© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Hello, everyone, and welcome to Siltronic's conference call on our Q3 results 2018. Please note that this call has been recorded and also streamed on Siltronic's website. The call will be available as an on-demand version later today. Your participation on this call implies your consent with this.At this time I would like to turn the conference over to Petra Muller, Head of Investor Relations and Communications of Siltronic AG. Please go ahead, madam.
Thank you, operator, and welcome, everybody, to Siltronic's presentation of the Q3 2018 results. With me are Chris von Plotho, our CEO; and Rainer Irle, our CFO. Both will comment on the market development and our financial performance and will be available for Q&A afterwards.Please note that our presentation contains the usual safe harbor statement and applies throughout this call and the presentation. Today, we have published all documents relating to our Q3 reporting. They are available on our website.And now I turn the call over to Chris for opening remarks.
Good afternoon, ladies and gentlemen. After a positive start in the first half of 2018, our performance in Q3 was excellent again. Wafer demand remained strong. We do not see any negative impact from memory. Also, logic and as well automotive and industrial applications were strong. ASP continued to increase in the third quarter of 2018. Due to further ASP increases and less headwind from the U.S. dollar than originally expected, sales came in at about EUR 380 million, 5% higher than sales in Q2. Our EBITDA in Q3 reached EUR 160 million, and our EBITDA margin was 42%. EBIT came in at EUR 138 million and ROCE was an outstanding 67%. Net profit for the period improved again and came in nearly EUR 115 million. And net cash flow reached EUR 93 million in the third quarter. We recorded net financial assets of EUR 716 million at the end of September. Because of the successful business development, we slightly increased our sales forecast to slightly above EUR 1.4 billion for the full year 2018.Now let's have a look at the market. Total wafer area in Q3 was up by about 3% compared to Q2 average and by around 9% year-on-year. However, demand is on a sustainable high level and overall, customer raw wafer inventory are still not back to normal levels. We were able to increase ASP again in Q3. And going forward, we see an environment, which should allow further price increases. IHS Markit forecasts wafer demand to grow by around 7% in 2018. We, however, believe that underlying demand, especially for 300-millimeter is higher but limited by available wafer supply. The high demand is driven by 1 stable and 4 growing end applications. While the downturn in PC demand has stopped, we see good growth in silicon area from smartphones, automotive and industrial applications and a high growth in solid state drives.DRAM is one contributor to higher wafer demand. Biggest driver are computing followed by smartphones. Both PC and smartphones are in a replacement cycle. The year-end growth is driven by growth in servers and by increasing DRAM density in servers and smartphones as well.Now let's talk about end markets and growth drivers for NAND. Content in smartphone is growing faster than unit growth. This is mainly driven by memory and more features. According to smartphone for users, content will, in the future, increase by integrating more memory, more cameras and more sensors. If we take the next life on generations as an example, you clearly see the shift towards more NAND memory with iPhone XS, being the first one to have 512-gigabyte as an option. And we see strength -- a strong growth in CMOS image sensors because these are -- there are more backside cameras for iPhone and 3D-sensing capabilities. Both trends have been popularized by Apple and are now being picked up by its competitors.Now let's have a look at the storage market. Storage has been growing mainly by Internet data traffic, and we saw SSD expand faster than HDD despite the significant cost cut. NAND is the preferred storage in high-end data centers and high-performance cloud apps due to the data transfer speed and data compression advantages over HDD. Despite high data growth, NAND supply has been catching up fast, which recently lead to sufficiency ratio over 100% of our customers. Subsequently, NAND prices started to come down. SSD demand in PC is price-sensitive. In the recent history, higher NAND prices slowed down the migration to SSD. Today, with falling NAND prices we expect ratio of SSD that we see to grow. However, the time gap between prices coming down and higher migration might be well 6 to 9 months.If we assume bit increase of 40% per year, the bit density increase of 50% every other year and 20% processing capacity loss due to higher complexity of each generation of NAND, this results in nearly 1 million additional wafers per month needed to match the demand growth over the next 2 to 3 years. So the net result is that NAND players will continue to invest in new capacities to meet the market's demand.Now let's moved to automotive. Content is growing by 7% CAGR between 2017 and 2022. Biggest contributors are advanced driver assistance systems and connectivity, but all other applications in the car are also contributing. We already talked about more sensors in the car in our Q2 presentation. And here, we come across CMOS image sensors again, we referred to when we were talking about smartphones. And we already talked about fast processing and large memory needed in Q2, so I won't repeat myself here. But let me emphasize that all the advanced process nodes for logic are on 300-millimeter, which is growing faster than the other diameter for car applications. Overall, macro indicators might become more mixed but are still mostly positive. The long-term memory demand is intact, smartphone content will increase as well as automotive seeing an increase in content over time. The U.S.-China trade war is causing concerns among corporations, but we expect limited impact on wafer markets like mobile, computing, SSD or automotive. However, we expect some impact on power devices related to white goods. And Chinese phone inventories saw some volatility but do not have an impact on long-term growth. Overall, we do not see any significant change in the market demand. Nevertheless, of course, we cannot rule out political events.With this first overview, I would like to hand over to Rainer for a detailed presentation of our Q3 financials.
Thank you, Chris, and welcome, everybody, from my side. Our sales were up by 5.1% compared to Q2 this year and reached EUR 379.8 million. Compared to Q3, last year's revenue increased by 23%. ASP increased further in Q3. And in addition, we saw a onetime sales effect of EUR 8 million as we converted a customer from sea freight to airfreight. This customer had a critical low raw wafer inventory level. The exchange rate effects had a slightly positive impact quarter-on-quarter as the U.S. dollar averaged $1.16 against the euro in the third quarter after $1.19 in the second quarter.Cost of sales decreased slightly quarter-over-quarter, from EUR 209 million to EUR 204.8 million despite higher area sold. And as a result, our gross profit and gross margin in Q3 compared to Q2 last year increased from 42% to 46%. Higher ASP also drove our profitability, EBITDA at EUR 160 million in Q3 2018 was nearly up 10% compared to Q2. EBITDA margin reached 42.2% after 40.4% in Q2. The EBIT came in at EUR 138 million and was up even more at 11.2% compared to the previous quarter. The EBIT margin went up to 36.3% after 34.4%. ROCE increased to almost 67%, which is due to our high profit and low capital employed. Going forward, as you know, we will see a normalization with higher CapEx.Due to the ongoing positive business development, the steady increase of our net profit continued in Q3. We reached a profit for the period of EUR 114.6 million. This is an increase of 16.9% over the previous quarter. Income tax expense in Q3 were EUR 22 million. The corresponding tax rate was 16%. As you know, there's a bit of a volatility. Our guidance remains that we will see for the full year around 15% to 20%.Our equity increased to EUR 841.7 million. This is primarily due to the net profit for the period. Our equity ratio remains stable and solid, 48%. In Q3, we received further customer prepayments and expect some more in Q4.Our net financial assets increased further and reached almost EUR 716 million, which represents new all-time high. We received year to date about EUR 200 million in customer prepayments and paid back EUR 28 million.Net cash flow increased significantly to EUR 93 million in Q3 2018. CapEx in Q3 was about EUR 63 million. For the 9 months, CapEx is up to EUR 141 million. Apart from MOB that mainly relates to capacity expansion for the capacity coming online in early 2019, the new crystal-pulling hall in Singapore and ongoing automation projects. Our CapEx guidance for the full year 2018 is unchanged, and that means that Q4, we'll see a higher load with EUR 120 million to EUR 140 million and therefore, net cash flow will very likely be negative in Q4.And with that, I would like to turn hand over to Chris.
Thank you, Rainer. Ladies and gentlemen, we remain very positive for the year 2018. Based on the successful business development over the first 9 months of 2018 and what we know about the fourth quarter, we slightly increased our sales outlook to slightly above EUR 1.4 billion. The rest of our guidance remains unchanged.Wafer demand is on a sustainable high level and currently, there is no reason to believe that this will change going into 2019. And we have no reason to believe that wafer prices might go down in 2019, even though we do not, as always, give a price guidance going forward. All this being said with a view to our industry, we can, of course, not rule out any macro influences, mostly driven by political decisions.With this, we close our presentation, and we are now available for your questions. Operator, please open the Q&A session.
[Operator Instructions] We receive our first question from Mr. Amit Harchandani, Citigroup.
Amit Harchandani from Citi. And thanks for taking my questions. Three, if I may. To begin with, firstly on demand. Maybe I missed this, but could you kindly confirm if you still see 300-millimeter demand growing at a 12% CAGR over 2017 to 2020, something that you talked about in the previous earnings release? Secondly, with respect to supply and capacity, maybe a sense for where do you think industry capacity is at the moment. And do you still continue to believe based on your checks that all the players are being rational, and here I particularly refer to capacity announcement plans say from the likes of GlobalWafers and others? And then finally, in terms of your long-term agreements, is there anything different about those agreements versus, say, what you would have signed in 2010, '11 time frame? In other words, can customers break or walk away from those agreements if demand picture deteriorates materially?
Well, Amit, thank you very much for your 3 questions. Let me try to answer the first one first. I saw this question already in your statements before you ask it now in the call. And I was really questioning with Rainer. I think we both do not remember that we ever mentioned CAGR of 12% in demand. What we did is that we showed the picture adding up all the quantities in additional capacities that our customers will create as additional capacity. And we said at the same time that in this time frame, 2017 to '20, these additional quantities, which were I think 2.1 million, could not be supplied out of the existing shares. So therefore, to deduct from this 2.1 million wafer starts addition from our customers, that CAGR will be 12% is mathematically correct but cannot be fulfilled out of greenfield -- without greenfield investment. Supply and capacity picture, your second question, did not change significantly. We still are in talks about additional LTAs with customers for future time periods. Rainer mentioned that we will get additional prepayments in Q4 coming from LTAs that we sign for the future. And about competitors. And what competitors do or what competitors announce, I think we have comparable information channels. So I read the same stuff that you do and therefore, I do not know any more than you. I think the addition, which was officially announced from GlobalWafers in Korea, 150,000, I think it was only the confirmation of what was earlier said more than 6 months ago. And I'm pretty convinced that 150,000 additional wafers should compare to the shipping quantity of somewhere between 6 and 6.3 in Q3, will not have a significant influence on the market. So LTA. I know that somebody, whoever that somebody is, said over -- in the past that LTAs are not -- maybe not worth the paper that they are written on, I do believe that it's slightly different. Our customers sign these LTAs because they are well convinced that they badly need these quantities in the future and they were willing to pay significant prepayments for that. And I do not see any reason why in the short-term, mid-term future, we should come into a situation that demand goes down that significant, that this picture will change. And by the way, these are -- these contracts are at fixed quantities at fixed prices with significant prepayment and the understanding is if the customer does not fulfill his taking obligation, then we keep the prepayment.
We receive our next question from Mr. Achal Sultania.
Well, on the Q3 P&L. So your revenue, if I strip off the EUR 8 million that you called out, if I strip that out, your revenues basically increased by about EUR 8 million -- EUR 10 million give or take, and then your gross profit increased by about almost EUR 23 million. So I'm just trying to understand, like what -- was there any benefit -- what was the kind of benefit that you saw in the quarter? Because usually, we think like 100% of that incremental revenue should fall to the bottom line. And here, it was much, much higher. So did you sign like additional new contract at a much higher pricing during the quarter?
Achal, I mean, it's a combination of 2 effects: it's the price effect and it's cost effect. We had a very good cost performance in Q3. We are happy about that.
Okay. And how should we see that? Is that sustainable going into Q1 -- I'm sorry, into Q4 and Q1? Or were there some things which may not be repeated again going forward?
So the EUR 8 million is a onetime effect. The good cost performance is something that we work on continuously. And we try to do everything to maintain that good performance.
Okay. And a follow-up. So looking at -- you mentioned that the wafer volume growth is still like very strong, 9% year-on-year. And I think we had 5%, 6%, 8% growth in Q1, 6% growth in Q2. It seems like, if I remember, like every, all of -- most of the fabs were already running at 100% utilization for all of the last 6, 7 quarters. So I'm just trying to understand, like, are you still able to do debottlenecking projects, which is allowing you to cater to more wafer demand? Or you basically -- you are not growing volumes but maybe your competition is, which is why, obviously, you're still seeing such strong shipment numbers?
What I think the picture did not change from previous calls. We are fully loaded now for, I would say, 8 quarters in 300-millimeters, maybe a little bit lower number in the quarter for 200- and 400-millimeters. And of course OEE is always going on and continuing, but we always said there will only be a little contribution, low single-digit contribution coming from quantity increase in 2018 comparing to 2017. And this picture did not change.
Okay. So how are we able to satisfy this 10% -- almost close to 10% growth that we've seen this year in the first 3 quarters? Where is this supply coming from?
I think you asked the wrong person. It's obviously not from Siltronic. Like we said. It's coming from somebody else, who that somebody is, I don't know.
Our next question is from Mr. Jürgen Wagner, MainFirst.
Actually, I have 2. When do you decide on your capacity additions for 2020? And how flexible would you be in case the increase from the customer demand is less than what you -- what they tell you at the moment? And the second question is your supplier, TePla, keeps announcing these deals from these Chinese companies. How do you see the competitive threat from China today?
Well, to answer your first question, capacity decisions for 2020, the capacity additions that we will bring up in 2020 are decided and the materials ordered. So there is no possibility anymore, maybe a little bit today, to even do bigger investments for 2020 compared to what we did up to now. And by the way, I want to remind you that we are coming closer to the maximum of our brownfield capacity. Second part of your question was whether we had the possibility to slow it down. Yes, of course, there is a possibility to slow it down. But when I look at the demand side from our customers, I think it is more theoretical question than anything else. And if now I want to give you some more details regarding this aspect. Yesterday or a day before, we -- there was conference call at hynix, and hynix was asked to make statement regarding memory bit growth in 2019. And after some discussions, they gave 2 figures: 20% for DRAM and 40% for NAND. If we translate this into wafer starts, this is something like a high-single digit. And high-single digit is probably more than what capacity additions will be. And your second question, can you please remind me what your second question was again?
On China.
Yes, it was about PVA TePla in China. So our picture on China did not change at all. We have a little bit clearer picture about what their capability is because we had the possibility to look at some properties of wafers which were produced in China. We still have the situation that Zing Semiconductor, they made very precise announcements for the first quarter 2018. I want to remind you, it was 150,000 output, 22-nanometer and 28-nanometer design rule. They didn't consider this in Q1, not in Q2, not in Q3 and we didn't see anything else in Q4 either. We saw some wafers that are going to reclaim those and then will be used as test wafers. We look at them, comparing them to our properties of test wafers. And all indications that we have really confirmed that they are somehow way behind. So we still believe that what we always said, wafers out of Mainland China, 300-millimeter, will not play a major role in the next 5 years, unless they buy the technology. And we already talked about PVA TePla when they signed the contract with GCL. Now there is another one, but the content is still the same. We talked about the hardware. And you remember that I always said, if we get the hardware from SUMCO and we are supposed to run it to produce ingot, we wouldn't have the capability and SUMCO wouldn't have it the other way around because it's a typical combination between the hardware on one side and software and some details, like the hot zone and so forth. So we are confident that we are in a good position.
Our next question is from Mr. Martin Jungfleisch, Kepler Chevreux.
I have 3, if I may. The first one is on your contract structure. Should we assume that the proportion of LTAs is remaining more or less the same? Or would you also not be opposed to moving to 60% or so of LTAs if your customers would ask that? Secondly, on 200-millimeter and smaller diameters, have you seen still price increases in the third quarter? And do you see further price hikes there? Or should we assume more flat pricing going forward? And the last question is on your capacity expansion that you plan up -- that you plan to ramp up next year. Can you provide us an update on the progress? And when should we expect it to be fully ramped and if we should model any ramp-up costs in the first half or so?
Thank you for your question. Let's start with the last one, the ramp. We announced for the beginning of ramp-up, plus 70,000 for 300-millimeter beginning of Q1. We will be maybe a few weeks earlier which is good and the full ramp should be done by midyear and we are very confident that's what we will do, maybe a few weeks earlier. And ramping cost will not be higher as we assumed at the very beginning. So our contract structure did not change. We are slightly above 40% with our LTAs, around 40% on a quarterly level. And the rest is somewhere between 6 months and 1 year. And there is no significant change. To be a little bit more precise, LTA is what we say we are at a fixed price and we are at a fixed quantity over the period. Of course, we have contracts, which have longer running period than 1 year but not fixed pricing. They have price set points after 6 months and after 1 year. And the second question was referring to price development on 200-millimeter and smaller diameters in Q3. Please understand that, as usual, we do not disclose pricing development by diameter. We give a general statement, we saw price increases in Q3 and we see an environment ahead of us, which should allow us to move further increase prices.
We receive our next question from Mr. Guenther Hollfelder, Baader Bank.
The first question just I think you were talking about power semiconductors used in white goods in China that you're seeing or you heard about some weakness in this area.
Well, it's not a weakness in that area. In the trade war, this preference between the U.S. and China, this is the argument which some thought that maybe less white goods will be exported from China. And as white goods also contains some semiconductors, we might see their influence.
Okay. And this mainly affects 200-millimeter wafer demand primarily or...
I would say so. Mainly smaller diameters and 200-millimeter. But please keep in mind, power business is -- or power applications, are to a large extent, automotive and general industrial applications. So white goods does not play a major role. I was only referring to it because when people are asking, "Are there any influence on the trade war?" The product category which is always called out is white goods. And I didn't want to hide that white goods also contains some semiconductors.
Okay. And the second question would be about the wafer inventory levels you mentioned. I mean, we have some large players, like TSMC and Texas Instruments that announced lower capacity utilizations now for the coming months. So would you expect that your shipment into those -- into customers that are affected in general, would remain unchanged and the customers just use the shipments to bring the inventory levels back to normal?
Well, of course, we also heard about the news that you were just referring to. We do not see from customers any slowdown in demand. We have customers where they have a contractual obligation. And within the contract, they have the possibility to call for an option for additional quantities. And these calls for additional quantities, they are still existing and they are still used by the customers. So I do not see any reason that they -- that this should change. On the other hand, there are a few customers where we have a very detailed view on how much inventory do they have with wafers from Siltronic. And this is something as we know our share -- our supply share, at these customers we can calculate the days of inventory that they have. And when I refer to 1 -- to 2 very significant customers that we have, this inventory level is below 5 days. And we all know that the desired level is probably more between 30 and 60 days.
Okay. Good. Maybe as a last question on the customer who moved to sea freight. Do you think...
They moved from sea freight to airfreight.
I'm sorry. Yes. Did you expect this customer to move back any time soon? Or is there a potential that this is offset by -- in the fourth quarter already? Or this...
Well, I do not expect any effect on that on the fourth quarter. For the future 2019, 2020, I simply don't know. But what I know is that the inventory is, at the moment, at such a level that he can't go back.
Our next question is from Mr. Rob Sanders, Deutsche Bank.
Maybe coming back to Achal's question about the mitigator and the production rising to meet the demand in recent months. There is sort of chatter going around the market is that Shin-Etsu asked Mimasu, their reclaim partner to move the test wafers to prime basically. I was just wondering whether that's a likely scenario. And whether that could explain how this growth in wafer demand has been met by wafer supply. And then I've got a couple of follow-ups.
Well, we are not the right people to comment on details what is the deal with partners because we do not have any more information compared to what is available to you.
Got it. But just in theory. If you were to do the same thing, presumably customers would be annoyed because then you would have less testing capacity, right?
In the time of oversupply and underutilization of our capacities, test wafer was a positive outlet. And of course, consequently, prices were significantly lower than the price for prime wafers. In the meantime, it's somehow different. And we, and all the other wafer players, try to avoid test wafers as much as we can, but there is a certain percentage that it always has the properties of the test wafers. And this quantity is, I would say, not completely in balance with the demand of the customer. The customer requests some percent points more. So I do believe if we and our competitors decide okay, let's convert 5% of our test wafers into prime wafers it wouldn't help us because then, our customers do not have enough test wafers to prepare their lines for prime wafers.
Got it. And just looking ahead let's assume the tariff concern blows over and you have to think about an expansion -- a greenfield expansion. How are you thinking about the relative benefit of doing, as of today, more brownfield expansion in either Germany or the U.S. or doing greenfield expansion in Singapore? And presumably, you've already kind of exhausted most of the worthwhile brownfield expansion already in your current plan. So how are you thinking about the relative merits of those 2 things?
I think theoretically you might come into a situation where your question do I do the next expansion brownfield or do I do it greenfield, in reality it never happens because as long as you have the opportunity to only put equipment into existing building, which means brownfield, you would do brownfield. And when brownfield is done, then you don't have an alternative to greenfield. So it's basically in the sequence. First, you do all your brownfield possibilities. And when you are done with that, you go to greenfield.
Got it. And you have the required land next to your existing facility in Singapore to do greenfield presumably? Or is that still a work in progress?
No, no. You're right.
Got it. And just my last question was just on the LTA question from before. This -- I think it's a valid question, is it not? That if there was to be a severe slowdown, that you might want to lock in prices for longer. You obviously have very high return on capital right now. So is there a scenario you think where, for the sake of securing very high return on capital for the next 5 years, that you would actually go to customers and say, "okay, there's going to be a bit of a slowdown. Let's lock in price x for the next 5 years." Because I mean that's what we're seeing in passage already. It seems like it would make sense for you guys to lock in very high profits for as long as possible.
But again there we are more on the theoretical side. If -- up to now, there are some people who believe that there might be a slowdown. We do not. But let's imagine that it will really come to a slowdown. If then we go to customer to lock in high prices, I don't believe that the customer would accept it.
Got it. So that's what I'm getting at. You don't want to be -- you don't want to do it when the slowdown has already happened, right? So I kind of assuming that your LTA percentage would be better off going up. In anticipation, because you don't want to do that conversation when it's already clear as day that there's a slowdown.
So you want to say we should do more LTAs? Is that the message?
I mean, I'm just wondering why your percentage is not going up, it's kind of related to the previous question. Is there some reason that I'm missing that you prefer to -- because in the past, the reason you didn't sign LTAs is because you ended up being underwater 12 months later on those LTAs. But presumably, you're not anticipating another 40% rise in pricing so maybe that risk is much more diminished.
Well, but you know, you can only change it when it's under a renegotiation level, and this is obviously not the case. And please keep in mind that the spread that I gave you 40-40, and something like 20% for in between, is total company and the picture for 300-millimeter is different. But don't ask me for the detail because I won't give it to you.
Our next question is from Mr. Amit Harchandani, Citigroup.
A couple of follow-up questions from my side, if I may. Firstly, let's just stick to your strategy towards adding capacity. You have commented in the past that you would look to add capacity to maintain your market share in the industry. But we've seen some news flow, and it is technically a scenario, in which it's possible that one of your peers or competitors decides to add capacity more than what their market share is today in order to gain market share. In that scenario, would you be -- have you thought about how you would react? Would you be looking to maintain market share? Or would you be willing to take a step back to protect profitability? So maybe I'm making a scenario analysis here, but just thinking if that situation was to come because we do hear this chatter about some of the smaller players adding more capacity. How would you react to this scene? And then I have a follow-up.
Let me try to answer that question. You know, when the industry came into this situation that price increases were possible, everybody wanted to know, "Chris, by when will you start to invest capacity in Siltronic?" And we clearly said at that time we need to see significant price increases, which we achieved in 2017. And then we announced the additional capacity. At that time, we said we add capacity with the expectation to stay relatively static in market share. This is always based on a certain assumption of growth of the market. And you can be wrong with this assumption, what you add in capacity under 2-- because you have 2 preconditions. The market might grow different from what your assumption is, and your competitors might behave different. So up to now we saw slight deviations from that. The deviation for that results is that there is no need to react. And I do not believe that if somebody announces a big greenfield, this will directly implicate that Siltronic does the same. And by the way, I know that the analysts and investors are a little bit nervous about greenfield coming on-stream. And I want to repeat one example. Let's assume that there are 3 greenfield investments happening at the same time, and that all the 3 are ramping with 150,000 per year. This will bring an additional capacity for market of 450,000 wafers per month compared to just by then fully build out brownfield capacity, which is around 7 million. So we talk about 6%. Will 6% additional capacity really disturb the market? When I looked at the expectations that our customers have for growth, it won't have an influence on the balance or imbalance situation.
That's helpful, Chris. And secondly, if I could, just maybe from a supply perspective for your suppliers that is. I mean, you're clearly evaluating at what stage do you think about doing the greenfield capacity addition? Is there anything in the supply chain -- I mean, do you think your own supply chain is ready if you embark on a greenfield capacity expansion at some stage down the line? Just trying to gauge if there is anything that needs to happen in your own supply chain before you actually go and make potential greenfield plan?
No. As far as we judge our supply chain I think everything is fine. Everything looks good.
Our next question is from Mr. Peter Testa, One Investment.
Just a couple of questions, please. When looking at the prepayments and LTAs, obviously, the gross drawdown remains pretty small at this point. I was wondering if you can help us understand at what point you would expect the gross drawdown from the LTAs to pick up. And how significantly -- from the prepayment, sorry. Yes.
Obviously we misunderstood or didn't understand your question. Would you please repeat it?
Just when looking at the prepayment balance, which is EUR 200 million in, EUR 28 million out, roughly. Just trying to understand at what point you think some of the prepayment balances are going to start to get used a bit more? Is this with the capacity that's coming on in 2019? Or are they related to longer contract structures? Just trying to understand when we should start to see some of the cash prepayment being used?
So let me -- if you look at the current portion of the prepayments, you can see right from the balance sheet what the amount would be that we refund over the next year, which is a bit north of EUR 50 million. And on the other hand, we will see an additional inflow of prepayments. The amount of what we know, I think is nothing that we would want to disclose in detail.
Right, okay. And when you look at the LTAs you're putting in place, to what extent are you starting to see this from increasing use of LTAs from Chinese -- new Chinese semiconductor customers?
Yes. We can talk about contract situation in general. We can talk about LTAs in general. But LTAs by region, specifically by country, no. We don't disclose details at a customer level. Sorry.
Okay. But when working with the Chinese customers, would you typically want to work on an LTA basis to get visibility?
There are not that many existing users of wafers in China. China is still relatively small. The ambitions are very high and the investments they are doing is very high.
And then just 2 quick financial questions. One is on the EUR 8 million of sales that went from sea to airfreight. What is the margin impact or relative profitability of that EUR 8 million? Is there anything that we should be aware of?
No. Nothing that you should be aware of. It's just like selling additional volume. Nothing else. It reduces inventory and it goes with the revenue and the COGs that were associated with the manufacturing. There is no impact on margin.
Okay. And the last question, just on the guidance -- the 40% EBITDA margin guidance. Given the 9-month stage, you're basically there and from what you said pricing is good. Is there anything we should be aware of that would mean that you would not be -- I'm sort of curious why you didn't increase the guidance and therefore, wondering whether there's anything about cost to do with the launch of the new fab? Or anything else that we should understand that would have prevented you from increasing that 40% guidance at this stage.
Approximately 40 means that it's close to 40, either from the upper side or from the lower side. And with a slight increase that we made on revenue, we didn't see a necessity to change the outlook on the EBITDA.
Okay. But there's no extra cost or something coming in with preparation for the ramp in early Q1, for example?
No. This question is very precise. The answer is no.
Our next question comes from Mr. Paul Leming, Water Street Capital.
I just wanted to ask about a number in your slide presentation this morning. You show a forecast from IHS that the overall wafer market would be up, I think at 6.8% year-to-date. And if I back out the actual shipments through the first 9 months, the implication is that fourth quarter industry shipments will be down 14%. I am assuming that, that's just kind of a function of an old -- older-dated forecast and not really reflective of what you see going on in the marketplace today. But would love any color you could give us on that.
I do not see any reason to disagree with you.
It is a dated forecast?
Right.
We received another question from Mr. Guenther Hollfelder, Baader Bank.
Just one question on -- there's a large semiconductor maker out there with shortages on the 40-nanometer process node. And I was just wondering without, of course, saying the name of this semiconductor maker, whether you can comment in such a situation, do you see increased demand for wafers from these customers right now like compared to 6 months ago with the forecast? And also relating potentially also the next quarters that this is a situation where wafer demand is increasing and higher than previously expected?
You don't want to name the customer but you want to know what our situation at the customer is. I don't get it because I don't know which customer you are talking about. And we do not disclose details by customer. But let me try to give you an answer answering it indirectly. If whatever player on our customer side, there is a certain schedule for advanced node utilization, and he's behind that schedule and he wants to fulfill his obligation to his end users, then typically, this means for a certain period of time more wafer demand.
There are no more questions. [Operator Instructions] We received another question from Mr. Thomas Angermann, UBS.
Just a last question. If -- when is the latest point in time, you have to decide about your greenfield? And at the moment, I see -- look at your orders and therefore, you want to fulfill these orders. But orders filled and talked about decline in demand, which might come, might lead to a decline in your order intake going forward -- or might not. That is to be seen. But what is the flexibility? How much or how far you can wait until you decide about these greenfield?
Well, to keep it very simple, all the obligations that we have to date with customer, short term, mid-term and long term can be fulfilled without Greenfield.
Yes. Okay. So but that would mean if we see this weakness coming, you can push out first and wait maybe for another year. Then you said before in the call, the current brownfield will be used -- fully used by end of 2020. And of course, greenfield cannot be done overnight or...
Greenfield cannot be done overnight. It's more than 3 years between decision point and first wafers out. And by the way, I want to remind you, Siltronic did not decide anything on greenfield. So we cannot talk about pushing out greenfield. There's nothing to push because we didn't decide.
No, that was the question. How far you can push this decision? I know that you haven't decided yet. But if you say your capacity is running out by 2020 so this decision must be on your agenda.
The future is always on our agenda, but like we said in the past, we need to have additional LTAs and orders to justify greenfield investment. These LTAs must be really long terms, which means minimum 3 years, preferably 5 years with fixed prices and to basically cover a certain percentage of the rent. And the prices must be higher than the prices we see today. We are not yet in that situation. Therefore, we won't take the decision for certain period.
[Operator Instructions] We received another question from Mr. Veysel Taze, ODDO BHF.
Veysel Taze from ODDO. Just briefly as I joined the call really very late, what were your comments on the ASP. Did it increase in Q3? And what is your gut feeling going into Q4? Do you see still rooms for ASPs or with the signs of the slowdown in semiconductors, do you think now it's a period of further flat ASPs?
Well, first of all we saw ASP increase in Q3 compared to Q2. Secondly, we see an environment where further price increase are possible?
Okay. And then second one. From the earnings release, from some of your customers, particularly auto, industrial, so what we hear that the slowdown or some indications from customer side has arrived over the last 2, 3 weeks, where they think there'll be a type of inventory correction or customer ordering behavior has changed very sharply over the last weeks. And I'm just -- what is your experience from the past cycles in your business? How fast can things go down in the sense that people might delay their agreements, might start to talk about prices again, renegotiating them? So what is the experience from the past cycle you might share with us?
The experience from the past are not bringing a good picture for the future because if you take the worst case, you take Lehman crisis, and Lehman crisis within 1 week significantly. But is that's the picture that you'll see for the future? We might see a slowdown in growth. So we are still in a growth mode and there might see a slowdown in growth. That is also what customers are saying so...
Let's say, we play...
For the future, it's not fruitful to refer to cyclicality of the past.
Okay. Let's say we take the 2015 when we had the slowdown in the auto industry and then for 2, 3 quarters, there was inventory type of correction. How was the behavior of your customers back then? What was your visibility? I mean, did then, things move very fast with prices? People cutting their capacities? Or did you get like a little bit of visibility?
First of all, the cyclicity in 2015 was there. You're right. It had nothing to do with the automotive industry. It was too high inventory in the logic and in the foundry environment and the automotive industry. Let's put it the other way around. One is the total share of automotive. It's around 17%. So let's assume it slows down by 10%, which is huge, and we have the effect on the wafering, you can calculate, at 1.7%. Does this have any influence on the balance between demand and capacity? No, it doesn't.
Great. One final one, if I may. The price increase, what I sense from your statement, it was not that aggressive in Q3 like in Q2. Would you say there's more on the 200-millimeter wafer where you can get still price increases or 300-millimeter at this stage?
Nice try. We had a similar question before, and I said we do not disclose pricing development by diameter. I'm sorry for that.
There are no further questions. I will hand over now to the speakers.
We all thank you for participating in today's call. Our next release will be on January 31 when we publish our preliminary financial figures for the full year 2018. We will sign off the call, and wish you all a good day. Goodbye.
Ladies and gentlemen, thank you for your attendance. This call has been concluded. You may disconnect.