Good morning, ladies and gentlemen, and welcome to the Elmos Q3 conference call. We are still operating in a very challenging market environment, heavily impacted by the ongoing semiconductor shortage and by persisting restrictions of the coronavirus pandemic. Once again, Elmos was able to post record sales numbers in Q3 and could further improve profitability. Our solid operational performance, we actually faced less disruptions than we feared this quarter, made the past quarter a very successful one. As usual, you have the opportunity to ask questions at the end of my presentation. The COVID-19 pandemic is still part of our daily lives, so I would like to start my presentation with a brief update of the situation at Elmos. After almost two years, I can say that Elmos has managed the pandemic very well.
Since the outbreak of the pandemic in January 2020, we have introduced extensive protective measures at all Elmos sites. This has enabled us to maintain our own production and business operations until today without any major disruptions. Since the end of the summer, we have eased some of the corona-related restrictions regarding travel and business meetings, as well as, of course, less important, in our company fitness studio and sports program. We have also decided to gradually return to a normal office routine with around 50% of our employees now on site, and have introduced a new company agreement that applies for the remote office working in the time after the pandemic, which can be summarized as a day per week remote working is okay.
We hope to be able to ease further restrictions and to finally return to a more normal business life when the current wave is over. However, we remain cautious and are prepared to quickly tighten our safety measures once again, should this become necessary due to higher infection rates or new virus variants. Let me now talk about the current market environment. As we all know, the high demand for electronic products and components across all industries is heavily impacting the entire supply chain for semiconductors since the end of last year. The soaring demand levels for semiconductors are much higher than the global IC manufacturing capacity, which led to a severe allocation situation, which is far from over. Automotive OEMs and Tier Ones are trying to secure as many semiconductors as possible, forcing IC manufacturers to prioritize their deliveries based on the real demand of their customers.
The bottlenecks along the entire value chain led to massive delivery problems and line shutdowns at the car OEM level. In its latest study, AlixPartners estimates that 7.7 million cars cannot be built this year due to this shortage. In its September forecast, IHS, also a forecasting agency, has significantly lowered its light vehicle production forecast for this year and next year. The current chip shortage has more and more effect on other industries as well. Apple, or even Apple, is likely to slash production of its iPhone 13 by as many as 10 million units due to the global shortage in semiconductors and other supply chain issues. Also PCs, tablets or gaming consoles are in short supply right now. I guess we should all order our Christmas presents very early this year or go for homemade cookies, as I will do.
The Elmos focus is to meet all of our delivery obligations based on the real demand of our customers. Every day we are fighting for a sufficient allocation of wafers from our foundry partners. The current situation requires close coordination with the foundries at the operational and on the management board level. Our own fab in Dortmund is also fully utilized and running very well. Overall, the situation is demanding a lot from us, but our efforts have paid off. So far we have been able to supply all our customers so that Elmos had not caused any production line stoppages among the Tier Ones or OEMs this year. Please remember that our higher inventory level at the beginning of this year helped us through the tight situation in 2021.
As our stock levels are now very low, we will not be able to use inventory like we have this year also in the future. It is obvious now that the semiconductor shortage will most likely persist in 2022. This may limit our growth potential next year. I will further comment on this later in my presentation. Of course, you all want to know when the allocation ends, but to be honest, nobody knows for sure when the allocation situation will end. Coming to Q3. In this very dynamic environment I just described, Elmos was able to continue its positive business development in the Q3 with a further increase in both sales and earnings compared to the previous quarter.
With an increase of 98% to EUR 80.8 million, sales in Q3 almost doubled versus the previous year, which was of course heavily impacted by the coronavirus pandemic. This once again is a record sales within the quarter and slightly above the midpoint of our quarterly sales guidance. The EBIT further improved to EUR 13.6 million in the Q3, supported by higher volumes, but also impacted by allocation-related effects and higher prices from our foundries. The EBIT margin improved to 16.8%, which is at least close to our midterm target of 17%. At 16% of sales, R&D expenses continue to remain on a high level. Our R&D activities support the numerous serial launches and new projects in all application fields.
We were able to acquire attractive new projects in all of our product segments, and we are still on track to meet our ambitious target for new design wins this year. Our investments also remain on a very high level, as already explained in the last conference call. The CapEx program is mainly targeted for the testing area to expand the capacities at the Dortmund site and at our partner sites in Asia, which are necessary to secure our delivery capabilities for this year and for next year. In addition to the back-end CapEx program, we have invested in a new building for our software division in Q3. As you may know, we are strengthening our software activities at Elmos. The acquisition of Online Engineering at the beginning of the year was an important component of our software strategy.
Now we are further expanding our activities and building a bigger AMOS software organization. With all these measures, we will be able to further enhance our innovative semiconductors with high-performance software functions and to increasingly pursue in-house software development for our customers. In total, we have invested EUR 18.6 million or 23.1% of sales in Q3 2021. At EUR 23.6 million, the cash flow from operations was supported by a strong net result and slightly positive impact from working capital, mainly by increased trade payables. The higher cash flow from operations, combined with the ongoing higher CapEx spending, resulted overall in a slightly positive adjusted free cash flow of EUR 2.7 million in Q3 2021.
Due to our successful public share buyback program, our net cash position decreased from EUR 47.3 million at the end of June to now EUR 8.3 million at the end of the Q3. We are confident that the last quarter of this year will also be a successful one. We estimate that sales for the fiscal year 2021 will overall reach EUR 320 million ± EUR 5 million. The full-year EBIT margin is expected to reach our midterm target of 17% ± 2 percentage points. I've mentioned it before, the global semiconductor shortage and therefore the challenging market environment and uncertainties will continue into 2022. For Elmos, the wafer capacity situation for 2022 has gradually improved compared to our last communication.
We could secure additional wafers from our foundries, but we still need a lot more wafers to cover what we call the bare minimum demand of our customers and to eliminate the risk for delivery constraints and line shutdowns caused by Elmos. The worst-case scenario, however, a noticeable revenue decline in next year, is not likely as of today. Still, the wafer shortage hurts growth, and we have to put a lot of effort into resolving that. The negotiations with all of our foundry partners are still ongoing. We are pushing very hard for additional capacity, also together with the OEMs and Tier 1s. We do our utmost to ensure that Elmos will be able to continue its successful business development in the next year and beyond. We are sure we have innovative products. We are sure we have leading market positions.
We are sure we have a strong balance sheet, and we are just as sure we have a great and highly motivated team, so we look with confidence to the future. Thank you very much. I'm now opening the floor for questions.
Our first question is from Mr. Veysel Taze. Please, your line is now open.
Okay. Hi, everyone, and thanks for taking my question. Maybe I would like to start off with understanding your guidance a little bit more. Could you explain what's really baked into the upper end of your margin guidance? Because, I mean, really looking at what's implied for the Q4, I think the upper end would be above 25%. How realistic would that be? And if so, what would be the levers that could be driving this?
Well, I mean, we have a rather broad range because a lot of things can happen in this dynamic environment. We ask some customers for certain payments for engineering efforts. This is one thing that should happen, but does not necessarily need to happen. There may be other one-time effects that could happen on the good and also on the bad side. That's why we think a relatively broad range is actually appropriate now. Of course, revenue is also a driver, as you know, but this is reasonably obvious. Revenue is a volatile thing. I mean, we look back at the Q3 and we saw a lot less disruption now towards the end of the quarter than we feared.
We had disruption within the quarter, but not towards the end. It's not really visible in quarterly numbers. Due to COVID in Asia, we had shutdowns and capacity restrictions at some of our suppliers. The world isn't completely back to normal. This is what we just have to accept. It may be transitory problems, but it's still not back to normal.
All right. A second question on the whole allocation topic. You have just said that you have been able to avoid or you're gonna be able to avoid a significant phase decline. Are you, from an allocation perspective, at a level where you would be able to grow next year? Or have you basically reached a situation where you can expect stable top line from an allocation perspective?
Well, the first question is, of course, always in units or in euros. In euros, we would say it's certainly not a decline. It should be some sort of growth. We can't tell you how much. I mean, we do not have price agreements with most of our customers for next year because these are just in the making. We have initial allocations, but we are kind of fighting for more wafers really every day. We have a lot of trilateral calls. We have some bilateral calls. Some of our customers have bilateral calls. There's a lot of heat currently in the system to resolve the wafer issues, which are still unresolved. I mean, they are partly, slightly resolved.
We are trying to do our utmost that no serious and long shortage reaches the OEM and leads to plant stillstands there. This is a very dynamic thing.
All right. Maybe a last question on CapEx. You've been spending quite a lot on expanding your testing capacities. How much more should we expect for the Q4? If it is gonna be some substantial similar to this one CapEx spending in 2022?
Yeah. I mean, currently the CapEx level is maybe not that bad. We will continue spending money in the Q4. On 2022, it's hard to tell for the full year because actually what we spend in the H2 of 2022 depends more on our expectations of growth on 2023 than on anything else. On that I really can't comment because I don't know yet.
We don't know how the end of this allocation will look like, whether it will be smooth, whether it will be a prolonged allocation because everyone now wants additional safety stocks once we're able to build it and we're coming into a phase after the kind of plant shutdown allocation to a allocation where everyone wants to build safety stocks and openly and transparently wants to build it or whether we get a hangover after the party. We fundamentally do not know how this whole thing will play out and end, and this is a key driver, of course, for investments. I'm sorry, I can't tell you how 2022 allocation looks like.
All right. That would be it for now. Thanks.
Thank you, Mr. Taze.
Our next question is from Malte Schaumann of Warburg Research. Please, the floor is yours.
Yep. Good morning. Can you please elaborate on the OpEx levels? It has been quite stable during the past quarters, so with some more reopening, more travel activities, what's the expectation going forward? What kind of additional costs might come through when you could do more marketing activities, whatever?
Yeah. We do see a return of travel to some extent, but actually, on a relatively low level. The whole trade fair thing may be kicked off again with the CES next year. Maybe not, who knows, what will happen? We somehow expect it to return. We do kind of see that travel is by far not back. The video conferencing is convenient, it's very effective also with customers. We also need travel, but there is a certain focus on the key travel activities. I mean, you could dare to say maybe there will even be a lasting effect on travel costs.
Maybe this is going too far, but currently it feels like there will be a lasting effect on travel costs because people think twice before they fly to the other end of the world, which is, I mean, in China you do not really get in well and get out well. Also the inner-European travel. There is some substitution going on with video conferencing.
So far, probably wise going forward, but not too much.
Yeah. It won't be too much. I don't think it will be a big financial impact. At least not, I mean, if you're referring to Q4. Over the next years, this is a long shot, and I don't really know.
Yeah, sure.
This year, I don't think it'll somehow burden our Q4 too much.
Yeah. Early next year, do you expect further increasing dynamic next year or early?
I would say it will be a slow development.
Yeah.
I mean, it of course depends whether you believe that the pandemic will be over, kind of in a big flash and then, corona is gone. My assumption is more that it'll drag out a little bit, and a lot of things will become possible but burdensome. Also the learning effect from this video conferencing and, there have been a change in habits, and a change in kind of meeting options. People go for this virtual options in a way that wasn't imaginable before.
On this Q4 sales level, the increment of sales in comparison to the Q3, is that already driven by additional wafer capacity going forward? Do you have kind of then similar or should we take a Q4 run rate or available capacity that the number might indicate, as available capacity then going into next year, early next year, and then as a starting point?
I would love to, but it would be wrong. Yeah. I mean, our implied Q4 guidance is more than we could. This means that there's actually some payments that are not directly linked to wafers, some price increases that are necessary to cover additional costs affected in. But also some one-time effects that will likely come. I wouldn't calculate as easily for next year as the timing for Q4.
Yeah. Okay. R&D funding, et cetera, also plays a role, probably. Then-
We have to get that money. I mean, every million counts.
Yeah. Yeah. Sure. On your visibility in the negotiations, if you can elaborate a bit on that, how negotiations proceed. Is that kind of a structured process? Is that kind of a black box? Is it, you know, fixed dates until when you should have final visibility for next year? Is that really ongoing, and you might still discuss the Q1, Q2 about capacities in the H2? Yeah.
I believe it's a structured process. I mean, it's a fact-based process as good as people can base it on facts, but it's not pre-structured. There are no fixed deadlines or auctioning off dates of capacity that we have seen with memory years back. It's just a set of exchanges between the interested parties, the OEMs, the Tier Ones, and the foundries. This is actually a pretty continuous process. So you wouldn't expect that, when you have a certain gap in wafers, that you get that filled at once or on a single day or even a single month.
You may talk about a very small increment, and then you agree that certain increment is doable, and then you have a few thousand wafers more, if it's a few thousand or sometimes less. Yeah, this is just a moving, developing thing. The talks continue because they need to continue.
Yeah. Do you think that the decision-making process has changed at your supplier side?
Fundamentally, I think no, but I mean, when I read in the newspaper that Apple is not building iPhones, that I believe highlights how tight the world there really is. I mean, if you read in the newspaper that a Chinese toy is not being built, that's sold for $10, okay, that tells you really nothing because it doesn't matter in this world. If you look at iPhones, I mean, these are relatively margin-rich devices. You may earn more on an iPhone than on a basic small car. If iPhones are not being built, that tells you something. It tells you it's really super tight. The trade-offs that the foundries, namely TSMC, for instance, have to make, they are real.
They are not kind of virtual and in reality there is enough for everyone. No, these are real trade-offs. You're deciding is an iPhone not being built or is a car not being built? You try to get that all squared and that both things are being built, but it's a hard thing to do. You make trade-offs on both sides.
Yeah. Finally, final question would be, do you see the risk of losing market share and suppliers in dual supply situations where maybe competition has more capacity availability to in-house production or other reasons?
We see a certain development that some customers are forced by their decision-making bodies to evaluate options to ease the supply situation by qualifying other chips. We're also asked a lot, "Could you do that?" In certain cases, and under special conditions, we can help, and in other cases we can't help. I don't think there will be a structural shift in market shares.
Okay. Fair enough. Thanks.
Thank you, Mr. Schaumann.
As of the moment, there are no further questions. Ladies and gentlemen, please give in the key combination nine and star key one to raise a question. Our next question is from Mr. Robert Sanders from Deutsche Bank. Please, your line is now open.
Yeah, good morning. Thanks for taking my question. Arne, last time you talked about TSMC not adding any eight -inch capacity next year. Has that changed, or have you seen some reprioritization by that foundry, or have you found wafers from elsewhere? That's my first question.
Well, on the capacity side, not to my knowledge, that changed. Yes, we got some more wafers, and that applies to all of our foundries.
Got it. My second question is it possible that an IDM, like an Infineon, could engage in some foundry business in order to support the overall industry? You know, we have seen in the past some IDMs doing outsourcing, and would that be something you would ever consider doing with a potential competitor?
Well, we certainly would consider it. I mean, you have to ask them whether they wanna offer it. I believe in this current environment there are more people kind of on the asking for a capacity side. I mean, we've been asked whether we could kind of give a little bit of our fab capacity to a certain foundry. We said, "Honestly, no," because we need it for our own. I mean, what an idea. I mean, going out of such an allocation phase and going into a phase where you actually need to fill capacity, this may be a viable model. I mean, you shouldn't say no to such discussion.
Got it. This year, I think everyone understands that the microcontroller was probably the biggest path dependency and reason for line downs. Looking into next year, it looks like a lot of that demand will be served by TSMC. A lot of kind of 95% finished cars that are sitting around will come onto the market in the H1, I think it's fair to say. Is the issue then next year that microcontroller is less of an issue, but sensors and power discretes or something else becomes an issue? I guess what I'm trying to understand is there gonna be a sudden kind of increase in production in the H1 as microcontrollers deal with that issue, but then do you become the cause of line downs?
I think in the past you've talked about being potentially the cause of line downs in Q1.
Yeah. This is very hard to say. I mean, on relatively stable volumes, we do our utmost to be not the cause of line downs. Any sudden jump is of course a risk. That is true. We are all in the industry facing the problem that there is never complete transparency, where material is, how quick it moves, what cars will be built in six months' time. If you do not act now, then problems will not come tomorrow, but they will come months and months later. Yes, there is of course a risk that you solve one problem or put out one fire now and then another fire lights up in a few months.
We try to be as transparent as we possibly can with our peers, with the OEMs to the foundries to show what is needed to avoid that second fire. We had some partial success with that, not complete success. We're still in a risk zone, I would say. The discussions are also ongoing. Let's see how much we can, with all the trade-offs that are going on in the industry, move out of that risk zone.
Got it. Last question would just be on despeccing. You know, there is this idea that some OEMs like GM are willing to sell cars with slightly reduced specs in order to get them out of the production, you know, out of the parking lot. Whereas I think Ford were publicly saying that they would never, ever sell a car that didn't meet the spec. I think in the case of some of the products we've talked about, you know, climate control maybe being despecced to one zone rather than four or this kind of stuff. What are you seeing in terms of behavior at the OEMs, and how does that affect you guys?
Well, we have not seen that too much. Maybe because our product portfolio, I mean, we do have some things that can be specced as extras, but a lot of the things I believe would be considered core for buying the car. It's hard to leave away an airbag. It's virtually impossible to go from one lighting concept to another and still have the same car. With the ranging thing, this is something that is a last customer feature. Taking that away means people would switch brands. Per se, it's not a bad idea, I think, to think about despeccing if that solves a shortage problem. I believe in reality, it is not really easy to do.
It's easy to do on a rarely taken extra. It's not easy to do on the core of the car.
Got it. Actually, one last question was, would be, you know, obviously this year the OEMs have prioritized selling premium cars 'cause that's where the margin is. How do you—I think historically you've been more exposed to kind of the mid-range rather than the ultra-premium segment. Just can you if there is big snapback of basic vehicles, how does that affect you guys in terms of content trends? Obviously, the guys in the premium segment have seen a big jump in content relative to history in the current year, but obviously that's not gonna last forever. I was just wondering how that affects you guys if there's a big snapback away from premium back to kind of basic cars.
Yeah, I wouldn't guess too much. I mean, we are also exposed to the premium cars. The only thing is when we think about a product going only to an S-Class, only to the big SUVs, it won't make the necessary volume to really justify a typical product because it's not millions and millions of ICs a year that can go into super premium cars. So I would guess that we have a lot more chips, of course, in a premium car than in a basic car. So in a way, we like premium cars being sold.
Overall, we need the volume cars, and we need the electrical solution in the volume cars and the innovation in the volume cars. I think that is a move that is already happening. You can buy full ADAS functions in a Golf or almost full ADAS functions in a Golf, even in smaller cars. We do not really fear the shift from premium back and forth. This is not as strong an issue for us.
You're agnostic to the move. I mean, it looks like that plug-ins are kind of gonna be taken out of the roadmap in four or five years and replaced with battery EVs. That trend seems to be accelerating quite rapidly. You're kind of agnostic to that trend as well, right?
Yeah, pretty agnostic. If anything, we like it being, I mean, we like it being more hybrids, but we also like it being more EVs or pure EVs. It's not super big for us in quantitative terms, but it's very nice for us, and it opens new chances. We are discussing a lot more with our customers on EV solutions now that in the predictable future, we will see substantial volume. Because always when you wanna do kind of specific ICs, you need to bring volume. It's just not enough to say, "I have 100,000 units.
Do you see that roadmap, major roadmap changes? That's what we see is basically cars that were anticipated to ramp in 2025 are being radically rethought, because of the shift to EVs. Are you seeing a kind of, a flux in the roadmap?
We kind of do see changes at the OEM level, but we do not see changes at our level because our products are actually I mean, for new EV products that you discussed, they get a little bit more momentum out of such things. For the broad part of our portfolio, we actually agnostic towards these changes because they will have no impact. You will take our ranging product anyhow, you will take our motor product anyhow, you will take our lighting product anyhow. Regardless of whether you cut out the hybrid and move to pure EV or you're a little late and have a diesel engine, it doesn't matter for these products.
With some new products, we like to move towards EV because this is even more electrification. This is good for us. On the first level, we are agnostic, and on the second level, we like it.
Got it. Thank you very much for your help, Arne.
Thank you, Rob.
Our next question is from Mr. Dylan [Belgen] of SKM. Please, the floor is yours.
Hi. Congratulations on another strong quarter. I was just trying to think from a longer term perspective, how are you thinking in terms of the revenue growth for the business given the uncertainty on the supply of semiconductors?
Well, I think we're going through a phase of allocation now. With all the semiconductor investment activity, which is very substantial, and there's also a tailwind from politics, we will get out of that allocation in due course. I believe in the long term, what is relevant is demand, because supply will adjust. We have a positive view on demand in our TAM. We see most people forecast that there will be a rising number of cars. I mean, our current level of cars being built is not. This is just not a lot. There's a rebound that is possible there, if you're talking now, on the mid and long term.
For sure, the cars will have more electronics in them. In general, we think we are in a great market.
Got it. Thanks.
Thank you very much.
As of the moment, there are no further questions.
If there are no further questions, thank you very much for your participation and your interest in Elmos. I would also like to remind you that we will publish our preliminary 2021 results on 17 February 2022. Finally, I would like to wish you all the best.