Earnings Call: Q4 2014

Jan 28, 2015

Summary

FY2014 revenue grew 13.1% to NT$140bn, while Q4 foundry growth included a one-time Fujitsu license fee. 2015 plans emphasized capacity expansion, a growing 28nm contribution, and margin management amid higher depreciation.

Welcome everyone to UMC's 2014 fourth quarter earnings conference call. All lines have been placed on mute to prevent background noise. After the presentation, there will be a question and answer session. Please follow the instructions given at that time if you would like to ask a question. For your information, this conference call is now being broadcasted live over the internet. Webcast replay will be available within an hour after the conference is finished. Please visit our website, www.umc.com, under the investor relations, investor event section. I would like to introduce Mr. Bowen Huang, Head of Investor Relations at UMC. Mr. Huang, you may begin. Thank you, and welcome to UMC's conference call for the fourth quarter of 2014. I am joined by Mr. Po-Wen Yen, the CEO of UMC, and Mr. Chi-Tung Liu, the CFO of UMC. In a moment, we will hear our CFO present the fourth quarter financial results, followed by our CEO's key message to address UMC's focus and first quarter guidance. Once our CEO and CFO complete their remarks, there will be a Q&A session. UMC's quarterly financial reports are available at our website, www.umc.com, under the investors, financial section. During this conference, we may make forward-looking statements based on management's current expectations and beliefs. These forward-looking statements are subject to a number of risks and uncertainties that could cause actual results to differ materially, including the risk that may be beyond the company's control. For this risk, please refer to UMC's filing with the SEC in the U.S. and our ROC securities authorities. I would now like to introduce UMC's CFO, Mr. Chi-Tung Liu, to discuss UMC's fourth quarter 2014 business results. Thank you, Bowen. I would now like to go through the 4Q 2014 investor conference presentation material, which can be downloaded from our website. Starting on page 3, the fourth quarter of 2014 consolidated revenue was NT$37.24 billion, with gross margins at 27.4% and operating margin at 12.2%. The net income attributable to the stockholders of parent was NT$4.56 billion, and earnings per ordinary shares were NT$0.36. On page 3, you can see our operating revenue deliver good solid growth compared to both last quarter and same quarter of last year. Utilization rate in the fourth quarter of 2014 was around 93%. Our total wafer shipment is about 1.431 million inch equivalent wafers. On page 4 is our comprehensive income statement. Revenue grew 5.7% on back of a one time license fee income from Fujitsu. As a result, our gross margin also reached 27.4% to about NT$10.18 billion. Net income in Q4 was NT$4.46 billion, translating to about $0.36 EPS. For the full year result, our revenue reached NT$140 billion, represent a 13.1% year-over-year growth. Gross profit was NT$31.8 billion, and operating income was around NT$10 billion. EPS for the full year at $0.97 per share. Our balance sheet is on page 6. Our cash position is still around NT$45 billion. The total asset is still more than $10 billion U.S. dollars. Our stockholder equity is around NT$225 billion. On page 7, there is a breakdown between our foundry business and new business for Q4 of 2014. For foundry business, the net income is around NT$4.6 billion, and for the new business, the loss was around NT$541 million. The net result is about 4.46, as I mentioned earlier. In terms of foundry ASP in Q4 last year, it was flattish compared to the previous quarter. For market breakdown, Asia represent 42% of our total revenue, and Europe is around 8%, and North America is 45%. For the full year, the change is quite minimum. Asia is around 44% compared to 43% last year, and Europe down to 6% from 8% in 2013. IDM represent roughly 10% of our total revenue, and fabless is the remaining 90%. For the full year, it is about the same picture, 91% in fabless. For segment breakdown, we have almost identical breakdown for Q3 and Q4, with communication representing the largest slice, around 54%. For the full year, it is the same message, around 51% coming from communication and 29% coming from consumer segment. Again, we are happy to see our 28 nanometer now has grown to 7% of our total revenue in Q4 last year, compared to 3% in Q4 2014. 40 nanometer also grow to 28%, compared to 27% in the previous quarter. For the full year, 28 nanometer represent about 3% of the full year revenue, and 40 nanometer is 21%, and 65 is another 28%. For Q4 2014, our total quarterly wafer capacity was around 1.577 million 8-inch equivalent. 12I in Singapore, we have some technology upgrade from 65 nanometer to 40 nanometer. So that caused a temporary drop in terms of available capacity. We will see the capacity number to be restored over the next quarter or two. For Q1 2015, we expect the 12A, our Tainan most advanced facility, continue to add new capacity. Yet, we also factor in some of the shorter working days due to Chinese New Year holidays in Q1 of 2015. So the number is 1.58 million. For CapEx, our CEO will give a more detailed guideline later on, but it is estimated about $1.8 billion for 2015 compared to $1.4 billion in 2014. So that is the summary of UMC results for Q4 2014 as well as full year 2014. More details are available in the report, which has been posted on our website. I will turn the call now to Mr. Yen, CEO of UMC. Okay. Thank you, Qidong. Hello, everyone. I would like to update everyone UMC's fourth quarter operating results. In the fourth quarter, foundry revenue grew 3.7% sequentially to TWD 34.74 billion. This figure includes a one-time 40 nanometer licensing fee from Fujitsu, lifting gross and operating margins to 30.2% and 14.6% respectively. The overall capacity utilization rate remains at 93%, bringing shipments to 1.43 million 8-inch equivalent wafers. Our 28 nanometer technologies represented 7% of our foundry revenue in fourth quarter 2014. The shipments from our 28 nanometer gate-last high-k metal gate process exceeded that of 28 nanometer polysilicon wafers. Excluding the Fujitsu 40 nanometer licensing fee, UMC's 2014 foundry operating profit grew 74% from the previous year. This rise in profitability was mainly driven by double-digit percentage growth in wafer shipments, and partly due to the rapid production ramp of 28 nanometer, which accounted for 3% of total 2014 revenue. Our collaborative technology efforts with our partners will enable UMC to deliver additional manufacturing solutions to fulfill new product specs, strengthening our position in the IC supply chain to take advantage of the continued momentum in market demand. The Taiwan government authorities recently approved UMC's application to invest in a 12-inch joint venture fab in Xiamen, China. This investment will create opportunities for UMC to benefit from China's enormous chip requirements by bringing us closer to the Chinese semiconductor supply chain. The cooperation highlights UMC's differentiated approach of global expansion, proven through Singapore's Fab12i and Suzhou China Fab8n successfully achieving economy of scale via mitigating customers' risks via geographic diversification. While we expand our production sites worldwide, we are also focusing on continuous organic growth by deploying additional capacity in our Tainan sites. As such, we will budget 2015 CapEx for approximately $1.8 billion. Our 2015 CapEx illustrates our strong commitment to meet customers' requirements and gain additional market share through efficient execution and strategic alliances. UMC's global expansion efforts, driven by manufacturing excellence, will strengthen customer services with increased operating scale to enhance corporate financial earnings and deliver long-term returns to our shareholders. Now allow me some time to summarize the recent highlights in Chinese。接着向各位说明公司现阶段营运重点。本公司2014年第四季晶圆专供营收较上一季成长了3.7%，达到新台币347.4亿元，其中包括来自富士通半导体一次性支付的40奈米权利金。使毛利率与营业利益率分别推升至30.2%和14.6%。整体产能利用率则维持在93%，出货量约当8寸晶圆143.1万片。第四季联电28奈米制程占晶圆专供营收比为7%。其中后闸极，也就是Gate Last High-K Metal Gate制程的出货量已经超越了Polysilicon的出货量。2014年本公司晶圆专供营业利益，富士通半导体40奈米权利金较2013年大幅提升了74%。成长动能主要来自于达到两位数百分比成长的晶圆出货量，部分归因于28奈米的快速上量。28奈米占2014年全年总销售额的3%。随着市场需求持续攀升，我们与合作伙伴将携手投入更多的制造解决方案，来满足新产品的规格需求，以进一步巩固联电在晶片供应链的地位。联电董事会于去年10月通过于厦门进行12寸晶圆厂合资决议，此申请近期已获得中华人民共和国主管部门的核准。这将提供我们进入中国半导体供应链的机会，进而获益于中国庞大的晶片需求市场。联电全球布局的成功经验已表现在新加坡Fab12i与中国苏州的Fab8n两个达到经济规模的晶圆厂。于厦门的合资案也将持续我们在拓展全球制造布局时的差异化策略，经由分散设厂的地理位置来降低客户的风险。在拓展海外生产基地的同时，我们也投入资本支出，扩充台南的厂区的产能，以达到持续的成长。本公司2015年资本支出预算约为美金18亿元。我们将以高效率的执行以及策略联盟，来展现我们对满足客户需求与提高市场占有率的坚定承诺。联电全球扩展的努力，将结合制造的优势以深化客户服务，在整体营运规模扩展下，提升公司的获利能力，提供我们股东长期的投资报酬率。I have finished my remarks, and now let me go over the first quarter 2015 guidance。The foundry segment wafer shipments to show an increase of 2 to 3%. The foundry segment ASP in US dollar will increase by approximately 3%. The UMC foundry segment gross profit margin will be in mid 20% range. The capacity utilization rate for foundry segment will be approximately 90%. 2015 CapEx for foundry segment will be $1.8 billion. The guidance for new business segment's revenue to be approximately TWD 2 billion, and the net loss attributable to UMC parent company to be approximately TWD 170 million. That concludes my comments. We are now ready for questions. Operator, please open the lines up. Thanks. Thank you, ladies and gentlemen. We will now begin our question and answer session. If you have a question for any of today's speakers, please press 01 on your telephone keypad and you will enter the queue. After you are announced, please ask your question. If you find that your question has been answered before it is your turn to speak, please press 02 to cancel the question. The first question comes from Randy Abrams, Credit Suisse. Please ask your question. Okay, thank you. Good result and outlook. The first question on the first quarter guidance, if you could go into what is driving the strength for shipments to grow 2%-3% normally in a low season, also particularly accounting for TSMC also guiding pretty well for first quarter. Yes. Our first quarter 2015 by application, the communication segment will be the strongest one, the computer follow the second, the consumer will be kind of slow. In the communication segment, the Wi-Fi, baseband, and ISP, a small driver interface, display driver interface is going up. The microcontroller unit is going down. Okay. If I could follow up on your 28 nanometers now seeing a stronger ramp-up, we are seeing the other foundries like TSMC adding capacity. How are you viewing the landscape for 28, just from pricing and margins? Do you still see to reach the 12-inch average in second half, just how profitability and pricing looks for 28 as you ramp up? Yes. Our 28 nanometer profitability has improved due to a higher product yield with an enhanced scale. On the other side, we foresee the 12-inch overall loading will also improve. So the 12-inch corporate average profitability will rise, which are pulling slightly ahead of our 28 nanometers contribution. Therefore, our 28 nanometer will still have a small margin dilutive effect in 2015. Okay. With 12-inch utilization rising, could you give color on 40 and 65, like the more mature nodes, what you're seeing, if you're still seeing growth from 40 and 65 as you also ramp up 28? Yes. We are expecting our 12-inch mature node, including 65 and 90 nanometer loading, will be improved in the second half 2015. So that will help to improve our overall revenue. Okay. The last question, I have one to ask on the CapEx, the $1.8 billion. Could you go through how much 28 nanometer capacity you now plan, and then also an update on the 8-inch expansion? Thank you. Yeah. Our 28 nanometer capacity expansion plan is by the second quarter 2015, we will reach 20.5K per month. We have additional capacity expansion plan at phase 5, our Tainan Fab, at the end of 2015 or around early of 2016. Okay. For 8-inch? That will be additional 9,000 wafer per month. Okay. So 9,000, so it would go to 29.5 if you do the second phase. Yes. Okay. Okay, then your plan for 8-inch capacity? Our current expansion for 8-inch is only available at our Chinese site in Suzhou Hejian, and they are expanding from previously, I recall, was around 45,000 wafer per month and gradually to over 50,000. I think they ultimately will go up to 61,000 per month, and that will happen earlier this year. Okay. Thanks a lot. Good result. Thank you. The next question comes from Bill Lu, Morgan Stanley. Go ahead, please. Yeah. Hi. Good result and good ramp on 28 nanometers. Congratulations. Just a couple of more questions on 28. I wasn't quite clear on Randy's question. If you look at the $1.8 billion in CapEx this year, does that get you to the 20.5K per month, or does that include the additional 9K by the end of the year? Bill, this is a cash-based CapEx, and you also need to factor in the $1.4 billion we spent in 2014. So we already had about 12K to 15K in production almost right now. So the additional 8K to reach 20.5K, most of the spending is done in 2014. The $1.8 billion also, a big portion will go for this 9K expansion at the end of 2015 and early 2016. Got it. Okay. That's very clear. So you're doubling your capacity at 28. Should I read that as you're now feeling more confident that you have more customers coming in at 28, or the existing customers ramping? Can you help me with that? Yes. After we entering the 28 nanometer production in early 2014, we actually demonstrated pretty good yield learning rate on both poly-SiON and high-k metal gate technologies. Based on our customer engagement, we are gaining more traction on 28 nanometer business. We are pretty confident on our 28 nanometer business growth going forward. Can you help me with how many customers you have now? How many you would need to get to by the end of the year to have the capacity full? On 28 nanometer, we have over 20 customers engaged and more than 16, 20 nanometer CapEx. Okay. What is going to be the percentage of revenues from 28 in the first quarter? It will be a high single digit in first quarter 2015. Okay, great. Sorry, just two other quick ones. One is there going to be an impact to dividend given the higher CapEx? And two is how we model depreciation. Thank you. The amount of depreciation will grow around 15%-20% year-over-year. So higher growth rate compared to the last few years. We certainly hope we won't impact our dividend payout. That won't be discussed until the board meeting later of this quarter. Great. Thank you very much. That's it for me. The next question comes from Gokul Hariharan, JP Morgan. Go ahead, please. Yeah. Hi. Great result, and thanks for taking my question. First of all, could you give us an update in terms of your China plans? When should we expect some 40 nanometer capacity being built in Xiamen? Secondly, also, beyond 28, when would you start ramping up some 40 nanometer capacity in Taiwan? Is that contingent for you to have 20 nanometer in China? That would be my first question. Thanks. Well, the 40 nanometer and also 55 nanometer capacity plan for Xiamen project is really aiming to penetrate into the domestic market in China. That's not really tied into our 40 nanometer plan. Our CEO will give you update on our 40 nanometer development later. As for the Xiamen project, we received the official approval from the Taiwan government, and we also inject the first $100 million yesterday from HeJian for this new project. The plan is really to see early production by the end of 2016, and expanding to maybe 20,000 wafer per month, covering both 40 nanometer and 55 nanometer by the end of 2017. So about three years away to reach 20,000. Okay. Okay. I'll need to update our 14 technology. Our 14 FinFET technology development is on track. We are expecting the process will be ready for customers tape out in the second quarter this year. Okay. When we talk about the Xiamen facility, is that compatible with the eventual conversion to 28, as you're building the capacity? And, is that something in your plan that once you have some 14 nanometer ramping up, you could start moving to 20 nanometer in Xiamen pretty much concurrently, or is that completely subject to government approval? Of course, it's completely subject to Taiwan government's approval. However, as I mentioned earlier, this 20,000 40/55 nanometer capacity is really aiming at specific domestic Chinese market for their specific products. And, that's really the goal, to penetrate into the local supply chain, if you will. So I think at this point, as I mentioned, it takes three years to reach 20,000 wafer capacity, so it's still a long way to go. Okay, got it. Thanks very much. Thank you. The next question comes from Daniel Heyler, Bank of America Merrill Lynch. Please ask your question. Yeah, thanks very much. Ji-Tong, could you tell us what the gross profit margin was, excluding the Fujitsu license fee that came in from Fujitsu in the fourth quarter, please? Q4 was 26.8%. Okay, great. Okay. Is there any more of that coming through this year, or is that done? That was one-off. Our Xiamen plant, for example, 40/55, will also be licensing technology from UMC. So, we will get also licensing fee income from our Xiamen project. We do not know when that is going to happen. Okay. That will happen, I presume, when the revenues start coming in. We do not know yet. Yeah. I do not know. Okay. Then you had mentioned that you have got 20 customers. Dr. Yen had mentioned 20 customers on 20 nanometer and 60 tape-outs. I wanted to reconcile that with the high single-digit percentage of sales. So it seems like you have got a lot of small customers or maybe explain how many customers are actually in production. Because it seems like a small contribution of revenue for how many customers you have. Yes. It's very diversified in terms of the scale. It's very diversified products we have engaged. It normally would take at least several quarters after the product tape out to enter into- Okay. That makes sense. Yes. Okay. You've had, I think, maybe one important partner. How many kind of partners are you in production with now at this point, in volume? It's more than five. Okay. Excellent. I wanted to maybe get a little bit more color on the 65 and 40 challenges there, as you talked about last quarter. I was wondering maybe if you could elaborate on some of the dynamics there. Why isn't utilization at that nodes higher? What's it going to take to get those fabs filled up? Yeah, we have engaged some new application nodes. For example, the 55 embedded high voltage for the LCD driver and the 90 nanometer embedded flash with some customers, and that is the main area we expect to grow our mature nodes, 12-inch mature nodes utilization rate. Okay. Could you give a sense of timing there? Because I think you said earlier that you didn't think you could maybe achieve higher utilization there until second half of this year. I'm wondering why it's taking that long, because I guess it seems as though your competitors' fabs are pretty full on the mature nodes. To take share in those nodes, do you need to be maybe more aggressive in pricing, or you need broader technology? I'm just wondering why you're not taking more share there. Because typically those nodes are more price sensitive, and you would be able to take share back if you're aggressive on pricing. I'm wondering why you're not able to attack those markets with more aggressive pricing or something to get the utilization up on the 40 and 65. Yeah. It's actually not a pricing issue. It's affected by the second half of last year because some applications did not take off as we expected. Right. But this year, we are more confident on gradually improve the loading of the 65 nanometer, starting in Q2, going down to the end of this year. Okay. Are you anticipating which of the small panel display driver market, which is moving increasingly to those nodes, because there's, I guess, Powerchip is coming in with their 12-inch fab on those products. I guess TSMC has quite a bit there in those nodes. Maybe explain for us what you're anticipating in the mobile and the display IC market for this year. Yes, we are anticipating, especially the small panel display driver, that the demand is very strong. Also some applications, they are migrating to the higher-end technologies. So we are pretty optimistic on that. Okay, great. Qidong, could you give us a- I know you've given us some of the details on the individual fab basis, because I know you've got different issues on fabs. You're bringing off some capacity and bringing it back on. What's the full-year capacity growth? Do you have that number in front of you? Yes. The full-year overall capacity growth is around 5% year-over-year. Excellent. Okay. Thank you, gentlemen. Congratulations on the results. Thank you. The next one is from Steven Pelayo, HSBC. Go ahead, please. Yeah. If I could just quickly follow up on Dan's before my question. The capacity growth of 5% year-on-year, I am curious, how does that break out? What is 300 millimeter versus 200 millimeter capacity growth? The 300 millimeters growth will be accountable for 7% growth, and the 8-inch will account for 3% growth. 3% year-on-year. Year-over-year. Okay. First of all, I guess I just want to make sure from an accounting perspective, the Fujitsu license was $50 million USD. We exclude that from the top line and the gross profit line, and that comes out to a foundry gross margin of 27%, excluding licenses. Do I understand that correctly? If you are using the consolidated number of basis, you also have to take out the new business revenue as well. Sure. Yeah, I was taking that as well. Okay. Just wanted to make sure I got that correct. Your guidance for next quarter is for mid-20s gross margins for the foundry-only business. I guess I am trying to understand, you just did 27% excluding the license. Are we guiding the gross margins down quarter-on-quarter despite higher ASPs? Can you give a little bit more specific on your guidance for the first quarter foundry-only gross margins? The guidance is mid-20, so it is a range. At this point, we cannot really say it is up or down, but it is mid-20. To answer your question, yes, we do have higher depreciation expenses for the year of 2015. As I mentioned, we are looking for around 15%-20% increase in depreciation for the whole year. For quarter one alone, around 5% increase sequentially. 5% quarter-on-quarter. Okay. Can you comment a little bit, I think you talked about having 12K to 15K in production of 28 nanometer, going to get it to over 20K by mid-year. Is this pretty much that 12K to 15K fully loaded today? Or what is your utilization rate at 20 nanometer? We are currently around 70%. Around 70%. Thank you very much. Thank you. The next question comes from Donald Lu, Goldman Sachs. Go ahead, please. Oh, hi. My first question, just to follow up Steve's question, did you say your 28 nanometer capacity utilization is 70%? 75%? Yes. 75%. Okay, great. Thank you. Just following up on that, if you do a quick math that based on your 28 nanometer revenue, your capacity is at 12K to 15K according to Qidong, and 75% utilization rate, then we get to ASP, it's not very high. As our CEO mentioned, currently, still some dilutive to our overall earnings if you compare to our 12-inch average, and this is largely due to the early ramp of the production. That's a function of yield. Do you expect that to improve? Yes, correct. My next question is on the Q1 guidance. The wafer shipment is up 2%-3%. Capacity is up a little bit, but then you said utilization is going to decline. It seems like utilization cannot decline if you increase your wafer shipment in Q1. Yeah, if you go to the first page of our handout, you will find out that last quarter, our shipment actually declined sequentially with the same utilization rate. There was some kind of pre-build of wafers or wafer in the pipeline in Q4. Oh, I see. I think it's a little bit larger than the normal volume in Q4 last year, which will be shipped in Q1 this year. Okay. So that was a little dilutive to your margin as well. Not necessarily, but it will distort the capacity utilization rate figures, for sure. Okay. Got it. What's the guidance for OPEX this year? OPEX this year will see a percentage of revenue to come down a little bit. The absolute dollar amount may stay flat. Oh, flat only. The total OPEX. Yeah, will be flattish for the full year because there's a beneficial factor that our IBM license fee has completed by the end of 2014. Okay. Got it. Would that be pretty linear throughout the year, like Q1, Q2, Q3, Q4? Well, it depends on our R&D project. It is hard to say yet. But the full year number, as I mentioned, is going to be somewhat flattish compared to last year. Okay. Got it. Thank you very much. Great result. Thank you. Next question comes from Andrew Lu, Barclays Capital. Please ask your question. Yes, Dr. Yang and Qidong, a couple questions. Do you have, this year, revenue growth target? Something, a range, will be useful too. Yeah, we cannot comment on that. Okay. Yeah, but we. Sorry. Any specific reason? But we do have some. Sorry. Yeah, we do have some outlook. Are we looking for double-digit growth as TSMC guided industry growth of about 12%? The foundry outlook is, according to our information, it's around 10%-15% range growth. For UMC, we are optimist the growth on 2015. Yeah. But we don't comment on the exact numbers. Thank you. In terms of CapEx of $1.8 billion, I did not know anyone asked, which you answered earlier, what's the breakdown of the CapEx? Is that all 100% 28 nanometers, or there's some CapEx going to the 14 FinFET as well? I can only comment on that the majority part, around 90%, is on 12-inch capacity expansion. 90% on 12-inch capacity. Thank you. Do you see the competition from SMIC on 28nm poly and high-k metal gate, because their schedule is quite aggressive as well. By the end of this year, they likely to have 10K-15K. So it is just one year behind you. Sorry, we do not comment on our competitors. However, we are still pretty confident on our technology readiness and manufacturing efforts on both poly-SiON and high-k metal gate versions. Based on our customer engagement, the demand is still pretty strong on that. Thank you. The capacity you have on 28, you already mentioned still dilution for this year, gross margin from 28 nanometer. What kind of monthly output to get the corporate average margin? I don't quite understand the monthly output, please. Yes, because Okay. Actually, there are some other factors that will affect our 28 nanometer dilutive effect. As I mentioned just earlier, the overall capacity utilization rate on 12-inch, that will have some contribution. As we foresee, our overall 12-inch mature node utilization will be improved in the second half of this year. So that will lift our 12-inch corporate average profitability. So that will still kind of keep a small margin dilutive effect for our 28 nanometer profitability. Thank you. It will affect also by the utilized rate of 12-inch mature nodes. Do we have a revenue target for percentage 28 by the end of this year? Yes. By the end of this year, we are targeting around 15%-20% of total revenue from 28 nanometer. 28, 15%-20%. Okay. The last one I have is solar business. I recorded some news indicate you sell some of your solar business to other firm. Is that already factored into the first quarter's guidance? It's already factored in to our new business guidance in Q1. But the actual merger date is July 1st. Oh. After July 1st, can we assume that revenue and the losses will be gone because, the ownership on this new solar business actually is a book holding, right? Is less than 20%. Yes. Okay. Thank you. Thanks. Thank you. And the next one is from Szeho Ng, BNP. Please ask your question. Hi, gentlemen. I just want to ask more question on the new business. Should we assume all the new business contribution, top line, and also the cost will be gone after July this year? Is it what you mean? No. Only the Topcell, the silicon-based solar business we had was merged to Motech. Right. And that were effective on July 1st. Okay. There were some accounting treatment starting become effective on this year already, January 1st, which is, we still book some of the top line and bottom-line losses by excluding some fixed cost depreciation. It is a little bit complicated. We can walk you through offline. Oh, okay. All right. But any rough indication about the revenue level once we take off the Topcell part? Yeah. Majority today is actually Topcell, so it is going to be quite minimum after July 1st. But the operating loss will be probably staying around $170 million. Yes, significantly reduced. Oh, similar to the Q1 guidance you mentioned, right? Well, it depends on the market condition at that time, but at least there will be no loss contributed by Topcell. Topcell. Okay. All right. Okay, second question. What's the exchange rate you used for Q4? 30.89. 30.89. Okay. All right. Okay, thank you very much. Great result. Thank you. Thank you. The next one comes from Michael Ngo, Thomson Reuters. Go ahead, please. Hi. Thank you for taking my question. I guess the first thing I would like to ask about is the China plant, one more time. I am just curious if you might be able to talk about how much of your CapEx growth will be in that plant or just in China in general, if you could talk about that. Thanks. Our CapEx growth of $1.8 billion did not include any of the Chinese spending. Currently, the Chinese spending is treated as an independent project. It is not in the group CapEx yet. Okay. Can you mention any sort of CapEx plan for that this year? Well, it takes about 18 months for the construction, and it will only cost minimum amount of money for building the shell. This year, the number will be quite small. Quite small. Okay. Do you have any just sort of overall goal for sort of percent of your overall capacity to come from your China operations? Well, the design capacity for the China or the Xiamen project is 50,000 wafer per month. But the more foreseeable plan is to reach 20,000 per month first by the end of 2017. Okay. Okay. All right. Thank you. Thanks. Thanks. The next one is from Donnie, Nomura. Go ahead, please. Hey. My first question is that the foundry gross margin was 27% in 4Q. May I ask how many percentage from the help of foreign exchange? I think we are in the same position as the other companies. Basically, we experience about 1.5%-2% NT dollar depreciation in Q4 alone, and every one percentage point of NT dollar depreciation will contribute about 0.4 to 0.5 percentage point enhancement in our gross margin. Probably you can do the math. Okay. Thank you. In terms of the good gross margin in foundry business last Q4, can we say that the 28 nanometer has already reached a positive operating margin in last 4Q? Well, we only guided that currently is about two percentage shy from our average 12-inch gross margin. Again, each quarter is very different. As our CEO mentioned, our 12-inch foreseeable loading is improving in the second half of this year, so, we don't really comment on that. Again, it's a small dilutive effect to our overall 12-inch operation. I see. So, you say that it's a 3% impact to gross margin? No, it's a few percentage point dilutive effect compared to our overall 12-inch operation. Okay. I see. So, in terms of the capacity, could you remind me that the 8-inch fab capacity will grow 7% year-over-year, and the 12-inch will grow 3% year-over-year? It's the opposite. Yeah. Oh, opposite. 12-inch grows 7%, and 8-inch grows 3% year-over-year. I see. What is the high-k metal gate and polysilicon capacity allocation? If you want to reach 20.5K per month in the first half and 30K wafer per month in the end of this year. In 2015, our high-k metal gate will take 12.5K. Pardon? Yeah, 12.5K among the 20.5K per month capacity. So high-k metal gate will be a major part. Okay, 12.5K, right? Yes. So the polysilicon will be 8K. I see. in the Q2. I see. How about the end of the year? It will be this kind of proportional down to the 30K, yeah. I see. The majority part will be still the high-k metal gate version. I see. In terms of the new business, for Q1 guidance, Topcell is still included in the consolidated financial statement, right? The profitability loss, it also includes Topcell? Yeah. As I mentioned, majority of the top line and bottom line is included all the way until July 1st. However, due to some accounting treatment, which is quite complicated, the fixed cost or the depreciation of the Topcell is excluded from the consolidation, starting from January 1st. Again, we can walk you through the details offline. Okay. My last question is: What is the FX guidance in Q1? Yeah, we don't take FX forecast, but we just follow the Bloomberg numbers. Okay, I see. Thank you. Thank you. The next question comes from Roland Shu, Citigroup. Go ahead, please. Thanks. To take my question, actually, most of my question have been asked and answered. I think one more question is, for our new business, are we going to invest more in the new business, such as gallium arsenide or other items? Do you have any plan on that? Well, first of all, we don't have any capacity expansion plan for solar related. However, we recently did some corporate restructure for gallium arsenide business. Not only we see the growth potential there, but also they will play a big role in transitioning our 6-inch wafer fab from CMOS base to eventually gallium arsenide base for the longer term to help us to phase out the 6-inch wafer plant in the longer term. That will be the key focus in terms of new business for the near future. Yeah. So how much you are going to spend to turn this 6-inch fab to do the gallium arsenide fab? Well, it's all existing facility with minimum upgrade, so it's not going to be much. And most of the CapEx will be supplied by their internal generate cash. Okay, so the investment of CapEx actually is not included into this $1.8 billion CapEx plan this year? No, that $1.8 billion is only for UMC Foundry. Okay, so does not include this. Yeah. What is your plan for the dividend this year? I mentioned, it is up to our board meeting sometimes in March or April. However, we will try to have a very stable, predictable dividend policy as we can. Okay, thanks. Last question is on your, given this $1.8 billion CapEx spending and also given the continuous investment in the Xiamen, and also with this potentially dividend payout. Are we considering to raise money this year? If you do, by what kind of method? Thank you. Well, each year we've been raising money from different instruments, but I think this year, we will continue to do so, but mostly will come from debt market. Okay, thank you. Thanks. Our next question comes from Rick Hsu, Daiwa Capital Markets. Please ask your question. Yeah. Hi, good evening and good results. Congratulations. Just a couple of questions here. I think, I probably missed the number here. Have you guys talked about your 28 nanometer capacity build by the end of this year? I know it's been 20.5 at mid this year. What about by the end of this year? Well, there will be additional 9,000 wafer capacity sometimes end of this year or early next year from our second phase, P5 facility. Okay. So I can fairly assume probably around 30K by the end of this year, right? Or early next year, yes. Or early next. Okay, good. Second question is, I would like you guys give me a bit more color about your gross margin trend throughout the whole year, because it seems to me that you talk about 10%-15% year-on-year increase in the overall foundry business, foundry market. I am not so sure how much you guys will grow in terms of your UMC Foundry's revenue this year. But you also mentioned about your depreciation is going to grow by 15%-20% year-on-year. That is much higher than last year. So I am wondering, it looks to me that your depreciation power this year is going to rise much faster than your total revenue growth. So how will that impact your overall margin trends throughout the whole year? Well, that is certainly a big challenge we face for 2015. But, again, if we cannot count on the top line, we can certainly not count on the bottom line. But we do see new business, I mean, the incremental revenue growth coming from our leading edge revenue. Hopefully, it will improve throughout the year curve improvement. We also have a plan for our mature 12-inch node utilization rate to go higher in the second half, which has been a drag for overall profitability. So altogether, there are positive and negative sectors for 2015, but we will try our best. Okay. Got you. Thank you. Okay, yeah, last question I have. Thank you so much. Thank you. Ladies and gentlemen, we are running out of time, so we are taking the last question. Our last question comes from Edward Ling, CIMB. Please ask your question. Hi, Ji Dong. Thanks for taking my question. My question is on the new business. Can you tell us what is the number attribute to the Topcell in 2014 in terms of the loss? Well, for example, for Q4, it was about TWD 500 million or so losses, and less than half come from Topcell. So yeah, possibly a little bit less than half come from Topcell. For the new business as a whole, should we assume that last year, the loss from Topcell is about 50%? Well, I can only have the Q4 number, but that may be the ballpark figures. Okay. Q1 top sale actually was profitable, I think Q1 or Q2 last year. I cannot remember exactly, but Q4, that was the case I mentioned. Thanks a lot. You mentioned that you are not going to invest in a solar business anymore, but will you further divest other solar operation? When I look at the balance sheet, I think there are still about three to four investments in solar related business. Will that be the direction you are going to continue to divest in solar business? Well, I did not say that we will not invest in solar business. I said we will not expand the solar capacity, but we will continue to invest for the sake of efficiency and profitability competitiveness enhancement. So we will continue to invest, but not for capacity alone. As for divestment plan, we do not really have any concrete plan. But again, anything can benefit our competitiveness or efficiency of our solar plan, we will do it. Okay, thanks a lot for the clarification. That is all the question I have. Thanks a lot. Thank you. Thank you for all your questions. That concludes today's Q&A session. I will turn things over to UMC Head of Investor Relations for closing remarks. Thank you everyone for joining us. We appreciate your questions today. If you have any additional follow-up questions, please feel free to contact UMC at ir@umc.com. Have a good day. Bye-bye. Ladies and gentlemen, that concludes our conference for fourth quarter 2014. Thank you for your participation in UMC's conference. There will be a webcast replay within an hour. Please visit www.umc.com under the Investor Relations, Investor Events section. You may now disconnect. Goodbye.