UMC 联华电子

Operator

Welcome everyone to UMC's 2014 Second 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 Events section. And now, I would like to introduce Mr. Bowen Huang, Head of Investor Relations at UMC. Mr. Huang, you may begin.

Bowen Huang

Thank you and welcome to UMC's conference call for the second quarter of 2014. I am joined today by Mr. Po-Wen Yen, the CEO of UMC and Mr. Chi Tung Liu, the CFO of UMC. In a moment, we will see our CFO present the second quarter financial results; followed by our CEO's key message to address UMC's main area of focus and the UMC's third quarter, 2014 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 Investor Relations Financial Release 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 risks that may be beyond the company's control. For those risks please refer to UMC's filings with the SEC in the U.S. and the ROC securities authorities. I would now like to introduce UMC's CFO, Mr. Chi Tung Liu to discuss UMC's second quarter 2014 business results.

Chi Tung Liu

Thank you, Bowen. I would like to go through the 2Q, 2014 Investor Conference presentation material, which can be downloaded from 2Q, 2014 quarterly results from our website. Turning on Page 3, the second quarter of 2014; revenue was NT$35.87 billion with gross margin around 22.9% and operating margin at 8.1%. The net income attributable to the Stockholder of the Parent was NT$3.48 billion and earning ordinary shares were NT$0.28. Loading in Q2, 2014 was around 90% and the wafer shipment is NT$1.42 million 8-inch equivalent. Please turn to Page 4 and here's our comprehensive or consolidated income statement. On consolidated basis, revenue growth 13.2% quarter-over-quarter to NT$35.8 billion. Gross margin was 22.9% or NT$8.2 billion and operating income was NT$2.9 billion or 8.1% operating margin and EPS was 0.28%. Page 5, you can see our first six months year-over-year comparison. Again revenue grow 13.2% to NT$67.5 billion and gross margin grow by 32% to NT$14.1 billion and operating income grow 167% to NT$3.85 billion or 5.7% and the total net income attributable to Stockholder of the Parent for the first six month of the year is NT$4.6 billion or 6.9%. EPS in the first half is NT$0.37 per share. On Page 6, is our abbreviate [ph] balance sheet. Cash remain at healthier level around NT$50 billion or NT$49.6 billion and our total asset is around $10 million and Stockholder Equity is NT$215 billion. On next Page, our operating segment details. Here we offer breakdown between our foundry business and also new business which focused on Solars. In quarter two, as you can tell; among the NT$35.8 billion consolidated revenue NT$32.5 billion coming from our foundry wafer business was net income around NT$3.5 billion and on the other hand, our new business contributed around NT$3.3 billion in revenue and net loss was around NT$491 million in second quarter of 2014. So following pages, we will focus on our foundry operations and from a pricing point of view on Page 8, in second quarter our ASP remains somewhat flattish, which is in line with our previous guidance. On Page 9, here we offer further revenue breakdown by countries or by geographic breakdown. Europe was 5%, which slightly come down from 7% in the previous quarter and Japan went up through 6% from 3% in Q1. In Asia and US remain somewhat unchanged, represented 46% and 43% respectively. In quarter two, IDM go up to 10% from 8% in the previous quarter. In Page 11, you can see the application breakdown and communication as we indicated earlier. That was the strongest segment growth to 49% in quarter two of 2014 and computer remains somewhat unchanged. On Page 12, we are happy to provide revenue breakdown to 28-nanometer which represent about 1% of our revenue in quarter two, 2014. And total revenue below 40-nanometer is around 22%, which is a continue to show further improvement. On Page 13, we offer quarterly capacity breakdown and in quarter two, we continue to show some mild capacity growth. However, in the third quarter we reduced some capacity upgrade in Fab8D and they we've caused some capacity loss in Fab8D in the third quarter, so total net capacity in third quarter, will show minor decrease in the third quarter compared to the quarter two. And last Page, here we have prepared our pay price budget unchanged. It was somewhat moving toward the upper end of our previous range of $1.3 billion for 2014. And that's pretty much summarized my report and also UMC's result for quarter two, 2014. More details are available in the report, which has been posted on our website. I will now turn the call over to MR. Yen. CEO of UMC.

Po

Thank you, Chi Tung. Hello, everyone. I'd like to update everyone, UMC's second quarter operating results. In the second quarter of 2014, UMC recorded NT$32.57 billion in revenue from the foundry segment. With operating margin from foundry operations of 10.2%. Wafer shipments reached 1.426 million 8-inch equivalent wafers. Our 28-nanometer business represented 1% of revenue, while 40-nanometer accounted for 21%. The second quarter foundry revenue grew 13.4% sequentially, fueled by strong communication segment demand that lifted capacity utilization to 90%. The market conditions also reflected a turnaround, with rising demand for portable computing devices, helping to drive UMC's 28-nanometer shipments. We project continued 28-nanometer revenue contribution growth in third quarter, 2014. As a result of the sustained demand for mobile and tablet computing. We are optimistic in the long-term that our 28-nanometer production ramp will strengthen our overall product mix and generate abundant opportunities for UMC to win additional foundry market share. For other technologies, we have strengthened the IP portfolio for our 55-nanometer Low Power platform to address low power and wireless applications by offering Kilopass' Gusto and Cypress's SONOS embedded flash memories. These enhanced IP solutions will further diversify our manufacturing offerings to fulfill a broader range of specialty technology products on our proven and versatile 55 Low Power platform. Designers in wearable and Internet of Things (IoT) applications have already realized product tape-outs using UMC's 55-nanometer low power technologies and IP. We expect the semiconductor demand strength to continue into the third quarter of 2014. For 28-nanometer momentum, we – our activities have increased, including IP verifications, customer tape-outs on standalone products and product reliability qualifications. These 28-nanometer customer collaborations will provide further traction heading into the second half of 2014. UMC will continue to broaden our customer base and penetrate additional high growth areas to expand customer adoption. We remain confident for the long-term, as our sound business strategies, solid engineering execution and strong commitment to customer service, will help ensure UMC's future business growth and deliver enhanced profitability to raise shareholder value. Now please allow me sometime, to summarize the recent highlights in Chinese. [Foreign Language] So I have finished my comment and now let me go over to third quarter, 2014 guidance. For the foundry segment wafer shipment has showed an increase of low single-digit percentage trend. For foundry segment ASP in US Dollar, we have remained flat. The UMC foundry segment gross profit margin will be in the mid 20% range. Capacity utilizing rate for foundry segment will be in low 90% range. 2014 CapEx for foundry segment will be $1.3 billion. The guidance for new business segment revenue to be approximately NT$1.2 billion and net loss attributable to UMC Parent Company to be approximately NT$706 million. That concludes my comments. We are now ready for questions. Operator, please open the lines. Thank you.

Operator

Thank you. Ladies and gentlemen. We will now begin our question-and-answer session. (Operator Instructions) the first question is from Nikki Hu from Credit Suisse. Go ahead please.

Randy Abrams

Hi, this is Randy Abrams from Credit Suisse. The first question I had, on just the outlook for second half and starting with third quarter for low single-digit growth. I guess, it's a little bit modest with 28-nanometer starting to ramp. So you could talk about just what you're seeing for the environment, if any of the end markets are slowing down as you look into the third quarter or slow down between 8-inch versus 12-inch. And was counting on that, fourth quarter the last few years has been down single digits. If you're expecting at least at this early stage similar profile for anything to support above seasonal fourth quarter.

Chi Tung Liu

The reason, the third quarter 2014 is low single-digit revenue growth. It's mainly due to our 8-inch capacity constraint. During the 2Q, 3Q 2014 and we – even though our capacity – we cannot provide additional revenue and margin offers from 8-inch operation. And in the meantime, the revenue increase mostly from 28-nanometer, which is currently in initial ramp stage. So our 28-nanometer business is not able to contribute meaningful margins and sales for the moment.

Randy Abrams

Okay and on the, I guess the 40, 65, 90 like the middle node to mature 12-inch. Maybe what's driving the weakness for 8-inch, 28 and if you expect to kind of, I guess a slower fourth quarter then or perhaps you're through with capacity constraint. The 28, 8-inch could help by fourth quarter.

Po

So your question is?

Randy Abrams

The question is more regarding the where the weakness is coming from on the mature 12-inch nodes like the 40, 65, 90 and whether you expect weakness against, I guess or normal decline into fourth quarter for the overall business.

Po

For 12-inch, our 40-nanometer node demand still very strong and for the mid-technology node for the 65 and 90. We have some percentage point, unutilized capacity. Mainly because we plan for the some application off site this year and especially that the driver high end driver products. Unfortunately given realized for the moment, so that's from percentage point to ensure operation utilization rates loss.

Randy Abrams

Other question, I want to ask about the Japan business, which doubled in the quarter. If it's a one-time or start of trend and maybe, if you could talk about the press maybe leaked out a bit. On potential to get more aggressive on Japan and I guess I'm curious your thought process on Japan, where you exited UMC Japan, but if the right opportunities come. If you see some new opportunities in the Japan market.

Chi Tung Liu

First of all, we don't comment on multi speculation or medial speculation having said that, Japan has always being a important multi-segment. We recently just hosted a large scale technology workshop in Japanese market and of course, this quarter increase in Japanese market is a bit too early to call it as a trend, but definitely we are putting lot more resources into new Japanese markets along with the, all sorts of training in Japan.

Randy Abrams

Okay, thanks and my final question on the new business. You're guiding pretty significant drop in sales. If you could talk about what's happening in the new business.

Chi Tung Liu

Mostly due to their very recent event of the so called antitrust banned by US and Indian market and happened out, we saw significant reversal of operations. One of our larger solar business in [indiscernible] post profit and because of the so called antitrust and anti-dumping and we are going to see quite significant dry up in business in at least the third quarter. So that will cause some significant reversal of the operation.

Randy Abrams

Okay, thanks for the color. Hope to get advocate [ph]. Thank you.

Operator

The next question is from Bill Lu, Morgan Stanley. Please ask your question.

Bill Lu

My first question is on second quarter gross margin, apologies. I didn't have the chance to go through all the materials you had, but total gross margin was 22.9% guided for foundry gross margin in the mid 20% range, can I assume that you hit that target and the overall was dragged by the solar business?

Chi Tung Liu

Yes, definitely I will give you breakdown. I mean, gross margin in second quarter was 25.3%.

Bill Lu

25.3%. Perfect, thank you.

Chi Tung Liu

And operating margin was 10.2% for foundry business.

Bill Lu

Okay, great and the second question is, you talked about supply constraints for 8-inch wafers is there anything to do about that?

Po

We actually, we're expanding our capacity inch fab, so we have for this year we are aiding 5,000 wafers per month capacity in our fab Suzhou, China. And we are planning to build more capacity in 2015 next year.

Bill Lu

Okay, so the Suzhou capacity comes in line 4Q, I assume?

Po

It's actually, the 2Q. It’s already done.

Bill Lu

Oh! I see, I guess I may have, is your more to come going forward.

Po

For the next move, we are planning to expand our the same, the Suzhou fab from additional 11,000 wafer per month for next year.

Bill Lu

Can you tell me about when next year?

Po

It's the same, the middle next year.

Bill Lu

Okay, got it. And then, on 28-nanometer previously you've guided for 5% of revenues by the fourth quarter, can you give us an update there?

Po

This will still remain that our target of 5% in first quarter.

Bill Lu

What is the percentage in 3Q?

Po

It's around 3%.

Bill Lu

So 1% in 2Q, 3% in 3Q. Got it. Look I guess, the one thing I'm little bit concerned about. Is there is been some press releases on Qualcomm going to SMIC for 28-nanometer. Now I think, you are going to be fairly with somebody else and you're so ahead of SMIC in your timing but, typically when SMIC comes in, they tend to price pretty aggressively, how do you think about that competition?

Po

Thank you for your question and let me start with our 28-ALP PolySiON version. We have a trend very good in EU enhancement and also the quality manufacturing profitability. We always provide our customers the best cost to performance ratio. So that can raise the level of UMC's competition and we have very high confidence to surpass our competitors and gain market share for 28-nanometer PolySiON version in the next few quarters to come. And for our 28-nanometer High-k / metal gate HPM version. UMC is in a unique position to provide our customers, the most reliable and service-friendly 28-nanometer High-k / metal gate technology. So we are pretty are confident, that we can have more business and gain more share on 28-nanometer technology.

Bill Lu

Okay, great. Thank you very much.

Operator

The next question is from Gokul Hariharan, J.P. Morgan. Please ask your question. Gokul Hariharan – J.P. Morgan: My first question could you talk a little bit about your 28-nanometer capacity plans by end of this year, as soon as some early plans in terms of what you're expecting for next year as well in terms of 28-nanometer capacity? And how should we think about 28-nanometer profitability as we go to 5%, when do we get to that mid 20s kind of profitability for 28-nanometer as well? If you're going to take until end of next year or if you're going to be a bit earlier?

Po

Yes, we will expand our 28-nanometer capacity from 10,000 wafers now to 12,000 wafers in 4Q this year and we also are have the plan to expand another additional 8,000 wafers per month, totally up to 20,000 wafer per month in mid of 2015. Gokul Hariharan – J.P. Morgan: Okay and on the profitability side, how should we think about the profitability? When we get to 20,000 capacity and let's say reasonable utilization, are we still going be at sub 20% gross margin or is it, when to get better by then?

Po

We expect the gross margin of our 28-nanometer, this is we're improved along with the year learning curve and also the productivity improvement and we believe 28-nanometer will be close to our 3-inch copper rate, gross margin areas in the few quarters. Gokul Hariharan – J.P. Morgan: Okay and I have one more question on 8-inch capacity tightness. I think you mentioned that you did prepare some capacity for drivers in 12-inch, 65-nanometer which did not span out this year. So what is the thinking of behind 8-inch capacity tightness? Do you expect a fair bit of conversion happening from 8-inch to 12-inch next year or is it still going to be 8-inch capacity is going to be tight for most of next year as well on a industry basis?

Po

In our case, let me answer your second question, first. Our perspective is to – the inch business demand is still very strong comes from the year, 2015. Yes, we also observe to certain structural changes behind high loading of inch board and the long-term demand is straightly driven by the growth of the IC content, which means that the number of chips persistent is going up and also the thigh size is getting larger. It means, the function integrates on a single thigh is getting more functions. So that's why, it drivers – in terms of wafer number. Our wafer demand is still very strong. Gokul Hariharan – J.P. Morgan: But you're not expecting a very big conversion from 8-inch to 12-inch board [ph]. Some of the high volume 8-inch applications like say smartphone driver IC's or power management IC next year; you don't expect that to happen.

Po

Yes, we do expect that will happen, but I cannot – for the moment, I don't have a clear picture, how much you will convert to a – for the 3-inch the high end technology for that driver IC's. We do have that planned. Gokul Hariharan – J.P. Morgan: Okay, all right. Thank you very much.

Operator

The next question is from Dan Heyler, Merrill Lynch. Go ahead please.

Dan Heyler

So, couple of quick ones. On the 65, 90 fab that's some are underutilized. What's the prognosis getting that business on the high performance driver business in the fourth quarter or is this going to take more than that?

Po

I will say, it will take more time. Yes. It quite depends on the customers, their design and business engagement. Yes.

Dan Heyler

Got you. Okay, is there some other products on the you think are on the pipeline. Now you've got now invest more time and find another application. So is the potential to fill that with other customers more likely towards the next year or could you fill with something else, this year?

Po

Yes, we do see a lot of opportunity there for those 3-inch technology will be better utilized especially for the high end driver IC and power management IC and microcontroller, we do see, there are many opportunities.

Dan Heyler

Okay, a bit more of 2015 story then. Okay and then on the 8-inch you talked about a structural pick up there. Frequently, what you find when supply chain is going through a big restocking process that everything fills up including your 8-inch FAB. So I'm wondering, what gives you confidence that the current high utilization on 8-inch is sustainable and it's not merely cyclical pick up.

Po

Yes, we think it's to me, to our case it's pretty that will be sustainable to end of this year, at least.

Dan Heyler

I mean, you said, you wanted to expand capacity there next year. Am I right and how much capacity, do you plan to add for 8-inch?

Po

11,000 wafers per month.

Dan Heyler

Next year?

Po

Yes, next year.

Dan Heyler

Okay, what's driving that?

Po

The same driver IC and power management IC.

Dan Heyler

Got it makes sense. Okay and then finally on the – to maybe just housekeeping on the depreciation side, what the quarterly run rate on your depreciation?

Chi Tung Liu

Next quarter, we're talking about less than 3% increase and whole year, 2014 will be less than 10% increase than 2013.

Dan Heyler

Okay, Chi Tung and what's the fourth quarter number?

Chi Tung Liu

You have to work it out yourself. Basically.

Dan Heyler

So less than 10%, you said.

Chi Tung Liu

Actually between 5% to 10%.

Dan Heyler

Can you give, a little bit more granularity on the full year numbers?

Chi Tung Liu

Closer to 5%.

Dan Heyler

So 5% year-on-year growth and depreciation.

Chi Tung Liu

Yes.

Dan Heyler

Okay, got it. Thank you. Great, thank you guys.

Operator

The next question is from Roland Shu, Citigroup. Please ask your question.

Roland Shu

First question for the new business we have lower revenue and higher loss, is lease just wearing off or are you expecting this to continue going forward?

Chi Tung Liu

Roland, we just highlighted like over the weekend new penalty issued by US mainly and also Indian market for anti-dumping, antitrust for including Taiwanese manufacturer and solar was doing better in quarter but still pretty shallow in terms of their profits and basically this sanction just turned the whole picture around and quarter three for now is kind of like sudden test to all the solar markers in Taiwan including our investment. We don't know, how our outlook going to develop and we will continue to monitor and there hopefully we will have better news in the next quarterly conference call.

Roland Shu

Okay, so this situation probably will be continued into the third quarter or even your going forward, if the situation does not improve.

Chi Tung Liu

Or if the big picture remain the same, of course we will continue to enhance our efficiency and try to find our way out, but so far visibility is rather limited to third quarter.

Roland Shu

Okay, thank you. Second question about talk about 28-nanometer capacity, said we are going to spend 12,000 by end of this year and also at another 8,000 wafer to 20,000 in mid of next year. Remember last quarter company said actually the money having spent for 10,000, all this year. So I just wanted to clarify, whether it's 8,000 additional capacity invested, actually the money has been spent in this year or there will be another capacity in next year?

Chi Tung Liu

Well, you know better than most of people, I assume that the lead time of the different equipment varies. So some of the long lead time, we probably have issued PO already this year, but majority of the 8,000 PO will be issued from this point onward and payment will follow accordingly. So the payment schedule is always a few months behind trailing the PO schedule.

Roland Shu

So actually, when your PO base release, but cash base will be still next year.

Chi Tung Liu

Well big part of it will fell into 2015.

Roland Shu

Okay, understood and so do you have any rough idea for how much CapEx spending will be in next year because you have at least right now, 28-nanometer capacity expansion and also have the 11,000 wafers 8-inch expansion. So how much will it be for the total CapEx next year?

Po

For 2015 CapEx, we are now under discussion and we could potentially take an aggressive approach on 2015 CapEx in order to take advantage of our unique position on 28-nanometer. So we believe that 28-nanometer is strong and [indiscernible] is numerous way off applications and customers and we haven't finalized the CapEx for 2015 now.

Roland Shu

In addition to 8,000 wafers 28-nanometer capacity and 11,000 8-inch capacity, is there any other capacity expansion plan other from that?

Chi Tung Liu

Roland, as we recall, we do have a new job in Shell [ph] has been completed and we are installing the facility at the moment. So P5, if we start to equip and there will be another new beginning for large scale dropping towards effects.

Roland Shu

Okay and still there P5 schedule will be in second half next year.

Chi Tung Liu

P5, I mean, we are equipping right now, but in terms of the installing equipment that will be partially driven by the market demand and that will be finalizing next year.

Roland Shu

Okay, thank you. Last question, is you talk about in 3Q you're doing some capacity upgrade in the 8D and then that result in your capacity. Can you give some color, what kind of upgrade are you doing at the 8D now and for what kind of purpose? Thank you.

Po

Our 8D Fab is mainly is a conversion from the copper lines to aluminum line to support the demand from the high end driver ICs.

Roland Shu

Okay, so previous copper line actually, so just asked which to 12-inch. I mean the application also.

Po

Yes, that's correct.

Roland Shu

Okay, thank you very much.

Operator

Your next question comes from Donald Lu – Goldman Sachs. Go ahead please.

Unidentified Analyst

The first question, I want to ask is about the customer base for 28-nanometer commercial product. Can you elaborate on how many real like meaningful volume customers you have for the 28-nanometer, that's number one and number two is, can you give update on your progress this High-k / metal gate because I think, that latter since like the unique opportunity you commented on? Thank you.

Po

Our 28-nanometer we have over 20 customers engaged and around 50 products announced already and we have more than five products now enter the pilot production stage and so your second question is regarding 28 High-k / metal gate version and yes, we have pretty common. Our 28 High-k / metal gate technology is on track. So and that we saw gaining the traction from many customers. So we are pretty confident on our 28-nanometer High-k / metal gate.

Unidentified Analyst

Just follow-up on that, for the five product in pilot production, is that from one single customer or from two or three customer and also?

Po

That's from more than three customers.

Unidentified Analyst

Two or three customers. Okay and.

Po

More than three customers.

Unidentified Analyst

Oh! More than three, okay and four the 28-nanometer High-k / metal gate, when do you think it will represent let's say 1% or 2% of your total revenues?

Po

In the Q4, that first quarter this year.

Unidentified Analyst

Oh! Okay. Great. Okay and my second question is more on the 8-inch. I mean, you're adding 11,000 in sooner. So clearly you have a pretty positive outlook on the demand. Can you, I mean you commented that there is driver and a power management IC are the key drivers there. Can you give us more color on why do you think, this growth is sustainable into next year? And I'll have to follow-up on the 8-inch.

Po

They're mainly because of the growth of IC's content. As I explained just now the numbers of chips persistent is going up. It's obviously and secondly the functions integrated on a single thigh is also a driver, the chip size becomes larger. So that will consume more wafers in order to support it, our customers. So the wafer in terms of wafer, the demand should stay very strong, in terms of that 2015.

Unidentified Analyst

But is that, do you mean that a content, semi-content on the smartphone or is that just in general has increased?

Po

Yes, many of the smartphone. Yes.

Unidentified Analyst

I mean, why can't you move the product to 12-inch.

Po

Yes, we actually have been in very intensive work we saw customers trying to port in those high end driver to 12- inch wafers technology. To ease our inch fab capacity constraints is.

Unidentified Analyst

All right, okay. A follow-up question on that, is the 55-nanometer SONOS embedded flash. I mean that, a pretty big market in theory. I remember a few years back. I think Cypress trying for the 65-nanometer SONOS to UMC. I mean, how much is that, if your revenue right now? How much is that, if you're 65-nanometer? And also I mean, what's the big advantage of this 55-nanometer technology and last how much, earning outlook for this technology?

Po

Is so far, we don't see the significant revenue contribution from this 65-55 SONOS technology and somehow we see the engineering progress is pretty good and I mean, we do for business for the next year.

Unidentified Analyst

I mean, why isn't it a 65-nanometer SONOS contribute revenues because you've got the technology few years back.

Po

I probably cannot comment on – at this moment. You want to add?

Bowen Huang

Hi, Donald. I think, we have a very good business opportunity impacted as solution at 55-65-nanometer. This is why, our CEO actually didn't comment on 65-nanometer ALP plus 4 and then we will see some of the IP, related to longer tier memory's and I think for the 55-nanometer or 65-nanometer the embedded solution will become mentioned in next few years instead of some kind of say along flash numero [ph] top product in the future.

Unidentified Analyst

So to think, the opportunity is more like in the next few years.

Po

Yes, correct.

Unidentified Analyst

Okay, great. Thank you.

Operator

The next question comes from Szeho Ng from Paribas. Go ahead please.

Szeho Ng

Hi, Szeho from BNP. Two questions from my side. With regard to ThinkPad [ph] actually express the production to start?

Po

Our ThinkPad [ph] technology will enter process core in the first quarter 2015 and ready for customers tape out in second quarter, 2016 and our ThinkPad [ph] involvement is right now way on track.

Szeho Ng

All right, wrapping your contribution should be something like in 2016, right?

Po

Yes, we are covering in 2016. Yes.

Szeho Ng

Okay, good and also for the R&D expenses, it has been sustaining the round NT$3 billion cash run rate on a quarter basis. What should we expect the expense to come down after and off the [indiscernible] at year?

Po

It will show some mild decrease by that same time, the R&D cost for 14 [indiscernible] also more expensive. So we only expect to see significant reduction in R&D in 2015.

Szeho Ng

Okay, all right and lastly just an update, what percentage of the revenue is coming from the specialty classes and what's your plan for next year for that particular market?

Po

Around 30% of total revenue is coming from specialty technology.

Szeho Ng

Okay, the percentage for next year should be similar or we should expect it to spike up?

Po

We –it will be going up, I'm not sure, its spike up. Actually is going up yet.

Szeho Ng

Okay, all right 30% similar like, 35%, 40% next year?

Po

See it also depends on the total base, so we don't have the figure yet, but the trend is definitely going up.

Szeho Ng

Okay, all right. Thank you gentlemen.

Operator

The next question comes from Julie Tsai, UBS. Go ahead please.

Julie Tsai

Hi, this is Julie from UBS. Couple of questions, two follow-up. Can you give some a bit of color into Q3 outlook for this segment or the sectors? Thank you.

Po

In the third quarter, the in terms of sector communication will be a stronger sector and secondly followed by computers. Computer segment.

Julie Tsai

And assuming consumer will be the weakest?

Po

Yes, correct.

Julie Tsai

Okay and also two more question. The second will be you did mention about the Fab upgrade at 8D, right and this is for the high-end driver IC. So will we see Q4 some capacity coming back to normal for that specific Fab?

Chi Tung Liu

It will be a minor dip in quarter three, but I think the conversion and also the addition of capacity will play out for the next few quarters. So we don't expect to immediately reach the previous level in Q4.

Julie Tsai

Okay, so could we fairly assume that the ASP for this particular product will be better than normal.

Chi Tung Liu

No, no more – driver ASP will be lower than any product using copper line.

Julie Tsai

Right, but I meant apple-to-apple comparison. If you're switching from copper to aluminum line and supporting your high-end driver IC, so will ASP will be better or is this just a request by the client?

Chi Tung Liu

It will be a better new transition rate overall in longer term, but not necessarily helping in speed.

Julie Tsai

Oh! I see okay. And then, in your presentation Page 14, your guide 2014 CapEx will be about $1.3 billion and you also separate them into 8-inch and 12-inch. Could you also give us a bit of guidance in terms of technology now that you're going to spend?

Chi Tung Liu

And most the 12 inches are 28.

Julie Tsai

And that's it. No 14, no, anything.

Chi Tung Liu

14 for the time being is R&D stage.

Julie Tsai

Okay, so my last question would be leading to, can we talk about 2014 a little bit. How is it progressed, no significant revenue contribution this year? Well maybe even for next year, but just wondering your collaboration will that be in how is it progressing?

Po

Again, our target it will be the processed core will be finished in the first quarter, 2015 and they're ready for customer to buy in, 2Q, 2015 and for our 14-nanometer node, there are a few early adopters and demand will come us, along our cost becomes attractive. So among those trends of technology revenues customer are also concerned on technology complexity and toolsets mismatch, which could potentially limit the multiple sourcing strategy. So in this regard, UMC can provide distinct value of manufacturing stress ability and the experiments with customized process and design in technology. So in general, we will find the position for UMC.

Julie Tsai

I see, thank you and maybe just one very last follow-up question. Regarding to your Q2 operating margin and also gross margin. It seems to be slightly under straight expectation well even my analyst expectation, is it the similar taste for you and why?

Chi Tung Liu

Well, we guided mid 20 and we achieved 25.4% so we are way on top, way on the spot.

Julie Tsai

For Q2, right?

Chi Tung Liu

Yes.

Julie Tsai

Okay. All right. Thank you.

Operator

The next question comes from Dan Heyler, Merrill Lynch. Please ask your question. Mr. Heyler, you're on the line, please proceed.

Dan Heyler

Can you hear me?

Po

Yes.

Dan Heyler

Great, thanks. So I had quick follow-up on the 28-nanometer for the you've got five customers in pilot production and you're trying to bring those up to volume. Is there any demand risk that you're seeing in the fourth quarter? Is there a potential for, there's been some softness here and there and LT expectations? So just wondering, how confident you're that those orders will continue to ramp. I know it's early stage, so I presume that there would be ramping up, but I just wanted if there is any demand risk to the fourth quarter targets.

Chi Tung Liu

Yes, very confident. We don't see risks from our demand sides on 28-nanometer.

Dan Heyler

Okay and then, did you have a contribution, expectation for the fourth quarter revenue contribution on 28-nanometer, did you have a target?

Chi Tung Liu

Yes, we have 5%.

Dan Heyler

Okay, so that's on track.

Chi Tung Liu

Yes.

Dan Heyler

And I want to revisit this, since this question on 14-nanometer came up, I just wondered one of the question management, whether you think, why it's necessary to pour significant amount of money into 14-nanometer given the opportunity on 28-nanometer, which is a fairly long node and the limited competition in 28-nanometer, other than potentially SMIC will off to see, but given the size of that opportunity and the desire for people to do SONOS from UMC to somebody else. Why not focus all your resources on executing 28-nanometer and diverting R&D dollars to 14-nanometer, which is a very new and very challenging technology?

Po

So actually we were aligning our customers on these, so we are the 14-nanometer fab, since most of the customers design is to 28-nanometer, so if it is a 14-nanometer, we will see a strong node for us and so there are customers, keep up high interest to engage with UMC for 14-nanometer and ask for 28-nanometer, as we just mentioned that we address a pretty good attraction from many customers. We see that in the long-term our 28-nanometer pretty good node and good return for UMC.

Dan Heyler

I mean, just to play a little bit devil's advocate, would you want to agree the issue isn't really customer demand, the demand for 28-nanometer is there for long time. Customers want it from second source, the issue is more about execution, isn't it?

Po

Yes, we agree that. As security is [indiscernible] for us.

Dan Heyler

It just seems to be that will be better off to put those valuable smart engineers working on 14-nanometer to focus on brining up 28-nanometer and not just, wonder why. It's pretty expense to spend on 14-nanometer.

Po

Yes, we already, we joined IBM alliance on [indiscernible] department, so we tried to leverage IBM's alliance and on their technology to be more cost effective and cost to performance ratio, to better cost to performance ratio to our customers.

Dan Heyler

Okay, all right. Well thanks for that. Appreciate the color. Thank you.

Operator

Ladies and gentlemen. We are running out of time. So we are taking the last question. The last question comes from Steven Pelayo, HSBC. Please ask your question.

Steven Pelayo

Agreed, a few questions here. TSMC enjoyed multiple quarters of weighted average ASP increases with their 28-nanometer way up. I'm curious with UMC saying realized to 5% of mix in the fourth quarter getting the 20,000 wafers capacity by mid 2015. Do you think, you can see blended ASP increases over the next couple quarters or the next three to four quarters?

Po

We are looking for the – our gross margin improvements in the few quarters on 28-nanometer. So I'm not sure, if I answer your question, but.

Chi Tung Liu

So Steven, we do think the 28-nanometer to play pricing enhancement, but again that's largely depends on the other bigger part of our revenue, as well. So we cannot guarantee the ASP increase in the next few quarters, but if everything stayed it is, of course ASP will be going up, if 28-nanometer play more portion of our revenue.

Steven Pelayo

Okay, let's dig a little bit more to that. I mean, the ROE's that you're generating right now at 90% utilization rate is still kind of mid single digits. I guess, well off to your full off seasonal declines in fourth quarter and first quarter. So we kind of have to wait to see maybe a year from the day, what maybe you can do and the next kind of seasonal peaks and I'm curious, how much do you think 28-nanometer holding back your margins, then or do you have a targeted ROE? I mean is there something you can help us try to quantify that impact?

Po

Well currently the margin is below 12-inch copper average, it will take a few quarters to reach the 12-inch copper average and of course we have targeted ROE's but put a lot of parameters together including loading ASPs and all the cost reductions. And third quarter, we do encounter some utility increase in Taiwan as well as the higher depreciation cost. So not necessarily fact the tool or the theoretical gross margin, we should have because of the key cost item increase.

Steven Pelayo

Okay, two more quick questions from me. Interesting, gain was kind of suggesting that maybe 28-nanometer little competitive thinking your very good reliable second source. I guess, I was thinking Samsung maybe has a excess capacity here at Global Foundries and Smith coming in, do you feel the 28-nanometer is any more less competitive than the four prior nodes?

Po

As I spend, 28-nanometer we have high HPM version, so called High-k / metal gate version. UMC provider gained at High-k / metal gate which is, we are very unique position. This is – we are one of the only two manufacturers can provide this version. So we are pretty confident on our demand and growth on this 28 High-k / metal gate version, which is at the competitive as you mentioned just now is not competing in this area.

Steven Pelayo

Understood. Last question from me, is just on customer concentration. Do you have any 10% of revenue customers and how many in second quarter or third quarter something like that?

Po

Yes, we have a one to two customers accounted for more than 10% to UMC's total revenues.

Steven Pelayo

Okay, so two customers over 10%. Great. Thank you very much.

Po

Thank you.

Operator

Thank you and that concludes today's Q&A session. I will turn things over to UMC, Head of IR for closing remarks.

Bowen Huang

Thank you, everyone, for joining us today. We appreciate your questions. As always, if you have any additional follow-up questions, please feel free to contact UMC at ir@umc.com. Have a good day.