you know, revenue downturn and still stay above 40%. That's the goal, and we think it's a pretty clear path. Obviously, we have to demonstrate that. We think we've taken the first step to that this last quarter. We think we'll see sequential improvement in margins again in the fourth quarter, and we'll continue to stay on that journey until we get margins where we'd like them to be.

Okay. And I guess, like, as I think about, you know, one of the things that when you look at the different kind of scenarios you guys outlined right like the incremental gross margin is is like in the low 50s so like but like why what would be driving low 50s gross margins I don't think any of the components within you know the different markets right even semi are like above 50 percent margin so like how I guess

Walk me through, like, how do I... Yeah, the key thing there is there's a substantial fixed cost structure, right? And the first thing, as I said, is we're actually reducing that as step one. The efficiencies of our scalable factories will actually allow us to scale at a lower variable rate even than we have today. I think that's another contributor to it. And so the new products will come with lower material and other costs because of their better higher values. So I think when you put all that together, you do get incremental benefit as we grow margins.

Okay. Moving down the P&L a little bit, like OPEX plans, obviously you guys have been very focused on introducing new products. So how should we think about R&D and just the investments that you want to continue to make there?

Yeah, so the first thing is during this downturn, we didn't cut R&D. We've kept it pretty steady. We've been able to afford that investment because of our diversification strategy. But we also knew that this is a critical time for us and our customers to get this new technology in the market. So it's very important to set up the next five years by investing now. So we've kept that at a high rate. At the same time, we streamlined the company. The peak to last quarter, we reduced our SG&A by about 12%. So we have made some changes in the company to streamline some of our acquisitions and other things to reduce costs. And as we go forward, the way we think about it is, as a general model, we should be able to add cost back at about 50% of the growth rate. So if we grow revenues 10%, we only expect to grow expenses roughly 5%. Now, how can we do that? Because we still have to pay salary increases every year. We have to do a number of things. But the reason we can do that is because we have a lot of leverage in our model. Since we didn't cut R&D, we don't have to add a lot back. We send it at the full level. And because our new platforms are very modular in nature, we can make derivatives and changes to those much more efficiently than we've ever done in the past. So we get a lot of efficiency out of the R&D investment that we've had. So we'll grow at some, but we don't have to grow at nearly the rate of what we put into it because we should get good leverage off the investments we've made. And similarly, we have a global consolidated infrastructure that we can leverage as revenue grows without having to add a lot to it.

Okay, that's helpful. Can you remind us kind of the capital allocation strategy? I know you guys have a focus on M&A, so maybe touch on that.

Sure. Our primary use of capital, we believe, is to add the most shareholder value by growing the company. We've done that historically pretty well. We've done a number of acquisitions. All those acquisitions are in the company today. They've grown. They're contributing good value. And we think there's a lot of opportunities in the markets we serve, given how fragmented it is, to continue to do that. So number one will be continuing to grow the company. We'll continue to stick to our strategy. We're a disciplined acquirer from a financial perspective. So Our goal is to add value over the long run with those funds that we can reinvest into the company through acquiring other companies like AE that we can add scope and then leverage the scale of the company. We continue to have an opportunistic share repurchase program. We don't buy stock every quarter. We know that the stock price can be a little bit volatile, and so we take advantage of that in times that we perceive there's additional value and we buy back stock. In fact, we bought some stock back last quarter in the low 90s. We continue to take advantage of that opportunity. Our goal is to offset dilution over time, but we've actually brought the share count down a little bit, as it turns out. So we do have the ability to do that. And finally, we pay a small dividend. I think that's a demonstration of the fact that we are cash positive in all markets. I think we've, our analyst today, we talked about that we've been cash positive, operating cash flow positive in every quarter over a 10-year stretch. So that's through, I think, two or three market cycles and And so that really demonstrates that we have the ability to sustain cash flow.

Okay. You know, in terms of M&A, Ray, like you guys had tried to acquire XP Power earlier this year. That didn't work out. And then I think you did some, you know, you did a little bit of balance sheet called optimization, if you will. So, like, I guess, how do we think about, like, the M&A pipeline? You know, is this the time that you're thinking about, like, I guess, How do we think about the opportunities there, especially given that you did do some optimization of the balance sheet? Is that trying to tell us that maybe there's not as many deals? And then I think in the past, you talked about valuation as being kind of the gating factor of doing some deals. So where do things stand? Is in the weakening kind of industrial market or in demand, is there more opportunities, less opportunities?

I'll talk about that, and Paul will talk about the financing. Perfect. You know, I did hire a new M&A leader, Kathy McKinnon, earlier this year. She's done roughly $4 billion in deals in the industrial space, primarily with Amatek. And she's got a very disciplined approach to the pipeline. And so we have a number of potential targets in our pipeline, and we'll continue to pursue them, to develop relationships with them, and be ready to engage when they're ready to engage. But to echo what Paul said earlier, we're going to maintain financial discipline and make sure they're a strategic fit in the company.

Yeah, from a balance sheet perspective, what we did is last quarter is we paid down our term loan A, but we essentially took that capacity and rolled it and a little bit more into a revolver, a line of credit. under the same terms that we had our term loan A, which are quite attractive. They're sitting at SOFR plus 75 basis points. So essentially, we have about a billion dollars of, I'll say, dry powder capacity that we can access. But we don't have to carry it all on the balance sheet ourselves, which saves in interest costs. So we have all the capacity we've had. I think we're able to restructure it in a way that's favorable to us financially and gives us all of the flexibility and capacity we had previously. OK.

Perfect. Maybe just to round up the discussion, anything that we didn't talk about we should have, or when you think about the story broadly, where do you think maybe investors are underappreciating or that we need to do some more work on?

Yeah, I think it's basically the power of our business model. And again, we're targeting high-end markets that are growing rapidly over time. We have the broadest set of technologies. We have the most power engineers of any player in the space. And we have the ability to put these technologies together quickly. So we can get to market quickly, we can differentiate using this combination of technologies, and we can get paid for it. And so that's really our business models because, you know, we have the best technology across the broad spectrum of applications.

Perfect. Thanks, guys. Thanks, Lacho. Thanks, everyone.