Aptiv PLC (APTV) Earnings Call Transcript & Summary

We are really excited to be back in person here at the Barclays Global Automotive Conference. I'm Dan Levy, I lead autos research coverage, really glad to be back at Barclays. And we're going to kick it off. We have a great lineup today. So we're going to start with Aptiv. So we can -- is the webcast live? Okay. Okay. For everyone on the webcast thank you so much for joining as we start day 2 of the Barclays Global Automotive Tech and Mobility Conference. Very pleased to have with us Aptiv, a large Tier 1 supplier who you all know very well, focused on products in the electrical architecture and electronic safety area or sorry, AS & UX. Joe Massaro, CFO, who you all know very well. So the format is going to be fireside chat. I will ask Joe a series of prepared questions. Anyone who has questions, please feel free to e-mail either of my teammates. So Jason Stuhldreher or Trevor Young. So trevor.young1 and jason.stuhldreher, who you -- probably have seen is e-mail around, they can ask questions for you anonymously or for the audience, please feel free to raise your hands and ask. But otherwise, we will kick it off.

So Joe, maybe a good place to start is just on the near term. On the third quarter call, you said potential for revenue guidance could be to the upside if China ends up in line with IHS, we've seen China revisions do fairly well, I think, in part because the purchase tax has helped. So maybe you could just give us an update, anything you want to say about how fourth quarter is trending.

Sure. No, I think for the most part, continue to be in line with what we talked about on our earnings call, right? We're -- continue to see North America be fairly resilient. Europe has been challenged. We called out the challenges that we thought were coming in Europe for a couple of reasons, back in the middle of the year, late May, early June. China is interesting. As we talked about on the call, they're clearly working very hard to hit higher production levels. And we saw that in October. We certainly saw some challenges, though, in November. And when I say we, certainly the industry just not Aptiv. I think you saw November retail sales were out in November a couple of days ago. Those were lower. I think maybe surprised a few people with how low they are. I do think that's reflective of what you're seeing in the Chinese market. There's clearly, an impact from COVID. We're not at the level of shutdowns that we were maybe in Q2. But there's certainly some disruption. And what we see is a continued push to hit higher production levels. But when you do that in a disrupted, sort of choppy environment, one, you may not get all the way there in terms of where you wanted to be from a full vehicle production, but it's also more expensive to operate in that environment. So I would say China, choppy production, clearly still trying to get to, call it, round numbers, that's sort of 27 million units. But I think November was a challenge. It will be interesting to see to what extent they can really recover over the course of December.

Great. The other regions, I think we've been hearing from some of the other suppliers as well that the choppiness or limited schedule visibility is that, that continues. One company mentioned yesterday that's happening in North America. Are you also seeing that?

There's some, I mean we talked in Q3, we had a couple of our larger North American OEs that were significantly impacted by semiconductor constraints, not on our products, but on other products, so it obviously impacted our deliveries to them. Yes, I think the choppiness, whether it's COVID and supply chain related in China or supply chain-related in Europe or North America, I think that's going to be with us for a while, certainly into next year. We cautioned on the Q3 earnings call that we did feel that the potential for choppiness and continuing into 2023 was there. Certainly, what we're seeing over the course of this quarter would continue to suggest that, right? And I think supply chain constraints have improved sequentially, I think, continue to improve. If you go back to what was a very low point of part availability in that sort of Q3, Q4 time frame in 2021, where you had broad-based impacts. The impacts now are much more acute. They tend to be on specific, particularly around semiconductor -- specific semiconductor products or a semiconductor provider that's having challenges. But they can lead to shutdowns over a couple of weeks period of time. And as we've talked about, and as I think others in the industry have talked about, you're -- one of the most expensive ways to operate in automotive is with a very choppy production schedule. And it's -- it comes down to just very basic things. If you're shutting down for 3 or 4 days in a week, because a customer shut down or because of parts availability, you have to do things like continue to pay the labor, right? Obviously, it's the right thing to do if you're sending folks home for 3 or 4 days. But just as importantly, if you're shutting down Tuesday, Wednesday, Thursday, you need them back on Friday. And you need to make sure the employees are ready to come back and ready to operate. And in a lot of cases, operate at sort of overtime levels at peak production to try to make up for things. So that type of environment is clearly continuing in Q4, and we expect it to continue into 2023.

And the weak link on production as you said that this choppiness should continue into 2023. Chips an issue. What are you seeing on Tier 2s, Tier 3s? What are you seeing on logistics, freight? Are these also constraints?

I would say at the moment, they tend to be more things that require to be managed. They're disruptive. We obviously spent the last couple of weeks and we're not -- Aptiv is not heavily impacted by rail in North America, but we did have some things on rail. So you work though, okay, how to pull everything off of rail for the next 3 to 4 -- next 3 to 4 months, try to get them on trucks, right? Just try to get ahead of things. So there's those types of potential disruptions or disruptions that require you to manage through them. And I think we try to do a very good job of looking around the corner, trying to think proactively about what could cause disruptions and try to manage through them. Unfortunately, you can't do that. You're not perfect all the time when you try to do that, some things come up. And then we've been saying this for a while. The other challenge is, it's ultimately the customer supply chain come through, right, because you can't build the car without all the parts. So most of the impacts we've had from choppiness aren't necessarily our ability to get parts or our ability to work through the logistical challenges, but we're getting impacted from customer schedules that are coming down because they're being impacted some other place in their supply chain.

Great. Let's dig in a bit more on chips because I think if you look at some of the chip data from a very high overview, actually chips are in oversupply right now, right? And you're hearing about consumer electronics slowdowns and so it's sometimes a bit difficult for people to reconcile how chips are still constrained in automotive when the rest of the semi industry is seeing actually excess supply. So maybe you can give us a sense of why on a deeper level, you are still seeing these constraints. It's just these older architectures or higher node counts. Is that still an issue? Are you seeing any relief on that whatsoever in availability?

Listen, I think part of part of what's important to understanding sort of the semiconductor constraint is understanding there were a lot of causes of semiconductor constraints in 2021, right? You have this supply-demand imbalance that started really with the V-shape recovery in auto. And I think the semiconductor industry was planning for much more of a U shape despite what they were seeing from ourselves and others with customer schedules. So you have that supply-demand imbalance. That is compounded by very long lead times, right? These products or the chips we put into our systems can have anywhere from, on average, a 20- to 40-week lead time, right? We have products that have gone over 52 weeks of lead time, right? So what that means is even if you get some relief somewhere in the bottleneck, right, you tend to have a period of time until that relief shows up in actual chips. So that imbalance is what we continue to deal with. What's gotten better is if you remember in '21, we were impacted by the fire and the Naka facility at Renesas, right? That was a plant that is heavily concentrated for automotive, right? So that was very disruptive in Q2. Q3, Q4 last year, you had COVID impacts in places like Malaysia, Indonesia, Southeast Asia, where a lot of test and assembly for semiconductors is done, right? That was an impact. So those are -- those have clearly gotten better. The supply-demand imbalance has improved. Like I said, you still have some very acute problems. As it relates to consumer electronics, and we get this question a lot, I think there's very few, if any, finished semiconductor chip that could go into a laptop or go into a car, right? There is not a lot of one-to-one sort of swap ability at the final chip level. Where you tend to see relief for capacity creation, if consumer electronics comes down, is all the way back to wafer manufacturing availability or things like lead frame availability, some of the basic materials that will make a chip. So slowdowns in consumer electronics ultimately could help industrial and automotive-grade chips. But you're then faced with that lead time issue, right? If corridors are opening up at wafer manufacturing and we're able to put more automotive chips into the process, it's going to be months. As I said, in some cases, we're averaging 20 to 40 weeks. So you're going to see a significant period of time before you wind up with more chips. I think the other thing that we're dealing with now relative to a 2018 or 2019, our view is the semiconductor industry typically worked with a lot of inventory in the various channels, not necessarily on our books, but throughout their system would have, whether it was die banks or inventory at distributors. And they were able to sort of insulate their end customers from problems in production, yield being lower than expected or some type of problem there, right? There was enough inventory in the system where the end consumer didn't necessarily feel it. I think inventories have been taken down so much over the past year. Just as we try to consume as much as possible to hit the manufacturing schedules of our customers that now if you have a yield problem, right, and you're running at x percent yield and you're 20% below that, we're now feeling that, right? There hasn't been enough inventory build back in the channel to sort of insulate. So I think that's going to take more time, which is, again, why we're just cautious on 2023. This is not going to be fixed January 1, right? Sequentially, it should continue to get better, barring some other larger event, I think just natural course, it will continue to get better, but it's going to continue to get better over time over the course of the year.

And the pricing situation around chips, there was clearly some price gouging or some price...

There was a significant amount of chip inflation, I will say, I'm not going to go any further with that in terms of how we might describe it. But we saw significant chip inflation throughout the course of the year. We were dealing with 25% price increase sort of on a quarterly basis. Those have slowed, the pace of increases have slowed. We do have some suppliers that are talking about price increases in the first quarter of next year. We're obviously pushing back on those. I think you just saw in Q3, though, we did a very good job of pushing that price through to customers. This year, we'll have passed through over $0.5 billion of price increases to our customers. They were obviously, in some cases, challenging discussions with customers on how to get that price. But we felt it was very important to push it through. About 70-plus percent of that increase is semiconductor related. So it was a significant amount. Now obviously, we're challenging our supply base to the extent they come back with more price increases and we'll continue to do so. And if they do wind up coming through, we'll obviously manage in the same way we've managed through this year.

Right. On that topic, maybe we can just talk for a second about costs. So you've taken on this year, it's roughly $300 million of COVID and supply chain costs. I think it's $215 million of COVID, $80 million of -- sorry, $215 million of supply chain, $80 million of COVID and then you've had some material costs as well. So maybe you can just recap that? And then give us a sense of what's sticky, what isn't and the mitigation efforts around...

Sure. No, that's a good place to go. So you're right. It's about $295 million this year of supply chain and COVID disruption costs. The COVID's $80 million, we spent about $20 million a quarter, mostly on PPE, safety protocols at the plants globally. And we continue to spend that. One, because it's all about employee safety. Our safety measures have proved very effective. We have not been in situations where we've seen sort of plant -- our plants closing down because of COVID. We've been able to keep our employees safe. And we're in places where -- manufacturing places where vaccination rates aren't as high as they are here in the U.S. or in Europe. So I'd expect those costs to continue for some period of time into 2023, again, about $20 million a month. Obviously, when we get to the point where vaccination rates and such, are at a point where we can spend less, we will. But our -- we're going to [ eer ] on the side of employee safety as it comes to spending that money. The remainder of that $215 million, very much supply chain disruption costs, things like premium freight, things like, as I mentioned, downtime for labor when you're paying the employees when the plants are closed. That number has been improving. I mean last year, that number, that $215 million was well over $300 million. So we've picked up about $100 million of improvement in that number on a year-over-year basis. There's nothing sticky in that number. As supply chain is -- inventory availability and the supply chain improves, we will do less premium freight, right? We will go back to our normal supply chain flows. There's nothing structural that's changed around that $215 million. So I think it's a matter of do those costs come down over the next, call it, 6 to 8 quarters? They do. How quickly really depends on how fast things improve at this point. But like I said, we have seen those costs come down, high confidence they're not structural. But it's really a matter of just how quickly things improve. And I'll make one caveat, in Q2 we incurred a little over $30 million of supply chain, COVID disruption costs in China alone with the Shanghai shutdown. So they can be lumpy as well, right? We saw a spike that we weren't anticipating, obviously, at the beginning of the year and wound up spending $30 million in a quarter. So they can add up quickly. But again, we don't view those as structural. On the material cost increases, again, over $0.5 billion, we've pushed those through to customers. We expect that to be in piece price in 2023. So we'll have sort of a gross up of revenue to match the gross up of cost and the cost of sales. That has been obviously our biggest remediation action. Over the course of the past couple of quarters, the team has done a great job of working with our customers. And it was a bit of a lengthy process, and I understand that. Our customers wanted to, one, make sure that we weren't coming to them first. I think that's fair. If I was sitting in their chair, I'd want to make sure other steps were taken to try to eliminate those costs versus just passing them along, and then obviously, you have to go through a documentation process with customers to show we are actually really incurring those costs. We took the approach of just really just passing along dollar for dollar the direct material cost. So we thought that was the right way to approach it and the best way to sort of expedite the discussions with customers. But that's really been our biggest remediation action. We'll have inflation. We're dealing with inflation around energy costs in Europe, for example, those types of things, which we're working through from a performance perspective to offset, and I think the team has been making good progress. So there's obviously inflation beyond material costs in the environment but certainly trying to deal with that through our improved performance.

Great. One last one on the near-term operational environment. Historically, I believe you said your incremental margins have been in the high-teen percentage range on a sort of an all-in basis. I think if we look at the latest IHS numbers, it's for, call it, low single-digit improvement year-over-year. To the extent that there are actually year-over-year improvements in LVP next year, should we expect a typical incremental margin? Or are there other factors at play?

Listen, I think a couple of things there. I think as it relates to vehicle production and some of the estimates that are out there for 2023. We are very cautious about how realistic those are. So I would -- again, we haven't provided numbers yet. We're working through our customer schedules. But I think just given all of the challenges, and I understand there's some arguments there for some [ bull ] cases. But certainly, from a planning perspective, we're not going to be on the sort of 4% to 5% up next year. Where we fall ultimately, we're still working through. With that said, we will flow with volume. I think our EBIT flow-through has historically been 18% to 22%. And given our growth over market, we're still very confident in our growth over the market of 8% to 10%. So even in a, call it just for example, a flat vehicle production environment or a vehicle production environment that's down slightly, we'd expect to grow in that 8% to 10%. And you will see at the EBIT line us flowing at 18% to 22% on that volume. There's certainly going to be margin upside potential on how quickly that $215 million comes off those supply chain and disruption costs. So there's certainly opportunity there. Again, I can't call the timing, but I would expect those costs to come off over the next 6 to 8 quarters. And then to the extent -- and again, I'd be very cautious on 2023, but ultimately, vehicle production, we view starts to come out of the sort of the low 80s, the sort of, call it, 83, 84 range and starts to move up closer to 90. As that happens, we are prepared to flow more strongly on with the combination of growth over market and higher vehicle production. We've continued to work on performance initiatives within the business. As I've talked about a number of times, we certainly dealt with the variable costs in our system given the lower vehicle production, but we did not close big plants. We did not consolidate facilities in part because if you're going to close big connector plants or close big electronics plants, those paybacks tend to be 3 to 4 years. And our view is by the time we would sort of get through that and realize that payback, we're going to be back to whether it's the high 80s or low 90s sort of vehicle production range, you need the capacity. So with that said, as we start to move off this vehicle production level to higher numbers, which we believe, over the next couple of years will happen, we believe we'll flow very strongly in vehicle production on top of flowing on the growth over market. Would just be cautious about -- we're going to be very cautious, assuming that happens next year though.

Right. You mentioned the growth over market. This is actually a good pivot point. Let's talk about your bookings. You're $25 billion through the first 3 quarters, which already is an annual record. So what is driving this really strong uptick? Is this just all the things we've heard about over the years, just aligned with the megatrends, CPV, I mean, we know about -- we'll go into high-voltage and active safety and SVA. But just from a high-level overview, what is driving this extraordinary surge in bookings.

I think you hit upon it. It's really the next leg of safe, green and connected. The types of bookings we see this year in Q4 will be another good bookings quarter. So the trend is continuing. The types of programs we're booking now with major global OEs as well as a couple of very strong Chinese OEs are the next generation of what we've talked about really over the past 5 to 8 years, right? If you go back to 2017, 2018, we were talking about domain centralization, getting large active safety wins that were consolidating all the individual active safety controllers within a vehicle into these larger L2, now L2++ systems. This generation of bookings is the continuation of that, right? We're starting to see large zonal controller which, again, a larger version of domain centralization where you're pulling more and more compute into larger platforms. We've had a couple of central -- CVC central compute wins. So it's really the next phase in the consolidation of the architecture from both a hardware and a software perspective in the vehicle. And it's what we've talked about. If you go back even at the Investor Day or CES in 2019, where we talked about benefiting over the next, call it, 5 to 7 years from domain centralization and the growth of Level 2, Level 2+, but ultimately, that becoming smart vehicle architecture and continuing that trend. And I think just given the capabilities we've established with large domain centralization, particularly in things like active safety or the large cockpit controller boxes that we've been able to do over the last couple of years. That puts us in a very strong position to continue to lead in the consolidation of compute in the vehicle. And again, for both the hardware and a software perspective.

Right. Let's unpack some of those trends. Let's just start with active safety, $5 billion of awards year-to-date. Maybe we can just take a step back because I think some people were questioning or had asked the question of your role as a Tier 1. We're obviously seeing one of your competitors who was -- or one of your partners who is a Tier 2 and is now starting to get into the Tier 1 space a bit. Remind us again your offering there and the awards that you're seeing year-to-date, what does that comprise of? Is that -- what's the hardware piece versus software? I know you've talked about more software abstraction, give us a sense of what's driving those awards and remind us of sort of your value proposition in...

Sure. Now the active safety systems that we've been deploying over the past 2 years, what I'll call the Level 2+ systems as well as what we're winning, Level 2++ systems, and we -- we won one system in Q2. It was almost a $3 billion award for a full active safety system for a global OE, and it's going to be their global standard. It's going to be the system they deploy globally. For us, what a Level 2+ system is, is exactly that. It's the full system, right? So you clearly have perception whether it's the radar, the vision, you're right, we do work with partners on the vision algorithms. But it's also the central compute. It's the ECU that drives the entire system. We're providing a lot of the software. Now we're open and we're big believers in open system. So if a customer wants to incorporate some of their software into that system, if there's software from other suppliers that need to be incorporated, again, we're fully capable of integrating that and delivering a complete system. But that's really what we're winning. It's -- and from a software perspective, you certainly have the vision algorithms. You have the radar algorithms, which is something we've done. We've been in the radar. We do over 40 million of radar systems a year at this point. We're doing the sensor fusion, the policy and the path planning software and again, incorporating other elements of software, whether it's the HMI from a customer. In some cases, a feature like maybe parking comes in from another supplier, those types of systems all get incorporated. But we're doing the full system. Certainly capable of doing an entire system. And we launched the system in the past year, where we did everything, including the highway pilot. We launched another very similar system almost in the same time frame where the customer chose to do the highway pilot system. So again, for us, it's important to remain open as we move forward with the Wind River software. Our view is that it's a very open architecture. And what's important about that. And it's a little bit back to my point of the next generation that we're booking, right? And Dan, you mentioned it. Software is going to be continue to be abstracted, right? We've gone over the past 5 or 6 years, where there were a number of different controllers that we're participating in, broadly speaking, the active safety in a vehicle, right? We're now down to one large domain controller. [ When ] continue to pull in various pieces of software into those larger controllers. As you get to SVA, there's going to be a larger controller that not only does active safety, but it'll do other functionality within the vehicle, right? So all of that software is going to wind up being abstracted into those larger control boxes. And that's really the space we own and have a very strong position of those larger controllers, they're going to continue to do more within a vehicle.

Great. A question on high-voltage. In the past, I think you've noted a very, very high win rate. I think it's in excess of 50%. Are you still seeing that type of win rate on high-voltage?

We are to date. I'll go right to your next question. If you back up for a minute, broadly speaking, on a vehicle electrical architecture within a vehicle, including low-voltage, Aptiv has content on about 1 out of every 3.5 vehicles manufactured globally. So we've got a very good strong position and a very strong sort of right to play in continued electrification of vehicles. The high-voltage business has been running higher than that share. We have that number you were referring to, we've got content on 50% of the EV launches in the next 2 years, right, a very strong position. We'll continue to grow that business. Our high-voltage business will end this year at about $1.2 billion. It's growing about 30% per year. We expect it to continue to grow 30% per year for the next few years. On top of that, we announced this morning, we completed the acquisition of Intercable, so that will be another EUR 250 million of revenue that's added to that high-voltage business. So that will continue to grow. It's going to continue to grow north of 30%. Intercable is growing north of 30% as well. It's likely that share comes down, though, right? One of the things we've done is we're very selective on the EV, not just the OEs, but the EV platforms that we take on. We want to make sure those cars get built, we want to make sure that our customers are committed to those platforms. I think as time goes on, and you've got, particularly in China, where you've got more OEs producing electric vehicles, we're not going to be on all of them, right? So I think our ultimate high-voltage share will be somewhere between that 1 out of every 3.5 and the 50% we're at now. But it's a great start for the business. Again, that's a product line if we were talking in 2019 would have been in the hundreds of millions of dollars of revenue, like I said, organically, we're at $1.2 billion. With the addition of Intercable, we'll be close to $1.5 billion plus next year. So very strong.

Great. I'll ask one more, and then I know we're running up on time, but I also want to give anyone an opportunity to ask questions in the room. SVA, I think in the past, you had -- you said you had all of these development awards. And I think the way that it works is you have a development award, you do some development work and then all of a sudden, that leads to a serious production win. So I think the number was 20 development programs with 10 OEMs, and you now have 3 awards, 1 with Geely, 1 with the German OEM, another with the same German OEM. So maybe you can give us a sense of how that's progressing. And I don't know if you can give us a flavor of the type of CPVs as these zonal control awards are coming in, what type of CPVs those are leading to?

Sure. And we'll provide -- we're going to have a Capital Markets Day in February of next year, and we'll update all of the CPV content then including SVA. But I would -- broadly speaking, think of SVA as higher than a Level 2+ system. We've talked about that being between $1,200 and $1,600 depending on what's in the system. So SVA, again, it's bigger compute. It's got more with it. Now it will depend on what you're doing. If you're just doing the controller, you'll obviously have a piece of that, but the TAM is going to be north of where we see the active safety business today. So it's a very attractive space for us. Advanced development, you're absolutely right. One of the things we've talked about in the past is this is how really technology gets introduced to the market in the automotive industry, right? And this started -- this is how we got into the Level 2 business. We had an advanced development agreement in 2012, 2013, for a very robust Level 2+ system with Audi, that became zFAS that we launched in 2016. And SVA is playing out very much the same way. We have a number of OEs who we've been working with for advanced development projects on what the architecture looks like. We get reimbursed for some of those costs so it's not pure R&D for us. It is sort of joint development that customers are willing to pay for in those formalized agreements. And one; obviously gives us a lot of learning. There's IP developed that we get to keep, obviously, in that process. There's IP that a customer may get to keep as well in that process. But obviously puts us in a very strong position to win work, once that goes out. Now automotive is automotive, right? Customers are going to go out and bid the work. We're fully aware of that. But certainly, the advanced development projects allow us to get a leg up. And I think that's one of the reasons you're seeing the bookings and SVA, what now, $5 billion in total bookings already. And certainly, those advanced development programs and the work we've done with customers have allowed us to jump and get an early lead. I would also say, we've talked a lot about -- when you get to Level 2+ complicated compute systems we've really widened the moat over the last few years from a competitive perspective. We've often talked about, really, at this point, from a large compute perspective, Bosch is our largest competitor. We've said that a number of times. SVA is large compute. So you're already starting in our view with a fairly wide moat, right? If you haven't participated in large domain centralization programs with OEs. It's going to be -- it's one, it's hard to leapfrog into the Level 2+ systems it's going to be even harder to leapfrog into large complicated SVA compute systems in our view.

Great. I know we're nearing time, but just wanted to ask a question on Wind River. Any updates on the closing the deal and this trend of software in the vehicle keeps on accelerating. How do you see the deal rationale is all fully in place?

We'll comment extensively. We continue to work through the regulatory process, fully appreciate. It's taken longer than certainly we had initially expected or that's the norm, but continue to work through the regulatory process in the U.S. No change at all in the deal rationale. I think one of the things we benefited from significantly given the extended regulatory approval process here as we -- and we've talked about this, had a commercial development agreement with Wind River in place prior to the acquisition, which targeted a specific list of customers that we were going to go as both independent companies and jointly approach and work with to develop a system. As everyone knows, you got to be careful when you're in an M&A process, you can't sort of go to market until the deal is complete, right? So we're certainly following all those rules. But given the fact that we had a preexisting development agreement, we're actually allowed to continue to work under that development agreement. So despite the extended regulatory approval process, we've been in front of customers that, under that development agreement, have received a lot of positive feedback when River was selected by Hyundai in May of this year to help them develop the Hyundai software stack for future generations, and we continue to get positive input and there's some really positive work being done with several customers under that development agreement.

Great. Quick one to close CES. Any Easter eggs you want to give us?

This CES for us. It's going to be a little different this year. It's going to be a heavy customer focused, particularly given we have the Capital Markets Day on February 14. So I think between CES for customers and February 14 for investors, you're going to see a lot of discussion around SVA and it sort of SVA, it's no longer sort of the concept of SVA, but what we're actually doing with customers. We have a -- CES is also sort of shortened by a day this year just given how the calendar fell. But tremendous amount of customer activity around SVA, around full vehicle architecture, around software in a vehicle. So we're going to be talking to customers, and again on -- and we'll take this through to investors on the Capital Markets Day. But talking to investors about really the -- our vision for the vehicle software stack, including middleware solution, things like containerization, how do you bring in all of the abstracted software into a single system, allow cloud enablement, whether that's development, deployment, enhancements through the entire life cycle management of the software. So I think what you're really going to take away from the next couple of months as we roll out technology is really sort of all of the concepts that we've been talking about through development agreement, the advanced development agreements really starting to become reality. And we're going to have SVA revenues in 2024 and 2025, right? We're going to actually start to see these bookings. We've gone from a advanced development to bookings. And over the course of the next planning period, actually see those turn into revenues.

Great. Great place to close. Joe, thank you so much. Very insightful.

It's good to see you. Welcome. Congratulations on coming back to Barclays.

Great to be back. Thank you.

Thanks, everybody.