TE Connectivity plc (TEL) Earnings Call Transcript & Summary

All right. Good afternoon, everyone. Day 2 Citi's Global TMT Conference. Asiya Merchant here. I lead tech hardware, tech supply chain researcher at Citi. Really pleased to have Terrence Curtin with us. He is CEO of TE Connectivity. There's a lot of questions here. I'm going to kick it off -- and please come in, there's a bunch of seats in the front here. I'm going to kick it off. It's a fireside. It's interactive. I'm going to leave a few minutes here for folks to ask some questions. Please do raise your hand. If you have interest in asking and we will bring the mic to you. So welcome. Thank you, Terrence.

All right. Let's just start with maybe just the most recent reported quarter. I know September is going to be ending year very shortly, too. But you guys just had a very, very strong fiscal 3Q, sales were up 14%, 12% organically year-on-year, you had record orders. Your EPS is growing much faster. Your margins are expanding. So -- and you are across a broad variety of end markets. It's not just AI, it's not just [indiscernible] but there's a bunch of end markets that you're [indiscernible]. As you think today and as -- with the quarter that's just been reported relative to kind of your through-cycle framework that you've shared 6% to 8%, where are some of the demand drivers that are sustainable as you kind of think about that demand?

Thank you for the question. And also thank you all for being here. And let me just -- I'll get to your question, I promise. But first off is, I think what's important is when you think about what TE does, we very much focus on where this data need connectivity as well as when you need compute and data, you also need power, and it's areas that I think drive the growth that I'll talk about. And one of the things that we've been very focused on TE is how do we continue to broaden our growth profile. It was a big part of what we teed up at Investor Day, and like you said, Asiya, there's a lot of it that I'm sure we'll have questions around AI and our momentum in our DDM business that has had tremendous growth. But I also think when you look at what we're experiencing right now is across our industrial businesses, anything that's CapEx exposed is really driving growth and faster than we would have even expected when we talked earlier in the year. So certainly, on the data side, what we see happening, and you've seen a lot of companies here today talk about everything that has to happen around AI and also the hardware around it, we're no different than that. Our DDN orders were up 70%. And while we grew $1 billion this year, that really sets up a very strong backlog for next year with the partnerships we have with our hyperscale customers. But it's just not what's happening in data in our Industrial segment, what we're seeing with our -- where we serve connections from an energy infrastructure, which are driven around power. We're seeing double-digit orders there, where you have a market that used to grow 1% or 2%, now is probably growing 3 to 4x what it used to, and we've really capitalized to make sure we can grow double digits in that business. And then you have those secular areas that we benefit from, like aerospace and defense, going to grow double digits due to everything happening there as well as a general industrial uptick that we've seen around the world. This is where we would play with factory automation where unique connectivity that brings that data out to get to the intelligence, be processed through the LLMs. So when you think about our Industrial segment, we're going to grow about $2.5 billion this year. About 80% of our growth is in that segment, and it's very broad. And then in the Transportation area, that's an area that we're sort of driving content growth in a flat market, and there are areas that are also around data and power, self-driving cars, software-defined vehicles, also an electric vehicle which in Asia is very strong and we're strong. So when you think about that framework, which we're growing about twice, we see that momentum going to continue, and it all starts around data and power. And certainly, the CapEx elements of where we play, we just see that motoring along and orders up very strong to record backlog as we enter into our '27 that there's no reason to say that's going to slow down as we go into '28 when you think about all that's happening on the data side and the power side.

Okay. And then on the -- if I may just as we -- before we get into each of the segments, EPS is growing faster than revenue. So what's driving that? I mean, you not only do you have top line growth, but you also have nice margin leverage. And what's the algo there? Is it pricing? Is your pricing is higher? Or is it other sorts of leverage that you're experiencing that's driving EPS to be faster?

Yes. On our strong growth this year, which will be mid-teens, probably 1 point or 2 is pricing where we're actually managing through inflationary effects where we're actually getting recovery across the portfolio. So pricing is really a cost recovery tool right now. The rest of it is around volume leverage on what we've done to make the portfolio healthy as well as our footprint healthy. So it's really good operating conversion. But the other thing that I just want to highlight is we're investing significantly back into the business, especially to those areas that have really strong growth. We're adding engineering in our DDM business. We've added factories in our DDM business. We're adding capacity, but we're driving that earnings leverage to really good conversion on the volume while also making investments for the future to make sure we capitalize on that growth ahead of us. So I feel the levers how the team is operating is very strong. And certainly, our cash element. We have a very strong cash generation model. We've done some acquisitions as well as returning capital also helps [ EPS ] line as well.

All right. Okay. Well, maybe we can just dig into each of the segments, right? I mean, clearly, I'll start with AI DDN because -- or DDN just because it's top of mind here, it's a technology conference. So DDN sales up in the [ 30s, ] orders obviously significantly higher. You guys did have a target $3 billion in cloud and AI that you kind of shared at your analyst event. You talked about that target continues to shift left, meaning it's being achieved much faster. So when you think about your size now of that DDN, AI DTN, cloud and AI -- sorry, cloud and AI segment and kind of the growth trajectory that one should expect, just given hyperscaler CapEx and that's just continued to rise.

Well, first off, being our orders this year with our DDN business, which is where that's housed, is up 70%. So I got a couple of great questions a day. Terrence, how do you grow 30-ish percent, but your orders are up 70%. Those orders are really for 2027. Right upon it. So there are programs that we've won, certainly, with our hyperscaler customers around the copper connectivity that we do for them to really get their infrastructure up on their GPU networks. And that's going to be -- sets up really strong growth for '27. And as you said, any target we gave out just 9 months ago related to AI, we've been ahead of that and that $3 billion shifted in and that's going to continue to shift to the left. But I feel very good about the momentum there and our position is very strong with our hyperscale customers.

And beyond 2027, does the orders look like we are seeing demand very, very strong. I mean I have some component players. We were talking about HDDs that are talking about demand further out, just given the supply tightness. So how does it filter to the connector side on your side? Like are you seeing orders maybe stretch out beyond just on fiscal year?

We're typically into '27. So we don't have that supply tightness that some of your other people that you cover do. Really, when we're looking at it, we'll probably start in '28 orders, coming into early into '27 when we will see it. But I think the real element that we get excited about, as you go into '28 is the engineering activity we have with our customers today are next-generation architecture. One of the things that's important is our customers are going to pick a semiconductor, they're going to pick a power supply, then you go work the connectivity, and what's really nice about where we lean in is with our engineers, with our customers is those next gens are being worked on already. So the orders we have today are for things that we have to take into production next year, the engagements that we have about what's the next-generation architecture, who's looking at 800-volt, who's looking at liquid cooling versus air cooling, where does optics come in, in the rack over time. All of those discussions are happening real time, what gets us excited is that the momentum is going to continue in...

Talking about optics, that's all a big debate, right? And I mean you guys have talked about it, you're like, it's not just copper or optics is going to be copper and optics that are going to continue. So just as you're sitting here, you're looking at all the workloads, you're looking at all the AI architectures. I mean, they are changing, right? At OCT Asia, there was a lot of chatter about what's changing, what's not changing. You have scale up, scale out, CPO, NPO. So when you look at your content opportunity in fiscal '26 increasing in fiscal '27, just help investors think about as you go into fiscal '28 and '29, what sort of -- what are you looking at from a content opportunity?

So a couple of things, first, and I appreciate you said it's going to be copper and optics, and it was great at the lunch today. The Marvell CEO actually said that as well. So you hear that a lot, and I do ask it, it isn't a versus because we're really looking at architecture being maximized. And copper is going to play a role. Anything in the power architecture is always going to be copper-based. And even if you go to a genetic where you bring CPUs in, you got much more CPU, that's going to be copper-based. Where we play, certainly, we play in to scale up. We don't play as much in the scale out. But in the scale up, we do look at where you'll have NPO and CPO coming in more at the switch level, is where you're going to first see it. There's a lot of scaling that has to happen there when we look at the architecture we're working on with our customers. And we have solutions that can actually play in that space that I'm pleased with our customer engagement on that would help enable that. And that's an increased TAM that we would have versus our copper portfolio. So we view copper is going to continue to grow over this period. Certainly, the architectures, as you said, they're very customized. They're changing a lot as people are trying to optimize different parts of the workload. Copper is going to be a key part of it. But what we get excited about is where we can introduce basically a fiber attach unit, which is essentially, we would view it as a fiber connector that would actually help enable CPO or NPO as that comes in.

Right. And there's just a lot of activity going on in that [ FAU ] side just to make it scale, right, because I put up to this point, even at a scale. So just as you think about that space evolving, what are some milestones to keep -- that investors should keep in mind to say, okay, now this is starting to gain momentum and can scale to the level -- obviously, it's not fiscal '27, but into '28, '29, but you kind of see some of those...

Certainly, it starts with our customers' architecture. They have to hit their intercept points. Secondly, our customer engagement. And where are they seeing? And certainly, that's where we're focused on right now. Our engineering engagement on that intercept point hitting engineering qualification. And then like everything in the optics space, has it scaled to the needs that the units that our AI customers want us to have in our hyperscale customers. So I think they're going to be the things we're looking at the road map. Certainly, we're investing ahead. We made a $200 million acquisition earlier this year to help buffer the portfolio there, and we'll keep investors apprised of our progress as we go through.

All right. And then you talked a little bit about power, 1/3 of the connect within the DDN AI space is power, right? And the rest is data connectivity. So obviously, there's a lot happening over there. You're talking about 400-, 800-volt architectures, like help us understand like as those transitions are happening, again, what's the content opportunity growth that you guys are seeing? What should we be looking out for to say, yes, this is happening at an inflection perhaps for TE?

The one thing you have to realize is not everybody is going to move to one voltage architecture. Go back to the customization. What's really great is as you bring in these sidecars, that brings in a whole different architecture where connectivity even becomes more powerful. It's also you're bringing in a lot of different voltages into the rack that actually have very different material science, also have different electromechanical interference elements to them. All of those are things that we excel in. . So as our customers are experimenting with that, that we have such a large portfolio of power products to begin with, it's very essential that as we move through that journey with our customers. Some will not go to 800-volt but that will create increased content just due to the connectivity in there. Also, as you move to optics, you're going to need more power in the rack. And I believe when you think through the AI discussion, a lot of our discussion with TE is around the data side, I actually think the discussion needs to be as much about the power side because all of those inflection points create engineering trade-offs that our engineers do great with our customers and they're figuring out the architectures and constraints they're working through. And what that will create is an increased inflection point on the content. It won't be cookie-cutter by rack, but it will be something that will increase revenue content for us as we move to that, I think.

Okay. And then just stepping away from it, like where is TE's moat when you think about all these changes, whether it's on the power side or on the optics coming into the rack, like -- where would you say is TE's moat? And why you think that content continues to stay with TE, potentially even expand while you guys are providing the solution?

I think when you look at connector manufacturers, there's moats, 4 pieces to our moat. The first one is we need to be engineering close with our customers. Connector business is not we design a connector and bring it to a customer and just pick it off the shelf. That's not what a connector manufacturer does. It is they're trying to work right at the design center with our customers, and it's a very engineering-intense business. And we're part of that architectural design. And it's why when you look at connector manufacturer, we all don't play in the same markets. Now building up engineering and all the design centers around the world are -- is hard, but significant investment in different markets, and you typically see us play in certain markets. The second thing is we need to be deep in the technology. Whether it's AI and you're dealing with 448 gig speeds of signal integrity or voltages going up to rates with 448, you need to actually understand that architecture to get in. And you can hold that true into an AI domain. You could say it could be utility infrastructure, where we're doing power connections or even into a car. The third element is you need to basically have the science expertise that comes with it. In some cases, we're dealing with space, we're dealing with different materials because we're in the different applications you would have on earth. And then the fourth one is we have to have the supply chain that executes at the rate at the supply chain of our customers. It isn't cookie cutter around where we make. We have to be very close to our customers. If you take us in AI, we've added 3 new facilities in Southeast Asia to really make sure we support our China plus 1 over the past 3 years. We're adding more right now. and it needs to be able to execute at the pace our customers want, and it's something that continually evolves. So the moat that you sit there, which each 1 by themselves may not say, "Well, is [indiscernible] special." When you put them all together, and the amount of SKUs that we cover for our customers from an execution perspective is very important from the design all the way through in an area where cycles just get tighter. And every 1 of our markets are innovation cycles are tighter than they need to be. And our customers expect it, and that's how we have.

Yes. That's execution definitely. Energy, the energy segment, that -- it's nice. You had an acquisition, Richards, and you've also added to that. But it did grow 33% organically, I think, in the last quarter, last fiscal. You have grid hardening there. You have aging utility infrastructure. Obviously, you have the data center power build-outs. So is this like how should investors think about it? Is this a mid-teens growth business that we should think about for energy? And -- yes.

I actually do think it is. And I think when you sit there, this is we're doing connections for energy utilities. And really what you have, and I mentioned it earlier, a market that used to be a 1% to 2% market that I would have described to investors 5, 10 years ago as a cash cow, and we would be using to fund other things is really a market that's probably growing 3x what it used to where we position ourselves both organically and inorganically is here in North America is our strongest position around grid hardening, it's also around making sure we get the power to the data centers. We benefit from that. About 30% of our business does come from that, which does drive the growth you've talked about as well as what's happening with undergrounding and just protecting the network where there's deferred maintenance. So it's an area where no different than the data side, the power side has very long tailwinds on it. I know it's probably not a discussion as much here today in a tech conference as elsewhere. But it's something where we built a $2 billion business for ramping up this year, that's going to be growing double digits. And I actually think there's a lot more opportunity because I think we're very unique, not only what we do with the rack, but also getting the power to the data center actually creates both growth vectors on the data and the power side.

And then you recently acquired Astrodyne, right? And then how does that accelerate sort of the momentum you're seeing in energy?

Actually, on the Astrodyne, while it does relate to the power infrastructure, it's not really in our energy business, it's really going into power and power filtering that actually goes into specialized equipment for semiconductor manufacturing goes into aerospace applications, and it's just another 1 of the things around the power theme we like. We've always done power filters to really make sure you get power quality going through. Astrodyne does some of that as well as niche power supplies that we brought in. but it plays right into the power theme, but it doesn't go to the utility side like our energy business. So it's more adjacent to it.

Okay. Same theme, but just a different end market. Okay, which brings me to ACL, which also did grow double digits in Q3. You said now factory automation possibly has hit an inflection point here? Yes. So what's been behind that? And when you just think about AI and as it's kind of impacting various industries, including perhaps manufacturing and -- how do you think about that demand there for factory automation in ACO?

What we get very excited about because I know when we talk about AI, we always talk about the data center, let's realize, when you're in a factory and you're trying to buy productivity, it's the data on the factory floor that makes that happen. So when you think about what we do here, this is where we're getting more devices connected, and that's actually bringing that data that the central AI can do, and we've seen the inflection point be all around the world. This is a space where general automation was slow for a couple of years post COVID. We've really seen every region in the world pick up. It is around the connectivity for us, it's very much a data play on the hardware that we support the major automation players of the world. And it's something back to the CapEx trend I talked earlier on, we're seeing that accelerate. And we get content on top of that due to what's happening with data and the connectivity solutions we provide our partners. So it's something that not only getting a content play, I would also say we're getting a cycle uplift back now after a couple of years at...

And then the acquisition that you talked about, just how does that -- the acquired assets from Astrodyne fill up that portfolio?

It rounds out our portfolio from a power filter globally. We have more of a European and Asia [indiscernible], they were in North America, so it helps transit out. And it also gives us an entry way more into semiconductor manufacturing equipment, where we were not as strong. They were strong there, and it's just something where we see the trends around data and what our semiconductor customers have to do from an equipment side. It's something that's going to give us a new market entry. . The business that's going to grow double digits. It's accretive to our earnings. Certainly, will fit into our cash model as we go forward.

Okay. Let me just ask the audience, any questions? Okay. Let's talk about automotive, transportation. I mean that's a big -- within transportation, automotive is one of the segments. That's a pretty meaningful portion of your portfolio and the unit production in autos is kind of more or less flattish, maybe slightly down. You're still talking about content growth outperformance relative to that. Any -- in any given quarter, though, sometimes it undershoot some of what your target model is. So just help us understand like when you talk about 4% to 6%, what gives you confidence that you should be able to grow at that rate organically?

Yes. I mean, you said it well. Production environment in automotive globally is down a little bit all around the world. but we're growing 4 to 6 points above that. That is how we think about it in TE, we are very global. So what we're very fortunate about is we're not how to wait in any region and our strength in Asia really has been helping offset some weakness in Western World in prior years. When you think about how we drive content performance greater than production, it does come back to data and power again. What used to be of the electrified vehicle was a bigger driver, has absolutely flipped over to the data in the vehicle. So things that are around autonomy in the vehicle, things that are software defined. They are things that are driving a bigger part of the 4 to 6 than we would have had 3 years ago, which we would have talked more electric vehicle. In addition, we continue to benefit from our strong position in Asia, which is still very high penetration of electrified. And then the last element is the other things that happened in the car architecture zonal architectures, you come into any comfort or safety features that bring electronics, which just comes into what you need when you have a connectivity supplier. So all of that creates the 4 to 6. And with some of the EV element behind us now, like North America was a tough EV environment this year. Actually, we feel very good about the 4 to 6. We've been hanging at the low end of that range, hopefully being able to move it up here as we move forward in a flattish auto environment.

All right. Defense. I think we heard from [indiscernible].

Very strong.

Yes. What's going on there? I mean, obviously, there's a lot of spending that's being directed to defense. But as it relates to your content, what's driving? And how should we think about that growth in that end market?