Bowen Huang – Head, IR Po Wen Yen – CEO Chitung Liu – CFO.

Randy Abrams – Credit Suisse Bill Lu – Morgan Stanley Daniel Heyler – Bank of America Merrill Lynch Andrew Lu – Barclays Capital Roland Shu – Citigroup Steven Pelayo – HSBC Donald Lu – Goldman Sachs Szeho Ng – Exane BNP Paribas Eric Chen – UBS Rick Hsu – Daiwa Steven Pelayo – HSBC.

Welcome, everyone to UMC's 2014 Third Quarter Earnings Conference Call. (Operator Instructions). 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..

Thank you and welcome to UMC's conference call for the third 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 focus and UMC's fourth 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 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 our third quarter 2014 business results..

Thank you, Bowen. I would like to go through the third quarter 2014 investor conference presentation material which now can be downloaded from our website. So let us start from page 3. The third quarter of 2014 revenue was 35.21 billion with gross margin around 25.1% and operating margin of 4.8%.

The net income attributable to the stockholders of the parent was 2.92 billion and earnings per ordinary shares of 0.23 NT$0.23. On page 4, you can see the comprehensive income statement and revenue was down 1.8% quarter-over-quarter but mainly due to the decline in solar related new business.

Foundry alone actually posted 2.9% quarter-over-quarter sequential growth.

Gross profit on consolidated basis is 7.55 billion or 21.5% of gross margin and operating income after around 600 million asset impairment loss from one off solar subsidiary NexPower, around 600 million is now around 1.68 billion for operating income in the third quarter of 2014.

And the third quarter is always a peak season for our dividend income and therefore, we do have 1.3 billion of non-operating income in the same quarter. So total net income in the third quarter was 2.57 billion or an EPS of NT$0.23.

On page 5, you can see the cumulative performance for the first three quarters on a consolidated basis and revenue show around 10% year-over-year growth over the same period of last year and gross margin also improved from 19.3% to 21.1% in this year.

And operating income grows by almost 44% to 5.5 billion from here to 3.8 billion in the same period of last year. Last year, we did recognize a so called bargain purchase gain around 6.2 billion from the purchase of Heijan shares, but this year, without that, we see quite a big decline in the non-operating part.

But the net-net we still have around 7 billion in net income with 6.8% on net income margin. EPS for the first three quarters was NT$0.61 per share. On page six, which is a summary of our balance sheet and we still remain a strong cash position around 42.7 billion with a total asset of around 300 billion.

And on page 7, you can see a further detailed breakdown between our foundry business and solar related new business. And as I mentioned earlier, we actually post 2.9% sequential growth in foundry-related revenue in the third quarter. And gross margin for foundry operations in the third quarter was around 24%.

And on the next page, ASP in the third quarter is still relatively flat compared to the second quarter of 2014. Further revenue breakdowns starting from page nine, we can see the Asia revenue declined to about 44% from 46% in the previous quarter but the U.S. margin actually grow to 45% from 43% in the previous quarter.

On the next page, IDM represent around 9% of our total pie and on page 11, you can see the revenue breakdown by application – the communication pulls the strongest growth in the third quarter while computer actually show somewhat weakness, the percentage of revenue declined to 15% in the third quarter from 18% in the previous quarter.

On page 12, we are happy to see our 28 nanometer continue to grow strongly in the third quarter now, it is actually 3% compared to 1% in the second quarter and 40-nanometer also grow to 24% of our total revenue.

And in 65-nanometer, due to some upgrading of capacity, and also slower demand, right now, the percentage of revenue has come down to 26% in the third quarter from 31% in the previous quarter.

Page 13, we provide a quarterly capacity table here and one thing worth mentioning is our 12i in Singapore continue to do technology migration from 65 to actually 40 nanometers.

So the available capacity in Q4 of this year will show a mild decline and at the same time, we continue to add more 28 nanometer capacity in our 12A in Tianan site and that will mean the 28-nanometer had come out of the capacity.

So page 14 are – estimate capital expenditure for 2014 still stay around somewhat similar to 1.3 billion but the actual number in terms of cash payment will vary depending on the cut-off date by the end of this year. So that pretty much is the quick summary of UMC's results for the third quarter of 2014.

And 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..

Okay. Thank you, Chi Tung. Hello everyone and I would like to update everyone to UMC's third quarter operating results. In the third quarter, our foundry revenue grew 2.9% sequentially to 33.51 billion.

Foundry operating margin was 8.8%, the overall capacity utilization reached 93% led by increasing demand from the communication products such as handsets and tablet devices bringing wafer shipments to 1.462 billion 8-inch equivalent wafers, moreover, 24% of our revenue came from 48 nanometer.

While the 28 nanometer contribution grows from 1% to 3% quarter-over-quarter, demonstrating the sustained traction of UMC's leading-edge geometries. Our 28-nanometer year progress for quality sound and develop High K Metal Gate products has continued to improve which will drive significant production ramp during the fourth quarter.

This progress will help attract multiple waves of new customers and products to strengthen UMC's 28 nanometer growth and further diversify our customer base in advanced nodes.

UMC has collaborated with Asia Regional components for capacity deployment trends with the intention to expand our operating scale, strengthen efficiency and capture additional market share.

This cooperative model will establish regional manufacturing centers that will build economy of scale to increase productivity and serve local markets while helping UMC's global customers mitigate geographical risks in the supply chain.

UMC's recent announcement with Fujitsu Semiconductors and our joint venture in Xiamen illustrates these collaborative efforts.

Our foundry alliance with Fujitsu improves the 40 nanometer licensing agreement which will enable UMC to better serve the Japanese local market including automotive, industrial, consumer electronics, and other related sectors.

Recently, we also announced an investment venture with Xiamen Municipal People's Government and Fujian Electronics and Information Group to establish a 12-inch fab in China for 40 nanometers and 55 nanometer foundry services, subject to approval by Taiwan government authorities.

Meanwhile, UMC's dedicated engineering team continues to work on 10-nanometer and 14-nanometer advanced technology in our Taiwan headquarters to ensure the Company's next stage of growth.

Our differentiated approach via global expansion combined with the leadership in advanced manufacturing technologies, will serve as the engine that drives UMC's growth to enhance operating profitability and provide long-term returns for our shareholders. Now, allow me some time to summarize the recent highlights in Chinese.

[Foreign Language] And now, let me go over the fourth quarter 2014 guidance. The foundry segment wafer shipments will show a decrease of approximately 3%. Foundry segment ASP in U.S. dollars will increase by approximately 1%. UMC foundry segment gross profit margin will be in mid-20 percentage range.

Capacity utilization rates for foundry segment will be approximately 90%. We expect to recognize approximately $50 million from Fujitsu for 40 nanometer licensing fee. The guidance for the new business segments revenue will be approximately 2.5 billion and the net loss attributable to the UMC parent company to be approximately 400 million.

That concludes my comment. We’re now ready for questions. Operator, please open the lines up. Thanks..

(Operator Instructions). The first question is from Randy Abrams, Credit Suisse. Please ask your question..

The first question I wanted to ask about the China JV fab that you are entering and it is a few parts to the question, but I wanted to understand the – how the economics would work and so whether you would share profit proportional to ownership or receive management fees and so that is the first part.

And then the second part, if you could talk about time table to ramp it up, and the final part is how you prioritize ramping that fab versus your Taiwan fab and also the customers between your Taiwan operations versus the customers you target for the Taiwan operations..

Randy, for the Xiamen JV, the plan is we are going to start with one third of the ownership and so we are booked profit and loss accordingly in the beginning and our intention is to raise our holdings up to two third of the JV sometimes around 2019 to 2020.

And eventually, we hope we can acquire 100% of the operation subject to the Taiwan government's approval. And the operation in Xiamen will start with 55 and 40 nanometer, aiming for domestic market in China, especially for the somewhat customized or local required spec products such as simcard etcetera.

So we hope it will be an enhancement and launch our addressable market by having a local production capability or capacity to attract or benefit the Chinese kind of local requirements market and as for the priority, this is really equally important from UMC point of view, however, from a technology point of view, the Xiamen one will start from late 2016 at the earliest with 55 and 40 nanometer and the Taiwan site will continue to focus on 28 nanometer with R&D focus on 14 and 10.

So it is quite clear divided in terms of our effort.

From a financial point of view, the beauty of the JV is the financial requirement or the financial burden initially will be mostly carried by our partners in Xiamen and UMC will gradually increase our holdings by further injection of the capital starting from 2017 and hopefully, that will balance our overall tech requirement as a group..

Okay, and maybe one last follow up on that point. For the Taiwan government, in your initial discussions, does it look like any hold back on the approvals or how you expect the approval process to go through with Taiwan..

I cannot speculate the Taiwan government's decision process, however, the whole JV design and structure wise, all follows Taiwan requirements legally..

And the second question on – the recently, IBM was purchased by Global Foundries, is there any impact on the finFET, I think you had the agreement – and already paid for a lot of the license on the 14 and wanted to extend it to 10, but if you can talk about how that – if there will be any changes in the technology now..

Yes, there is no impact at all on UMC, IBM's 10 nanometer JDP program and UMC's 14 finFET technologies is quite on track and so our R&D in the next three to five years will still focus on 14 and 10 nanometer technology development..

And one final question, if I could ask on the CapEx, last conference, you mentioned you plan to do about 8,000 28 nanometer and 11k 8-inch, could you give an update on your plans and would imply for a rough range for CapEx for next year at this point..

We’re working on our 2015 budget right now, it largely depends on the market outlook and customer event, we don’t have the numbers for CapEx in 2015 yet. It will be available at the next earnings conference call..

Okay, would – are you roughly thinking about similar amounts of capacity for 28 and also continue to expand that Heijan – the China 8-inch fab..

Heijan is expanding to reach their full capacity or even a little bit beyond their full capacity of 50,000 8-inch per month, however, the CapEx month needed is a minimum compared to 12-inch.

And the 12A wise for 28, yes, we still plan to increase to 20,500 12-inch wafers per month for 28 nanometer sometime by mid of 2015 and beyond that, as I mentioned, we will follow the customer demand and our overall addressable market and so again, the number will be available next quarter..

Our next question comes from Bill Lu, Morgan Stanley. Go ahead, please..

First of all, on 28 nanometers, should I assume that you are sticking with the 5% target of revenues in the fourth quarter and hopefully, you can also update us on what your outlook is beyond the fourth quarter as far as revenue contributions and at what point that node could hit the corporate average gross margin?.

Okay, our 28-nanometer revenue contribution in the fourth quarter will be more than double to the third quarter. And we expect there will be a sequentially increase quarter-over-quarter on the 28-nanometer revenue..

Sorry, so the doubling is from 3Q to 4Q and the sequential increase beyond the fourth quarter, do I read that right?.

Yes, correct..

Okay.

Do you have a target for year 2015?.

Bill, I mean, it probably won’t be that meaningful, I mean we’re probably talking about a 3% range for the whole year 2014 --.

I meant 2015..

2015, we don’t have a target..

Yes, we expect the 28 sales in 2015 will be much higher compared to 2014..

And how about in terms of gross margin..

It is not available at the moment. Yes, we still maintain our previous statement on 28 nanometer margin improvement and so – as Chi Tung just mentioned by the second quarter 2015, our 28 nanometer margin will be close to trend's corporate average..

And following Randy's question on the IBM alliance, I just wanted to make sure that I understood the answer and so you have worked on 14 with IBM, you are part of the alliance for 10 nanometers, and so are you saying that when that changes ownership,, global foundries cannot terminate that alliance? Or how does that work exactly? Was there a certain time frame?.

The IBM JDP [ph] program is operated independently and so you are not affected by the Global Foundries acquire IBM's microelectronics business..

I see, so going forward, the technology developed will still be done on the UBM side..

Still by IBM and JV..

Actually the JDP expense will be completed by the end of this year and so the whole project in terms of JDP is completed by the end of this year..

Sorry, lastly, just kind of a housekeeping question, I remember the last call, you talked about 55 nanometers as a potential growth driver in the future, when that starts ramping, are you going to classify that under 40 or under 65?.

Yes, it will be a – take a longer time to recover the 55 demand and so we expect--.

55 will be in the category fourth quarter 55, rather than 40, it is different, so 55 and 65 will be in the same category..

So I guess I'm a little bit confused because you are doing conversions from 65 down to 40 and--.

In Singapore..

Yes and 65 as a percentage of total dropped by 5% this quarter.

Is that mostly temporary because of the conversion or how should I think of that going forward?.

It is a conversion from 65 to 40 is – comprise of our business engagement with customers and we have a – enough capacity in the Singapore site on 65..

And just one last question, the fab in Xiamen, I know it is for 40 and 55, but are those – is equipment going to be 28 capable?.

Well, so far the Taiwan government still doesn’t allow UMC to engage 28 yet..

Yes, I just want to ensure if that facility can make 28 nanometer chips or you can't have any tools that is capable of doing 28 or --.

You're probably going to have to ask them. We will keep our options open..

Your next question comes from Daniel Heyler, Bank of America Merrill Lynch. Go ahead, please..

Just a few quick ones. On the 28 nanometer situation coming back to that, I'm wondering to what extent are you anticipating some pricing action, TSMC talked about a cost down version of that node to obviously preempt competition from new players.

Are you anticipating a pretty aggressive pricing environment response to your most recent ramp up?.

Yes, I can only comment that our engagement is 28 nanometer customers, and so our higher 28 nanometer yield is increasing wafer demand, has given us more confidence to expand on the 28 nanometer capacity. And we believe that the 28-nanometer is a strong node that will deliver the better cost to performance solutions to our customers..

Okay, so for the time being your customers, you think are pretty sticky at this point? You have relatively good visibility into this ramp, I imagine.

I guess what I'm getting at is to what extent do you think that there is risk to this business – to your ramp rate in hitting your mid-year target next year? If, in fact, pricing were to fall, do you think you can hold onto this business because you're serving as a second source or are there other factors that you think this is a pretty sticky business?.

Yes, our near term deposits on the 28 nanometer business looks really solid. And as I mentioned, our – soon our yield improvements are gaining more traction. So we – demand is pretty solid on 28 nanometer..

So given that, I think since you've turned a corner here, I guess in volumes should we anticipate contribution to revenue to follow a pretty typical path? So say within four quarters this should be 20% of revenues? Is that a fair assumption at this point?.

I think it's too early to give that kind of commitment. But our CEO just mentioned, he does expect to see sequential growth starting from this quarter – next quarter. So I think that could be a target we are working to..

Okay, but not out of the ordinary. Okay. And then I wanted, Chitung to talk about some comments you made last time. You said that the eight inch capacity was pretty constrained and you're fully loaded and you were unable to deliver upside.

I'm wondering what you're seeing right now in terms of your eight inch loading going into the fourth quarter, it sounds like the rest of the business is declining more than single digit since your eight inch – sorry, since your 28 nanometers seems to be doubling to I imagine over 6%, 7% of revenue.

So what's happening in the non-28 nanometer and specifically eight inch?.

Eight inch wise is still in a similar position as a constraint and that's why we push Russia and China over the limit to have more capacity than their design capacity. So most of the weakness in Q4 probably coming from the mature 12 inch – say, 90.11 a little bit 65..

Yes, I wanted to focus more and get a better understanding of this 65 and 90 situation, because that seems to be where there's going to be the most excess capacity.

And how you can go about filling that capacity as – particularly as you have customers going for the 28, but what's your strategy to keep that part of the operation full?.

Yes, for the trends we see a – there will be a more and more application frankly from legacy note to a three inch, 65, 55 and 40 nanometers. So yes, we do see a lot of application coming on these nodes, 55 and 40, regarding the connectivity and the smart car, (indiscernible) car in the China market and also high voltage driver, the LCD drivers.

So lots of applications will – we are expecting they will migrate into these nodes. So we are looking for the second half of 2015 will have a better improved evaluation on those nodes..

Okay. Sounds like it's China driven and then finally a quick one on the housekeeping side, Chitung. This impairment charge, it was 792 million last quarter, 580 million in the September quarter.

Can you give us some guidance for modeling reference for the next couple of quarters and how big this is going to continue?.

Hopefully it will be done in the near future and this is highly related to the economic cycle of our solar business. We’re experiencing slowdown due to the regulation change in the third quarter, that trigger, solar related, as an impairment loss in the third quarter. But again, I cannot forecast that.

This is really enforced by all the regular quarterly impairment tests we have to conduct every quarter..

So could you elaborate this is specifically driven by pricing of your assets? Or what – walk us through the calculation on the impairment charge..

It's coming from our solar subsidiary NexPower. The outlook or the cash flow forecast has been dampened due to the change in the overall macro environment of solar markets and largely due to the regulation change..

Regulation change in what tariffs?.

Imposed by U.S. market and in that I mean tax..

So this isn't really impairment, this is loss – a net loss in the business?.

Yes, but that will trigger the whole five years cash flow rolling forecast to have a parallel shift..

The next question is from Andrew Lu, Barclays Capital. Go ahead please..

A couple of questions from my side, I think on Q4 the guidance on 28 pretty much probably sticks to 7%, that kind of number.

Do you have a breakdown on high-K metal gate or poly-SiON?.

Yes, we do. The high-K, metal gate ratio will greater than 50% of the 28 total..

Is high-K a higher margin compared to poly-SiON?.

Yes..

The second question on the Q4 guidance, 3% down on the wafer shipment.

Can you elaborate which application products are weaker than average, which ones are stronger than average?.

Yes, for the consumer is the weakest one, yes, quarter-over-quarter, especially on CMOS image sensor and GPS and communication is almost flat and computer is almost flat..

CMOS image sensor is the one fall the most. Okay. Can you talk about the 14 nanometer FinFET based on I checked – UMC is looking for two tile version. One is a TSMC-like and one is an IBM-like.

So have you decided which one you go direct? Like a (indiscernible) 90 or a 78 nanometer?.

Yes, our 14 FinFET, we combined the – pressure [ph] process with (indiscernible) integrated with IBM's 10 nanometer fundamental learnings, Came from the GADP and yes, our – the process and schedule is on track, yes. We’re ready for – we plan to have – to be available for customers stable in the first half of 2015..

In terms of guidance for Q4, we can see a mid-20% margins are much better than 21% margin in Q3 and apart from the low season reduction from solar business which you guided lesser – 900 million. So Fujitsu license, I think, also contributes some. I just wonder whether this Fujitsu license will continue into Q1 next year or next year.

Chitung Liu Andrew, I need to make a clarification on that. For foundry operation in Q3, the gross margin was 24%. For foundry operation in Q4, we guided mid-20% gross margin. Excluding --.

Oh that's for foundry? Chitung Liu Yes, excluding the Fujitsu license fee. So Fujitsu license fee is excluded from the mid-20 gross margin guidance, and that would only be in Q4 of 2014, nothing in Q1..

Po Wen Yen earlier mentioned migration to (indiscernible) 90, 65, 55, 40 from legacy.

Are you referring legacy from eight inch or like a 0.11 to 90? Are you referring the customers migrating to 65, 55, 40 from eight inch? Or you are talking about legacy part of the 12 inch?.

It's partly coming from the migrations from 130 or 111 nanometer. And many is actually we have some new business engagements..

Yes.

So do you see this possibility for eight inch legacy to migrate to 12 inch foreseeable future? Is there any technology difficulty to do that? Or advantage or cost disadvantage not going to the 12 inch?.

I would – from my perspective, there are only a small portion of those products looking for a better performance or higher demand. So there are some migration from those products and yes..

So big volume one and want performance, they want to go to 12 inch. If it's small quantity, small volume and they don't care the performance, it will stay at eight inch.

Is that correct?.

Yes..

The next question is from Roland Shu, Citigroup. Go ahead, please..

First question is for the CapEx spending this year is 1.3 billion. However the overall capacity grows only about 2% to 3%. I know it depends on your – this year you're doing a lot of the conversion.

So can you give us some colors, how many capacity of the wafers you have been doing for this conversion? And how many new capacity you buy or you're putting in this year? Thank you..

Yes. I think that the conversion only take a very small portion. Most of our CapEx is actually increased our capacity..

Yes but do you retire some old legacy capacity? Because if I look at the overall capacity, we see it only grow about 2% to 3% and with this and your 1.3 billion CapEx spend, I think this actually – I think the capacity growth looks too small..

Our capacity we increased 4% year-over-year.

Yes, 4% and with this 1.3 billion CapEx spending?.

Yes. And the three inch, we actually – we increased 6%, yes from 2013 and 2% on eight inch..

Okay, second question for the 28 nanometer gross margin.

Correct me if I'm wrong you said in second half next year – 28 nanometer growth margin is able to bring up to 12 inch corporate average right?.

Yes..

So the question is, is your 12 inch gross margin higher or lower than the corporate average?.

It's a few percentage points lower than the corporate average..

Okay, so 28 nanometer gross margin in second half next year will a few percentage lower than corporate average, right?.

Yes..

Your next question comes from Steven Pelayo, HSBC. Go ahead, please..

I know you don't have a specific number for next year's capital spending but you do have some targets here. You want to add about 8k of 28 nanometer, you want to 11k of 200 millimeter, you have some investments I guess in China and Fujitsu to make as well.

I guess I'm just trying to figure out what's a baseline number to work with? One would think that just adding an extra 8k, maybe it's a little bit more, 28 nanometer, we may get to the 1 billion alone there.

So can you just help us walk through how much would it cost just to add the 8k, the 11k, whatever small portion you're starting with in China and then just (indiscernible) portion? Chitung Liu Again, China is a JV so we will pull in aiding capital instead of CapEx.

CapEx will be paid in that particular JV and for the baseline numbers; it's always a good reference to use our free cash flow, roughly $1.5 billion, $1.6 billion per annum..

Okay. All right that matches what I was thinking. Actually what was the last thing I wanted to you? Oh, you had talked before – I don't know if it was last quarter or it was the quarter before about (indiscernible) as being required to get to this corporate average margin.

So if by mid-next year you're at this over 20,000 wafers is that the kind of scale that you can do? Mid-20s and above type gross margins? In 28 nanometer? Or do you need it to be even bigger? Do you need to exit the year at 30k of capacity?.

Well the more the merrier, of course. But the funny case is minimum magic number in terms of economy of scale, but again, that will help us to definitely hit the 12 inch average gross margin. But how to improve the overall 12 inch gross margins compared to the corporate average, that's another level of enhancement..

All right and my last question is just on that free cash flow. I think it's been negative for a couple of quarters here. I assume with the big CapEx that's still left to pay for this year, it's going to be negative again in the fourth quarter. Yet you talked about next year's CapEx target being related to that free cash flow ability.

Chitung Liu Managed from operation, so cash inflow from operations..

Okay, but what about – on a free cash flow basis for next year, do you think you're going to be negative free cash flow again?.

I cannot forecast on that. But if I need to rephrase, that will be our cash demand..

The next question comes from Lingling Hu, Goldman Sachs. Go ahead, please..

This is Donald on Lingling's line here. My first question is just to clarify Andrew's question. In Q4, your gross margin guidance of foundry business is mid-20% and then I think Mr.

Yen commented – does that include the $50 million from Fujitsu or not?.

Not. It doesn't include..

Doesn't include.

So basically your gross profit margin is somewhat flat in Q3 and Q4?.

That's correct..

Okay, good. And the second question is also to clarify. Is that the 28 nanometer gross profit margin when you have 20k capacity could potentially reach the corporate average.

Is that the 12 inch corporate average or the overall foundry corporate average?.

The 12 inch corporate average..

12 inch?.

Yes..

Okay, great. Another question is on the – first, congratulations for the 28 nanometer high-K metal gate ramp. It's not an easy thing.

How many customers are right now doing this process with you in Q4 in terms of volume production?.

Right now we have more than 20 customers engaged on UMC's 28 nanometer technology. And we have more than 50 charter tip-offs and five of the charters are in production now..

So this is for high-K metal gate or for the entire 28?.

They are both together, poly-SiON and high-K metal gate..

I see.

And for high-K metal gate, are you expecting more than one customer let's say, in the next quarter or two?.

Certainly. Yes..

Another question is more – for the JV in Xiamen. You said over the long term you want 60% of that $6 billion potentially $6 billion JV.

And where would be the cash coming from? Are you going to raise money through that offering or issuing new shares?.

That's a very long term plan. As I mentioned earlier, we will start from one third of the ownership with minimum contribution for paid-in capital. And eventually, we will grow up to 2/3rds of the holdings and ultimately we hope to hold 100%.

And once we reach cash flow breakeven, hopefully it will be able to generate enough cash to fund the whole investment. So it won't cost whole $6 billion for the whole project. A large portion of the investment amount will be self-financed by that particular project..

Okay.

And then the local government will – you will buy out the stake of local government?.

Yes..

My final question is on your – last quarter you announced a 55 nanometer embedded flash process. I think that's good for smart card etcetera.

Have you got any customers (indiscernible) out or doing really any new real product with that process?.

Yes, we do. We have, yes..

Okay.

Is this a Chinese customer?.

Yes..

The next question comes from Szeho Ng, BNP. Go ahead, please..

I just want to ask how you are going to account for the Fujitsu JV investment.

Is it on cost counting?.

No, because we will have the control in the Board, so it would be 100% consolidate into our book..

Okay, all right then.

Is it possible to give us some color on how to model the revenue and also the earning contribution etcetera?.

The initial start production won't start until the end of 2016..

Okay, all right.

So near term should be no impact at all apart from those (indiscernible) that you're going to incur?.

Yes..

And then secondly, for the solar business, after the impairment charge in Q3, should we expect their cost structure to be a lot more competitive? And if that's the case, what will be the breakeven revenue for that operation?.

Unfortunately not, so the impairment won't help the cost structure too much. Or it won't have big impact on the P&L. Instead still the outlook or the demand and supply for the solar market will have a bigger weighting in terms of the new business performance..

And last question, is it possible to share with us the operating margins for the foundry business? And also for the back [ph] in Q3?.

Yes, it was 24%..

That's gross margin. .

Oh, that's gross margin, yes. We never really disclose the numbers. But you can do a rough estimate..

The next question is from Eric Chen, UBS. Please ask your question..

And probably two quick questions, and the first into the 28 nanometer process and we talked about that we may have the 20,000 wafer capacity. And Chitung, could you give us the breakdown in terms of the high-K metal gate and the poly-SiON percentage? And we know right now over 50% go for high-K metal gate.

So once we have a 20,000 wafer and for the 28 nanometer, the capacity, percentage in between the high-K metal gate and the poly-SiON and what will it be? Thank you..

Okay, we actually can give you the more detailed numbers. For our 28 nanometer capacity in the 2Q 2015 is – the high-K metal gate version will take 12,500 and the 8000 is for poly-SiON..

And the other one is regarding to the JV.

So the – Chitung just mentioned we – eventually this JV and the total investment probably the $6 billion and the what kind of capacity we talking about investing?.

50,000 design capacity..

50,000 design capacity.

And based on 50,000, the wafer capacity that probably won't cause so much money, right? Less than a 10,000 wafer probably costs, like $120 million, $130 million, right?.

This is 10 years-plus plan and based upon future potential upgrade etcetera that we come out with this number. So it has no obligation whatsoever we have to invest this amount of number if we can generate good returns. So this is a good number to follow but right now I would say there is still some variables in there..

And one very quick clarify, and you talked about the ownership initially one third. I assume that will probably be for the coming two years. And two third in year 2019 to year 2020.

So it's different from we saw in the newspaper, the 60% ownership you are going to have in the mid-term, right?.

Yes, it should be the same. But I think my version is more accurate..

The next question is from Rick Hsu, Daiwa. Go ahead, please..

My first question is about your depreciation for next year.

Can you give us some color?.

Depreciation this quarter will be about 5% or less. I mean 5% more compared to the previous quarter. And for the whole year it's also about 5% more than the previous year..

Okay.

What about next year? Do you have any idea, roughly?.

Yes, next year is not available yet..

Okay. The second question is about your dividend policy.

Do you have any plan to pay how much cash dividend for next year?.

I think we have the intention to maintain our consistent dividend payout..

Consistent dividend payout, is that right?.

Yes, compared to the last few years..

My last question is, if we talk about your 28 nanometer revenue contribution, what level your 28 nanometer will reach your corporate average margin? You talk about 20,000 wafers per month before, when you reached your corporate average.

But when I – if I translate that into revenue contribution, would that be 15% or 20%?.

Well again that depends on the overall revenue for UMC. So it's not only the parameters of the 28 nanometer alone. Overall revenue and supply need to be taken into consideration as well. So we don't really have a breakdown for now..

Okay, but I can fairly assume that – because you're talking about your 28 to reach corporate average by some time mid-2015.

So I can assume by mid-2015 your wafer shipment per month will be roughly 20k for 28 nanometer, right?.

Corporate average in 20 – 12 inch operation by mid of 2015..

We are running out of time so we are taking the last question. Andrew Lu, Barclays Capital..

A follow-up question very quickly.

The first one is regarding the Fujitsu 50 million royalty, are we going to put in the revenue line or non-operating side?.

Starting from top line, all the way down. So both top line and gross margin will be roughly 50 million. On the operating expense level, there will be some portion goes to provisions of earnings – employee bonus will be budgeted. Some percentage of the 50 million will be budgeted as incremental employee bonus..

And second question is, please remind me for Q4 this year, the 28 nanometer will be how much capacity? Is it 12k?.

Yes, it's 12k..

It's 12k?.

Yes..

No change, thank you..

No change. 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 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. Bye..

Thank you ladies and gentlemen. That concludes our conference for third 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..