UMC 联华电子

Operator

Welcome everyone to UMC’s Q4 2012 Earnings Conference Call. (Operator instructions.) 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 at www.umc.com under the Investor Relations, Investor Events section. I would now like to introduce Mr. Bowen Huang, Head of Investor Relations at UMC. Mr. Huang, please begin.

Bowen Huang

Thank you and welcome to UMC’s conference call for Q4 2012. With me today is the CEO of UMC, Mr. Po Wen Yen, and the CFO Mr. Chi Tung Liu. 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 risks that may be beyond the company’s control. For these risks please refer to UMC’s filings with the SEC in the US and the ROC securities authorities. I would now like to introduce UMC’s CFO, Mr. Chi Tung Liu to explain UMC’s Q4 2012 financial results.

Chi Tung Liu

Thank you, Bowen. For Q4 2012 revenue was NT$26.09 million, an 8.5% quarter-over-quarter decrease from NT$28.53 billion in Q3 2012 and a 6.8% year-over-year increase from NT$24.43 billion in Q4 2011. Gross margin was 16.8%, operating margin was 3.7%. Net income was NT$1.1 billion and earnings per ordinary share were NT$0.09. In 2012, revenue for the full year was NT$106.00 billion with NT$9.08 billion operating income and NT$7.92 billion in net income, and NT$0.63 earnings per share. The above is a short summary of UMC’s results for Q4 2012. More details are available in the report which has been posted on our website. I will now turn the call over to Mr. Yen.

Po Wen Yen

Okay, thank you Chi Tung. Good day and Happy Chinese New Year to our guests and friends in the investment community. I’m here today to address you for my first quarterly conference call since becoming UMC’s new CEO. In Q4 2012, both UMC’s revenue and gross profit were in line with our expectations. The quarter-over-quarter revenue decline was primarily attributed to decreased wafer shipments. A total of 1.07 million 8-inch equivalent wafers were shipped with overall capacity utilization at around 80%. Due to the impact from declining wafer shipments and currency appreciation, gross margin for the quarter was 16.8%. Revenue contribution from 40nm continued its sequential increase to 15% reaching our internal target. As industry competition intensifies the bar has been raised with regard to customers’ expectations for leading-edge technology products’ time to market and foundry service flexibility. As UMC’s newly appointed CEO my first priority is to enhance R&D for advanced processes. In order to realize maximum benefits from our R&D efforts, we will increase internal resources while leveraging outside collaboration so that we can effectively integrate our R&D with manufacturing to ensure timely delivery of those key engagements and projects. For example, UMC’s FinFET development began in 2010 and was bolstered in 2012 through a cooperative licensing agreement with IBM for their fundamental FinFET technology. Our FinFET development team has since been making good progress. Meanwhile, UMC’s recently demonstrated the world’s first TSV-enabled 3DIC chip stacking technology developed under an open ecosystem collaboration. The 3DIC developed with an OSAT partner reached a major milestone by passing package-level reliability assessments. This successful collaboration reaffirms our strong commitment to elevating technology capabilities in order to fulfill the promise of providing high value added services to our customers. UMC’s operating results can be further enhanced through the management team’s efforts to increase capacity, control the costs and expand our customer base. Although UMC’s 28nm ramp has been slower than anticipated, we remain dedicated to advanced technology development and timely capacity deployment to serve our customers’ needs. Equally important is our customers’ unwavering commitment to bringing their new and existing 28nm products to fruition at UMC. In the short term we continue to see elevated demand uncertainty from customers along with supply chain inventory that will need additional time to digest. When demand recovers we are optimistic about the growth momentum from the strong mobile communications segment. 2013 is a critical year for UMC to demonstrate continuous success. In addition to acquiring Hejian Technology’s operation which will help UMC expand in the China market and broaden the company’s operating scale we will continue to invest appropriate CAPEX to further our diverse product development and expand the leading edge capacity. These endeavors will help expedite the company’s growth and strengthen competitiveness to enhance UMC and its customers’ overall value, and increase shareholders’ equity. Now let me provide you with the guidance for Q1 2013. For the Foundry segment, wafer shipments will be increased by approximately 6%. For the Foundry segment ASP in US dollars will decrease by approximately 6%. For the Foundry segment profitability, the operating margin will be approximately breakeven. For the Foundry segment, capacity utilization will be around 75%. 2013 Foundry CAPEX budget is about $1.5 billion US. There is some other guidance to non-operating income and the New Business segment. For non-operating income after-tax will be approximately $100 million US. For the New Business segment, revenue will be approximately NT$2 billion. For the New Business segment operating margin will be a loss of approximately NT$1 billion. This concludes my comments. Thank you for your attention and we are now ready for questions. Operator, please open the lines up. Thanks.

Operator

Thank you very much, sir. We will now begin the question-and-answer session. (Operator instructions.) The first question comes from the line of Randy Abrams with Credit Suisse. Please ask your question.

Randy Abrams

Yes, good evening. Thank you. I wanted to ask the first question for Mr. Yen just given your coming in as CEO. One of the priorities you mentioned was increasing the focus on the R&D – can you talk about how that translates? Do you expect in terms of headcount or R&D expense any step up on the R&D side? And if you were to go through just any other operational or strategy changes or you think it’s for the most part ongoing operations as it was prior to your appointment?

Po Wen Yen

Thank you, Randy. Your first question is about UMC’s increase of our R&D resources, and in terms of the headcount and resources we are going to increase about 15% in terms of our R&D staff. There are two focuses we are now putting efforts on – R&D, integrate our R&D and manufacturing size and to improve the technology strengths to put us in ramp up cycle [times]. The other focus is we are, besides increasing our internal R&D resources we are also leveraging external resources through consortia partnerships and collaboration which I will also explain in this afternoon’s conference. And this, however it will take some time to show the results.

Randy Abrams

Okay. And to clarify, the 15% R&D headcount increase, was that 2012 to 2013 – like that’s what you’re expecting for this year?

Po Wen Yen

That’ll be this year.

Randy Abrams

Okay, for this year, okay. And for Chi Tung, I wanted to clarify the guidance again or factor for it. I think the underlying Foundry business implies revenue flat sequentially when you factor shipments and ASPs. The margin, if you can walk us through again the margin where it’s coming down from about 4% to breakeven, the factors for the sequential change in the margins?

Chi Tung Liu

Mostly coming from ASP which was the result of relatively low 12-inch rate versus 8-inch and for the 28nm (inaudible), the operating costs, the manufacturing costs is also higher than the other nodes. So altogether we’re actually seeing a decline in Foundry profit margin.

Randy Abrams

Okay. And the other question I had, the $1.5 billion for CAPEX, how much of that is pure 28nm or 40nm capacity additions? And how much capacity additions do you expect with that? And how much would just be for technology adds and upgrades? If you could give a little more color on how you plan to split out the spending.

Chi Tung Liu

For the CAPEX right now we provided the numbers, which is 6% for 8-inch and roughly 94% for 12-inch. For this 94%, about 11% is R&D-related so the remaining 83% is mainly for 12-inch although we are also building the shell of the of the third 12-inch wafer fab in Thailand, the phase five and phase six. That will also take some of the CAPEX. So that’s the rough figures for 12-inch capacity for 2013, and the majority is 28nm related.

Randy Abrams

Okay. For the 8-inch fab, the Hejian fab, what’s the maximum output and what are the restrictions in terms of the regulation, like what can you actually do in China for manufacturing?

Po Wen Yen

The output capacity is 45,000 8-inch wafers and they can do up to .13 +/- 2 to represent UMC’s most leading mainstream technology.

Randy Abrams

Okay, thank you.

Operator

Thank you very much. The next question comes from the line of Szeho Ng from BNP. Please ask your question.

Szeho Ng

Hi, good evening gentlemen. For the New Business segment what will be the revenue level required to achieve an operating breakeven?

Po Wen Yen

It’s not so much about scale itself, it’s rather about the pricing and costs. So right now our New Business is mostly in the solar industry and is suffering from very weak pricing. So it takes the price to rebound to make the New Business be operating profitably. So (inaudible) may not be able to help on the profitability of the New Business.

Szeho Ng

Oh, I see. And second question, roughly what percentage of the Foundry revenue is related to (inaudible)?

Po Wen Yen

About 30%.

Szeho Ng

Okay, alright, thank you.

Operator

Thank you very much. The next question comes from the line of Jeffrey Toder from RBS. Please ask your question.

Jeffrey Toder

Hi. Two questions actually: first, on the New Businesses, how do we think about those going forward? I mean if solar prices don’t rebound should we assume that the revenue and operating contributions are the same or is there another way we should look at it?

Po Wen Yen

That’s a very volatile industry so it’s a bit difficult to focus beyond this quarter. However, as we mentioned we probably won’t put further capital into the New Business for the sake of capacity expansion. We will continue to inject capital for the efficiency improvements and the business continuity, so revenue-wise without the piping rebound I would say it’s pretty much at the current level. However, we do hope through the efficiency improvements the net income level or net loss, if you will, hopefully we will see further improvement throughout the rest of the year.

Jeffrey Toder

Okay, so we should assume stable capacity going forward. Do you have some guidance as to what that capacity number might be?

Po Wen Yen

Well, I think the revenue line at least for the current quarter will scale to about NT$2 billion so I think that’s the profile for the near-term.

Jeffrey Toder

Okay, and then just one more question on this. You mentioned that prices need to go up in order to reach an operating breakeven – can you give some indication of how much prices would need to rise to reach that level?

Po Wen Yen

We have two major operations – one in the [sim key] and one in the poly. The poly arm is doing slightly better as in they are reaching cash breakeven and moving to the cost breakeven. But the [sim key] is still a short gap to catch breakeven, so the [sim] part we will see we will need more effort to bringing the operation into a profitable operation, but the poly part looks like the opportunity is much better.

Jeffrey Toder

Okay. And then on the non-operating income after tax you’re looking at $100 million US or over $100 million US. That’s a pretty big number considering where your operating line is going to be. Can you explain where that’s coming from?

Chi Tung Liu

That $100 million US after-tax is mainly coming from the accounting treatments, so-called the volume purchase [gains]. So if you recall we settled additional 51% of Hejian’s shares from January 31 this year, and we now control 86% of Hejian. And since we’ve become the controlling party of that entity and we also only paid 55% of the booked value of Hejian, so under [FRS] the new accounting treatment we have to realize their gain at once which is where the $100+ million after-tax profit is coming from.

Jeffrey Toder

Okay. So that’s obviously a non-cash item then.

Chi Tung Liu

It’s a non-cash item.

Jeffrey Toder

Okay. And then can I ask one more question or do you want me to come back?

Po Wen Yen

Please yeah, please.

Jeffrey Toder

Okay. So I think in the meeting this afternoon it was a little bit disappointing, the progress that the company has achieved so far with 28nm. And it sounded like the whole process is delayed by about a year, so that UMC will be ramping up 28nm around the same time TSMC will be ramping up 20nm. So that’s actually an unusually long gap if you look at UMC’s progress over the last I guess ten years or so. Can you explain why 28nm has been so difficult and what the strategy might be to regain more competitiveness in advanced technologies going forward?

Po Wen Yen

I think our 28nm, our current status, there is room for us to improve, yes. But generally speaking our 28nm technology, we have [more favored] so this is according to our communications with our customers. We align our milestones with our customers’ so basically we are still following this milestone.

Jeffrey Toder

I mean I guess the question is last year in the second half, management – a different CEO, I understand that – but management seemed confident that the company would reach a 5% contribution from 28nm by the end of Q4. And now that seems to have stretched out a year later. I understand that yields have been low but I mean I guess the question is why was this node particularly difficult and if you’re starting your 28nm when Leading Edge is moving to 20nm, and then 14 FinFET follows within four quarters after that it sounds like it’s going to be an increasingly competitive environment going forward. And I guess, Mr. Yen, as the new CEO you are entering at a very challenging time. What’s your strategy for addressing I guess a more difficult market than it might have been five years ago?

Po Wen Yen

I think to come back to your first question, which is what we are facing – yeah, our year is low in the [benchmark] of the company and it is a certain gap, yes. So that’s why we’ve become a low priority among the suppliers to the first engagement customer. And however, we still have many engagements. As I mentioned, we have four [10nm] processes which have some of them are designed to UMC’s process, so those milestones are still on track with alignment with our customers. So mainly in the second half of this year we can have those processes and also the product qualification and start ramp up in Q4 this year. Regarding the strategy, in order to deliver our 28nm technology on time so we increase our R&D staff and also we put in some of our high (inaudible) technologies capacity have had to accelerate our R&D cycle time. So that’s mainly our current focus, to catch up the technology gap with our benchmark company.

Jeffrey Toder

And if you’re starting to ramp up 28nm as you said in Q4, and I think you mentioned something like a low single-digit contribution, when do you expect 20nm to be available for customers and when do you think you’ll see ramp up of 20nm?

Po Wen Yen

I would say that will be first half of next year, 2014.

Jeffrey Toder

It’ll be ready for takeout or it will be ready for commercial production?

Po Wen Yen

Commercial production.

Jeffrey Toder

So we could see commercial production of 20nm in the first half of 2014?

Po Wen Yen

I’m sorry, you’re talking about 20nm?

Jeffrey Toder

Yes, 20nm.

Po Wen Yen

20nm, we’re still communicating intensively with our customers and I think not every customer requires this 20nm technology. So we are getting our process pipeline back however, yes.

Jeffrey Toder

Do you think a lot of customers may skip? I think you mentioned that really 20nm is popular with AP producers and those will be the initial adopters from an industry perspective. Do you think a lot of other customers might skip 20nm and move to 14 FinFET or 16, whatever – it’s got a bunch of different names; will move directly to FinFET and skip the 20nm node?

Po Wen Yen

For only the time to market products of lead companies to the best of my knowledge will adopt the 20nm device for their time to market products. And there are more companies, more of our customers are considering because the per-system cost and performance gain is not conducive to go through this technology. So some customers are considering skipping this and migrating to 14nm FinFET technology. So we are focusing on revealing our FinFET technology in the second half of 2014, next year. So that is our current focus.

Jeffrey Toder

Okay. So if I were just to sum up the strategy, because of yield issues you’re entering 28nm slower than you had originally anticipated. It sounds like 20nm for UMC is going to be fairly small and you’re going to try to move customers pretty quickly into FinFET and then catch up with the industry or get closer to the industry with FinFET. Is that a reasonable summation of the overall strategy?

Po Wen Yen

Yes, that’s correct.

Jeffrey Toder

Okay great, thank you very much.

Operator

Thank you very much. The next question comes from the line of Steven Pelayo from HSBC. Please ask your question.

Steven Pelayo

Great, a couple questions. First, there’s a lot of moving parts here on the top line as well as with your margin. Let’s first start with consolidating Hejian. I understand your guidance with Hejian for shipments is up 6%, ASP is down about the same. Without Hejian shipments flattish ASP is just down a little bit I guess, so I guess I’m trying to understand even the incremental revenue in Q1 that you’re getting from Hejian and how that impacts because I guess I’m trying to understand this 45,000 wafer [starts] a month fab what it’s contributing, what kind of utilization rates is it running and ultimately that’s going to lead me to questions about the margin in adding and consolidating this business. So can we talk a little bit about how much revenue it adds in Q1 to UMC?

Chi Tung Liu

So for Hajien we only count two out of the three months for Q1 as we made the purchase at the end of January. So only two thirds of Hajien’s Q1 numbers come into UMC’s consolidated numbers. And Hajien in Q1 is pretty much breakeven as well so not so much a help on our profitability, only at the top line.

Steven Pelayo

I understand it’s two thirds of a quarter but can you help me understand the revenue, incremental revenue that’s coming off there so then I’ll get a good idea of the run rate as we go into Q2 as well when we have a full quarter?

Chi Tung Liu

We don’t have the breakdown available but you do have the capacity we provided in our capacity breakdown, which is represented by 8N as Hajien.

Steven Pelayo

Alright. Is the utilization rate of that fab similar to the corporate average of 75% in Q1?

Chi Tung Liu

That’s right.

Steven Pelayo

It is, okay. And then I think I had a question, I guess we need to see a little bit more of a consolidated balance sheet as well so can we talk a little bit about that consolidation and the implications for some of the assets and equity there?

Po Wen Yen

Relatively speaking it’s probably a 15% addition to UMC’s overall asset base. So it used to be the non-core items but now it’s moving up to the operating items, so balance sheet-wise it doesn’t really change much. You still need to subtract the minority shares at the very bottom of your income statement so it’s just moving from a standalone financial statement to a consolidated financial statement – everything else actually doesn’t change much.

Steven Pelayo

Okay. And then maybe just a little bit longer-term question: I’m trying to also reconcile consolidating more trailing edge fab with wanting to invest more in R&D for 28nm and 20nm as well. So I guess I’m really trying to figure out what UMC’s margin model now looks like. So is R&D still kind of an 8% to 9% of revenue number, or maybe if I ask it even a better way – UMC is going to be back above let’s say a $30 billion per quarter kind of run rate. What kind of margins would you expect and what kind of R&D would be spent at that kind of quarterly revenue number?

Po Wen Yen

As you just mentioned we’re beefing up our R&D resources by increasing another 15% personnel in the R&D department. However, that’s not going to cost much from a cost structure point of view. We’re probably talking about moving from 8% to 9% to now maybe 9% to 10%. So overall that’s our current focus. And your second question was…

Steven Pelayo

And then on the gross margin level if you do consolidate a fab that is now running kind of breakeven, if you do get UMC back over a $30 billion kind of run rate in the next quarter or two what kind of gross margin we can expect?

Chi Tung Liu

Our guidance is still trying to focus on Foundry only, so if we talk about Foundry only going back to a $30 billion run rate then there’s no reason why we cannot go back to the previous average profit margin, say 30% to 35%. But with the addition of the New Business which tends to be lower margin, that is why we are having difficulty to predict on a long-term basis as it’s a very volatile industry.

Steven Pelayo

Yeah, we’re all having difficulty predicting there. Then just one last quick one clarifying: in your guidance you talk about non-operating income having this $100 million gain but then you also talk about the New Business segment having an operating loss of $1 billion. I guess I just want to make sure that when you say non-operating income after-tax will be a $100 million US gain that does not include the New Business loss.

Po Wen Yen

The New Business loss now is operating level, so that will impact our operating profits on a consolidated financial statement. But the number, the bargain purchase gains from the Hajien acquisition will be on the non-operating item.

Steven Pelayo

Ah, I’m sorry. So the way it’s written here it looks like it comes under the section of adding to non-operating income. So the New Business segment actually does add $2 billion in revenue to your reported top line but takes away about NT$1 billion in operating profit in Q1. Okay, thank you. Sorry for all the housekeeping questions, thank you.

Po Wen Yen

No problem.

Operator

Thank you very much. The next question comes from the line of Donald Lu of Goldman Sachs. Please ask your question.

Donald Lu

Hey, good evening. My first question is on the gross margin. Chi Tung, can you give us an idea about the gross margin for the Foundry business in Q1? And also I know the New Business is hard to predict but just in general, as far as our modeling purpose, what’s your assumption for Q1 on the gross margin lines?

Chi Tung Liu

Well, the Foundry-only segment guidance is operating, approximately operating breakeven on the operating level. And operating (inaudible) is actually no different from the previous quarter so you can do a quick calculation.

Donald Lu

Okay. So Foundry OPEX would be similar to Q4.

Chi Tung Liu

The Foundry segment, yes.

Donald Lu

And for the New Business, is the gross margin positive or negative?

Chi Tung Liu

New Business, I would say gross margin is probably negative for Q1.

Donald Lu

Okay. And for the New Business, previously you included that in the non-op.

Chi Tung Liu

That’s correct.

Donald Lu

So basically we should take out that part.

Chi Tung Liu

Exactly, so net income doesn’t really change.

Donald Lu

Okay, got it. And for the Hajien, I mean I don’t see the press release that you acquired recently control up to 86%. Can you give us an idea of how much you paid for that? You said you paid for 55% of booked value?

Chi Tung Lin

We paid 55% of the booked value and we actually had two tranches of transactions. The first tranche was 20% plus 15% free shares and the second tranche which had been at the end of January was 51%. So overall we paid $233 million US for the 86% including 15% free shares.

Donald Lu

Okay. So how much did you pay, I’m sorry?

Chi Tung Lin

$233 million US.

Donald Lu

Okay, got it. And also the next one is for Mr. Yen: regarding the FinFET, UMC has recently licensed IBM technology. And I mean do you think, will that technology work based on what you get from your IBM? I assume you’ve looked at a tape or a library or whatever, is that worth the money? Because I think historically IBM, UMC has licensed other technologies from IBM that really haven’t worked well in the production. I mean are you really confident it will work now or is it kind of something that you might have to find a Plan B?

Po Wen Yen

I think we actually optimized this from IBM team’s technology platform. We actually made some progress recently. At least in our [trend sys] level we already had a kind of prowess. And I would say this time the IBM alliance FinFET technology is much matured in the 130nm technology mode. So I’m more confident on this license collaboration on the FinFET technology, yes.

Donald Lu

I see. I mean the reason I’m saying it is it looks like TSMC has its own technology and now Samsung, Global Foundry, [UM Tera] are all going to IBM technology. And just in case this IBM thing doesn’t really work why doesn’t UMC really take another approach, license it maybe from TSMC or something else? Then maybe you will be ahead of Global Foundry and Samsung? Would that be something you would consider?

Po Wen Yen

We are open to anything. [Every influence] is good for our company’s competitiveness but this kind of hypothetical question is difficult for us to provide you with a view. So currently we’ve already paid for IBM’s FinFET and certainly any addition or enhancement that UMC can acquire we would love to consider. We’ve also assessed our FinFET technology carefully with our customers and there are some inputs from our customers, so we are more confident of this FinFET technology from our IBM alliance, yeah.

Donald Lu

Okay, got it. And Chi Tung, can you give us the deprecation for this year?

Chi Tung Liu

This year because of the [RFS] it’s a little bit complicated. If I were to add for UMC alone we are talking about a less than 10% increase. If we are talking about the consolidated depreciation versus 2012 UMC alone deprecation, we are talking about a 15% increase.

Donald Lu

Okay, got it. Thanks.

Operator

Thank you very much. That does conclude the Q&A session for today. I will hand the conference back to the presenters.

Bowen Huang

Thank you again for your interest in UMC. Please feel free to contact us directly if you have any additional questions. Operator, back to you.