聯華電子 (UMC) 2013 Q3 法說會逐字稿

Welcome, everyone, to UMC's 2013 third-quarter 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 has 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.

Thank you and welcome to UMC conference call for the third quarter of 2013.

In attendance today is the CEO of UMC, Mr. Po-Wen Yen, and the CFO, Mr. Chitung Liu.

To start things off our CFO will present third-quarter financial results, followed by our CEO's key message to address UMC's main area of focus and UMC's fourth-quarter guidance.

Once our COs have complete their remarks, there will be a Q&A session.

UMC's quarterly financial reports are available at our website www.umc.com, under the investor relations financial release section.

During this conference we will 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 control -- beyond the Company's control.

For those risks, please refer to UMC's filing with the SEC in the US and the overseas securities authorities.

I would now like to introduce UMC's CFO, Mr. Chitung Liu, to discuss UMC's third-quarter 2013 business results.

Thank you, Bowen.

Hello, everyone.

For the third quarter of 2013, revenue was NTD33.41b, with gross margin at 22% and operating margin at 7.2%.

The net income attributable to the shareholders of the parent was NTD3.48b and the earnings per ordinary shares were NTD0.28.

And if you refer to our earnings presentation, you can see on page 3 our operating results summary.

Our third-quarter utilization rate was 87% that confirm 1.329m 8-inch equivalent wafer shipments over our quarterly capacity of approximately 1.5m per quarter.

And cash on hand in third quarter -- at the end of third quarter was NTD50.3b, with total assets approximately $10b.

If you turn to page 4 you can see our third-quarter income statement.

And, as we mentioned earlier, revenue grew 4.7% to NTD33.4b.

That comes from over 2% increase in ASP and over 1.5% increase in wafer shipment.

And gross margin was 22% or NTD7.3b.

We continue to tighten our cost control.

So operating expense has shown 2.5% quarter-over-quarter decline to NTD4.89b.

And operating margin as a result has improved from 3.6% in the previous quarter to 7.2% or NTD2.39b in the third quarter of 2013.

In quarter three we always collect the majority of our investee dividends.

So, total non-operating income in third quarter was NTD1.56b.

And as a result, our net income attributable to the stockholders of the parent in third quarter was NTD3.47b or an EPS of NTD0.28.

If we turn to page 5, this first nine-month performance, a year-over-year comparison, our revenue grew about 7.2% year over year, which is mainly due to the combination of revenue from He Jian.

And gross profit is 19.3% or NTD18b.

Our operating income, around NTD3.8b or 4.1%, which has shown quite an improvement from the previous year.

And the EPS for the first nine months is NTD0.95 per share.

On page 5 (sic - page 6) there's highlights for our balance sheet.

And other than the NTD50b cash, and we also have about $10b in total assets and NTD211b in stockholder equity.

For operating segments, on page 7, most of our revenue and earnings coming from, of course, the foundry operation.

And new business, which mainly composed of solar and some LED investment, is only about 5% of the total revenue and still in the loss-making condition given the tough competitive environment out there.

And for the earnings for foundries in third quarter is NTD3.5b.

For average selling price in the third quarter, it went up about more than 2%, mostly due to higher portion of leading-edge wafer shipment.

And if we turn to page 9 for revenue breakdown, as you can see that we had 44% of revenue coming from Asia.

And Japan has shown quite an improvement; it improved to 6% in third quarter from 3% in previous quarter.

IDM also shows a 4% increase from 10% to 14% in third quarter.

On page 11, the revenue breakdown by application did not have too much change.

Computer is 16% and communication is 52%.

And we continue to make way to finer geometry.

And 40-nanometers is still about 20% of revenue.

And total revenue below 55-nanometers has increased to 54% in third quarter from 51% in the second quarter.

And capacity, as I mentioned earlier, is approximately 1.5m 8-inch equivalent per quarter.

And we continue to add more leading-edge capacity.

In the third quarter and fourth quarter most of the capacity increase is coming from our Singapore 12i fab in the 40-nanometer capacity.

And that concludes my summary of UMC results for third quarter 2013.

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, Chitung.

Good day, everyone.

I'd like to update UMC's third-quarter operating results.

The foundry segment recorded NTD31.6b in revenues, with profit margin from foundry operations of 9%.

And wafer shipments hit an all-time high of 1.329m 8-inch equivalent wafers, bringing overall capacity utilization to 87%.

The new business segment recorded NTD1.83b in revenue.

40-nanometer business continued to grow and accounted for 20% of sales in third quarter 2013.

During the last couple of quarters, the engineering group has focused relentlessly on yield enhancement activities to improve our 28-nanometer manufacturing technologies.

These accumulated efforts enable 28-nanometer poly/SiON and high-k metal gate yields to accelerate its climb.

This accomplishment allows our pre-production customers to shift closer towards volume production and concurrently shortens the time required for new customer adoption of UMC's 28-nanometer platform.

In the upcoming months we look forward to further optimizing 28-nanometer yield levels to move more customers towards volume production.

In terms of specialty technologies, UMC continues to lead the foundry sector in the high-voltage manufacturing processes used for applications such as driver ICs in mobile devices.

Our unique high-voltage technologies increase our customers' competitiveness so they may further gain market share in today's growing display industry.

For the fourth quarter, we have seen a decline in wafer demand.

Primary factors are -- that lead to a weakened outlook stems from seasonal correction and supply chain inventory control and the uncertainty culminating in macro economy.

We firmly believe UMC is well positioned to benefit from the robust 28-nanometer demand in the mobile industry for the next several years.

To secure long-term foundry growth, UMC continues to deploy our advanced logic and mixed-mode and specialty derivative technology offerings.

UMC's diversified product portfolio will enhance our manufacturing flexibility to accommodate additional customer production requirements, broaden our business scale and solidify our position as one of the leading players in the foundry industry.

I'm now moving to fourth-quarter 2013 guidance.

Our fourth-quarter 2013 guidance are as follows.

The foundry segment wafer shipments will decrease 8% to 10%.

Foundry segment ASP will decrease by approximately 2%.

UMC's operating margin to be approximately breakeven for the foundry segment.

Capacity utilization rate for the foundry segment will be in mid-70% range.

2013 CapEx for foundry segment will remain unchanged at $1.5b.

Guidance for new business segment revenue will be approximately NTD2.2b and operation loss will be approximately NTD550m.

That concludes my comments.

We are now ready for questions.

Operator, please open the lines now.

Thanks.

(Operator Instructions).

Randy Abrams, Credit Suisse.

Yes.

Good evening.

My first question, I wanted to ask on the 40-nanometer how much continued momentum or potential for an additional or second-wave business, if you expect that to continue to grow as percent of sales.

And factored into the small ASP decline, is that due to product mix, where perhaps 8-inch or older nodes are holding up more than the 40-nanometer into fourth quarter?

Our 40-nanometer business has been a steady contributor and continues to be around one-fifth of foundry business.

So in today's environment, UMC has not felt too much pressure on 40-nanometer business.

Okay.

Do you expect continued -- like another second wave where that will grow from the current 20%, or would the incremental growth now be 28-nanometer as you look into the next year?

Yes.

I would say the second wave is still -- we still have a great opportunity to grow our 40-nanometer revenue share.

And yes, I think it still continues to grow in the coming quarters.

Okay.

And I wanted to ask on the CapEx, with utilization falling to 75% by fourth quarter, how's your -- how should we think on initial view on CapEx if the $1.5b is the right rough level for spending, whereas where utilization is, can you bring that down into next year or you need to invest to bring up 28 and more 40.

Randy, maybe I can answer this question.

Our cash flow pretty much can support $1.5b-ish CapEx per annum without really additional funding requirement.

So $1.5b is always our baseline.

And if we can do better in terms of the technology development in terms of schedule, we can do a little bit more than that, and vice versa.

So $1.5b is the number we feel comfortable financially, and that's the number we will be using as a benchmark in the near future.

Okay.

And my last question, just the two areas.

One, Japan took a step up.

Maybe is that a one-time bumper?

Do you expect some continued incremental IDM business from Japan?

And if you could give a sense, where you talked on specialty and high voltage, just where that's at per percent of revenue now.

Yes.

Our Japan's -- the business growth mainly coming from -- not from high voltage, it's from the ISP.

Do you expect -- that jump from 3% to 6%, is that a new level or even a level you see or project continuing to grow?

It's been low single digit for a number of years, but is there finally more potential for outsourcing to continue growing there?

It depends on our customers, on their business situation.

It depends on customer demand.

So you are not one-time jump.

It's we expect to continue to maintain the business growth in Japan.

Okay.

No, that's good.

And the specialty process, what percent of revenue is from specialty and high voltage?

Yes.

Our high voltage is around 40% of our total specialty revenue.

Okay.

And specialty as a percent of total sales?

Around 30% of UMC's total revenue.

Okay.

Great.

Thanks a lot.

Thank you.

Steven Pelayo, HSBC.

Great.

Thank you.

Just a reminder on 28-nanometer, I think before you were talking about a first few percentage of revenue by the end of the year.

Is that still what it's tracking to?

It sounds like you're making great improvements on yields.

And can you maybe also provide some color on number of tape-outs or number of customers?

How much is initially going to be poly/SiON or do you have some initial high-k metal gate as well?

Could you just give us a little bit more detail on 28-nanometer, both in terms of qualitative as well as quantitative in the mix over the next quarter or so?

Yes.

Our 28-nanometer production has started to gain some momentum.

And however, we still maintain our 28-nanometer revenue share will be in a low single-digit target at the end of 2013.

And --.

And any more details on number of customers or number of tape-outs?

Anything else you can provide some insight?

We -- yes.

We already engaged the more than 20 customers on 28-nanometer node.

And more than -- greater than 30 product tape-outs, yes, so far.

And about five -- more than five, yes, more than five products are in pre-production stage.

Okay.

Excellent.

I guess I want to follow up a little bit on Randy's question.

He was asking a bit about ASPs being down in the fourth quarter 2%.

So obviously 28-nanometer ramping up, that should be good for ASP.

So maybe you could talk a little bit about what's going on and why ASPs are down in the fourth quarter.

Is this on a like-for-like basis?

Is it product mix?

Can you help us understand the ASP decline in the fourth quarter given 28-nanometer starting?

Yes.

I think it's all -- it's mainly due to our shipment, this drop more in 12-inch wafers.

So it's less 12-inch wafer shipments that caused our ASP drop.

That is the main reason.

Okay.

And I know you've guided for the fourth quarter a mid-70s utilization rate.

Does that then imply that your 300-millimeter utilization rates are in the 60s or maybe you could talk a little bit about what that means from a 200 millimeter versus 300 millimeter given a mid-70s blended average.

Yes.

Our 12-inch wafers, our fabs, the loadings are, yes, greater than 70%, the low 70% range.

And 8-inch is around 80%, greater than 80% range.

Okay.

And last question from me is just I'm curious on your 40-nanometer margins.

You talked about revenues continuing to maybe grow from here next year.

Is the margin for 40-nanometer already above your blended average there at 22% for gross margins?

No.

Majority of the 8-inch facilities have been fully depreciated.

So for most of the new investment, from an accounting margin point of view it's very difficult to compare.

So it's not to compare average yet in terms of 40-nanometer margin.

Is that, you know, just a capacity thing, a timing thing, a volume?

When would you expect 40-nanometer to not be a drag on your margins, I guess?

Again, back to the accounting margin, it may have to be to the end of the depreciation schedule, which in our case is about five years, yes.

Understood.

Understood.

Great.

I'll get back in line.

Thank you.

Thanks.

Thank you.

(Operator Instructions).

Eric Chen, Daiwa.

Yes.

My first question regarding to the 28-nanometer process, I would like to know when we will see the revenue and over 5% of total revenue.

And also I would like to get idea the capacity.

And I remember you -- so far you guided like 10,000 wafers per month for the 28-nanometer process and so do we have any plan, as at end of the second quarter, what kind of capacity we should expect?

And that's it.

Eric, maybe I can jump in here.

For 28-nanometer, we still stick to our previous target, which is low single digits by the end of this year.

So hopefully, if everything goes well, we see low single-digit quarterly revenue in Q1 next year.

And beyond that it's really difficult to predict so I cannot really commit here when we're going to see the 5% [quarterly] revenue.

Of course, from management's viewpoint, it's sooner the better.

Okay.

How about the capacity -- capacity-wise?

Our capacity, you mentioned it's 10,000 wafers per month existing in 2013.

And our capacity target will depend on the results of our engineering collaborations as well as the amount of demand.

We'll keep track on this.

Okay.

So in that case, do we need to expend a high CapEx for next year once your 28-nanometer process, the yield rate, also the commercialized shipment and the catch-up first half next year?

For next year.

First of all, CapEx number for next year will be announced or disclosed in the next earnings call.

And as I mentioned earlier, our baseline CapEx largely depends on our free cash flow, which is, at current conditions, is around $1.5b.

And if we see greater opportunity, of course, we will be more aggressive.

So at this point, we will plan our capacity based upon multiple factors, including the cash flow.

Yes, very clear.

And in that case, the base, the $1.5b, let's assume $1.5b as CapEx, how big the capacity for the 28-nanometer process will be per month in second half next year?

Well, it's hypothetical so it's difficult for us to answer.

You know better than us how much it costs for 1K 28-nanometer high-k capacity.

So you can do a quick calculation.

But we also have other things we need to take care of.

For example, we are converting some of our copper-based 8-inch capacity to aluminum.

We are also installing additional 40-nanometer capacity in our Singapore facility.

And on top of that we are finishing up our new shell in Tainan Science Park.

So it all counts as part of our CapEx.

Okay.

I see.

Okay.

And last question regarding to the He Jian, and could you give us the highlight in terms of the capacity there or any the expansion planned in He Jian and for the China business?

Thank you.

Yes.

Our He Jian fab currently is 45,000 wafers per month.

And we are expanding up to 50,000 wafers per month by the end of this year.

Okay.

What kind of products are made in He Jian right now?

It's driver IC or what else?

Yes, correct.

Yes.

Okay.

Driver IC mainly.

A good portion is driver IC, yes.

Yes.

By the way, you talked about driver IC would be the high-growth product.

What kind of geometry, I mean for UMC to keep gaining the market share, what kind of geometry does UMC need?

And also in terms of revenue contribution, the main geometry for the driver -- what's the main geometry for the driver IC?

Thank you.

Yes.

We -- yes, because the demand situation, there's some changes from the with-RAM driver IC to the RAM-less driver IC.

So the recent change makes our demand move to legacy node technology node.

So I'm probably not going to detail the technology node, but anyway, this is the big trend here.

So geometry migration, right?

This is mainly to 0.11 micron.

Okay.

That's all the major geometry for the driver IC?

Yes.

Thank you very much.

I'm sorry?

Yes.

Okay.

Thank you.

Thank you, very clear.

Thank you.

Thank you.

Donald Lu, Goldman Sachs.

Thanks.

My first question is on depreciation.

Chitung, if your CapEx is on the baseline next year, can you give us a sense on how much depreciation will be increased year over year next year or decreased?

It will be increased about less than 10% from this year.

So less than 10% year over year next year?

Increase.

Increase.

Okay.

The second question is on the R&D spending.

How much percent would that be after your [annual] if there's any guidance in terms of the degree of intensity there.

I remember in the last call you commented it might increase.

Current conditions include the 10-nanometer FinFET development with IBM.

And this project should be concluded by the end of next year.

So from now to the end of next year it will be the current condition, current percentage.

And hopefully we see some mild decrease beyond the end of 2014.

Okay.

So the R&D as a percent of revenue would be the similar percentage as in Q3 or Q4 level into next -- in next year.

Q3, Q4 are very similar.

Are similar as a percent?

Yes.

Okay.

And just relating to that in Q4 your guidance is for operating loss.

Is that all from the new business?

Or is there anything special there?

Foundry only is breakeven.

I think it will be profitable.

And including new business, it probably will be very mild, very small operating loss if we include the solar investment, yes.

Okay, so that's the guidance.

The small loss would be from the solar business.

Yes.

And how is the revenue trend for the solar business in Q4 (multiple speakers)?

-- give a number of flattish, a small increase around NTD2.2b in Q4.

I see, okay.

Great, thank you.

Thank you.

(Operator Instructions).

Bill Lu, Morgan Stanley.

Yes, hi.

Good afternoon.

Going back to your guidance for the fourth quarter, are you guiding for breakeven for the semi business?

And I'm not sure if my math is right but that seems to imply a gross margin in roughly the mid teens.

And you also said that 12-inch margin is lower, it's dropping a little bit faster than the 8-inch.

So it seems to me like margins shouldn't have dropped that much if it's only utilization rate.

Can you talk about what are some of the reasons behind that?

Is it 28 and 40 ramping?

Or what exactly is causing that?

Thanks.

First of all, the currency is moving against us.

So NT dollars has appreciated quite significantly over the past few weeks.

And we also see ASP drop mostly due to 8-inch drop in shipment.

And our depreciation continued to increase, so does the electricity utility cost in Taiwan.

So altogether the cost structure has -- getting a little bit worse in the fourth quarter.

Okay, that's helpful.

You talked about seeing additional opportunities in 40-nanometers going into 2014 as far as share gains.

Can you help me quantify that?

How much can 40-nanometers grow year on year next year or what is it as a percentage of revenue?

However you want to quantify, it is fine.

Yes, we do see the second waver customers, their demands is still there and their peak revenue, peak demand is in 2015 and 2016 timeframe.

So we will continue to see and to engage with those customers to grow our 40-nanometer share.

Maybe I'll ask it a little bit differently.

If you look at the second wavers and the size of the opportunity versus the opportunity for the first wavers, what is the difference?

On what?

Just in terms of the number of wafers that's required by the second wavers versus the first wavers.

Yes, for this, the second wavers, some -- the scale is still pretty big and it's not small.

It's very big compared with the first wavers.

So similar in scale?

In total, second wave probably very similar in scale.

Okay, that's very helpful.

And then the last question is I wonder if you have visibility into the first quarter of next year, just given that I think the expectation is that we're going through an inventory correction, but that's potentially coming to an end.

You're ramping up 28, 40 and things look pretty good.

How does the first quarter look?

It's quite difficult to provide an answer on first quarter next year during the end of October.

So it'll be important for us to tune into the market and to pay attention to see the trend.

So we will need more time to observe the first quarter 2014's outlook.

Okay, great.

Thank you very much.

Thank you.

Szeho Ng, BNP.

Yes.

The Company turned into net debt in Q3.

So just want to know what sort of financial gearing UMC is comfortable with?

Debt to equity ratio, our level is around 30%, which is the current condition right now.

So we would like to stay around 30% for the foreseeable future.

Okay, all right.

And then for the tax, I just want to know what sort of number would be an appropriate one to have for modeling purpose going into 2014 and 2015?

Well, eventually the corporate tax is 17% and we still enjoy some kind of tax benefit.

So I think from the current point to maybe two, three years timeframe, eventually it will go up to 17%.

Right now I would say 15% or so is probably the number you should model in.

Okay, all right.

Thank you.

And last question with regard to your margin guidance for Q4, do you incorporate any one-off items in it or it's purely just the normal one?

It's just the normal range, nothing one-off.

Okay, all right.

Okay, thank you very much, Chitung.

Thanks.

Dan Heyler, Merrill Lynch.

Yes, thank you for taking my question.

Chitung, I have a question on the R&D comments that you made relatively -- I think you said relatively stable in 2014 versus the second half of this year, R&D as a percentage of sales.

How are you accounting for the IBM project?

Is that amortized on a quarterly basis?

And if so, how many quarters is that affecting the R&D number?

As I mentioned, the whole project will be concluded by the end of next year.

So the impact will be evenly distributed every quarter until then.

Okay, excellent.

And then on the capacity comments that you made for -- I know it's early to make a call on needed capacity for 2014 because that depends a lot on your 28-nanometer yield improvements or not.

Do you have any thoughts though on the specialty business?

How much capacity do you think you need to add there?

Specifically would you be interested in adding 8-inch capacity for specialty or would you be adding 12-inch capacity for specialty?

Maybe just give us some update on the specialty capacity requirements and whether you need to add wafers or not?

For 2013, our capacity expansion on specialty for example, the 8-inch wafer is around 9% of our total CapEx.

So it's 91% on 12-inch wafers.

Have you added any capacity at all on that 9% or is that all process related?

Yes, as I just mentioned we added our 8-inch fab in Suzhou from current 45,000 wafers per month to 50,000 wafers per month.

Right, okay.

And will you add more next year for 8-inch?

And if so, how do you look at maybe acquiring used equipments from IDMs?

Or how aggressive do you want to be in 8-inch?

Our Board actually has passed a resolution or agenda asking the management team to find some instant capacity in Asia, either 8-inch or 12-inch and mostly trying to fulfill the demand for specialty.

But there's no definite schedule on that kind of search.

So we're still looking for opportunity out there, but it's not really limited to any possibilities.

Okay.

And within your 8-inch specialty process are there and if so, how much can that be moved to 12-inch?

Are there any areas where you're seeing, any product areas that will be moving from 8-inch to 12-inch specifically?

Maybe you could tell us high voltage or areas where you could create more capacity on 8-inch if you needed to.

Yes, just like you mentioned, our high voltage, there's -- some capacity will migrate from 8-inch to 12-inch, 65-nanometer technology.

And there are some from the embedded and (inaudible) memory products also will migrate to 12-inch technology.

Okay.

So is that what's creating your -- that's what's your creating your demand for 40 is that you have customers that are mostly migrating from 65 to 40?

Is that where the majority of your first wave adoption is coming from?

Is it the migration from 65 to 40?

Yes, but it's not majority, but it definitely has some contribution.

Okay.

So you'll be able to use your excess 65 capacity, you'll be able to use that for high voltage?

Is that right?

Yes, correct.

Correct, okay.

And then when you talked about second wave adopters, I was just trying to understand what you're referring to.

Usually this is just the people on 65 that haven't migrated forward.

Is that what you're anticipating is a lot of your 65 now coming forward to 40?

Is that what you're referring to on second wave adopters?

Yes, correct.

Okay, great.

Thank you very much.

That's it from me.

Aaron Husock, ShearLink Capital.

Great, thanks for taking my question.

Could you clarify on your R&D expense, why it wasn't down compared to Q2 given that Q2 had the 28-nanometer start-up cost in the numbers?

Hi, Aaron.

We joined the IBM co-development program for 10-nanometer FinFET and that program won't conclude until the end of 2014.

So each quarter we will have to pay some payment to this joint development.

So that's the reason for the next few quarters, until the end of 2014 our R&D expenses will stay at the current level.

Okay, great.

And then your depreciation in cost of goods sold was down sequentially.

Can you help me understand why depreciation went down and how you're thinking about depreciation in Q4?

Well, first of all, our depreciation for the whole year 2014 should be less than 10% growth than that of 2013.

As for quarter-over-quarter movement, sometimes you have older equipment that the depreciation has expired.

So quarter to quarter, it's more difficult to predict.

But Q3 in 2013 the depreciation expense is probably the lowest quarter of the year.

And going forward the next few quarters, we will see mild quarter-over-quarter increase in depreciation.

And more importantly 2014 according to current forecast, we are seeing some less than 10% increase year over year.

Okay, great.

And then just to clarify on 28-nanometer, I know you have pre-production customers.

But can you just tell us if today right now you're in volume production with anyone for 28-nanometer outside of the one customer who has been in small volume for a while.

Because the number of customers is rather limited, I think we won't be able to tell too many details.

But I guess what we can say is we stick to our original target that low single digit production at the end of this quarter.

And hopefully quarterly revenue we'll see low single digit in Q1 next year.

Okay, great.

Thank you.

Andrew Lu, Barclays Research.

Thank you.

A couple of quick questions.

First one, demand outlook regarding Q4 based on your guidance, what kind of sector application are declining more than others?

It's -- almost each sector, consumer, communication and computers.

But for the sub-sector, for the consumer sector there are some DTV set-top box, [DSC] products that has declined.

And for the big surprise, communications, the baseband and ISP products also went downward.

And for the computer it's the touch and flash controller products (multiple speakers).

So can we say the consumer is worse than the communication and the communication is worse than the PC?

It's quite even and -- but of course the communication takes the most portion of our revenue share.

So from the absolute numbers, communication was downward more serious than --

In terms of percentage which one have the larger decline?

It's pretty even.

Even, thank you.

The second one is on the solar business.

You are guiding actually you're increasing the solar revenue to NTD2.2b.

But losses seems to be expanding.

Do we have any idea this solar business can turn in more breakeven level in the next few quarters, within the next few quarters or will remain at this kind of losses for quite some time?

It's difficult to predict.

And we have been pushing really hard on all those solar entities and the situation has actually improved quite significantly if you compare to 2012.

So we hope next year we will see continuous improvement.

As for when we can see profit, it's a little bit more -- it's a little bit difficult to predict right now.

Thank you.

The last one is the breakeven utilization rate.

Based on Q4 the guidance, pretty much breakeven utilization roughly about mid 70% quite similar to Q1 this year, but worse than year 2011 Q4.

In that time breakeven utilization is about 67%.

Do we expect any improvement on this breakeven utilization to be lower in future for the next downturn?

And what are we going to do to make this happen?

Andrew, I think that our 28-nanometer capacity is not very well utilized at the moment and that's the main reason for the higher breakeven point level.

And with our engineering progress on 28, it's improving significantly, we hope that the low utilization rate on 28-nanometer equipment can be improved as well.

So hopefully this is a short-term phenomenon and through our improvement in the leading edge capacity in our utilization rate our breakeven point should be able to come down from the current level.

Very helpful, thank you.

Thanks.

Roland Shu, Citigroup.

Good evening.

First, a question for the solar.

Since this is about 2% of your total revenue, however loss of about 17% of the bottom line.

So is there any plan for you to terminate this new business in the solar in the near term?

We have already stopped expanding any solar-related capacity for quite some time.

All the focus for the past 18 months has been putting on the efficiency and the capital reserves.

So first of all, we will not expand the solar capacity and secondly, as I mentioned earlier, the performance from a loss-making point of view has reduced quite significantly compared that of 2012.

So we are continuously looking for opportunity for internally or externally to improve the efficiency of the solar operation.

So, again hopefully 2014 we will see a better performance coming from our solar investment.

But 2014, I think it probably will be still loss making on the solar business.

In the near term, yes.

Okay.

Okay, second question is from the competition in China.

I think there is a new foundry in China called [Hong Li] (inaudible) and I think this is a new joint venture there.

I think their core product is doing 55-nanometer specialty IC for using their high-voltage technology and even for (inaudible) IC.

I think this actually is a direct competition with your specialty IC.

So from that actually are you seeing any pricing pressure or competition from this new [competitor] in China?

I think that in high voltage industry, we are -- our technology is in the lead position.

And of course we will feel the pressure from China's competitors, but I think UMC will still in the lead position on this area.

For your 4Q you guided ASP to decline slightly.

Is there any reason or correlation with the competition from China?

No, it's mainly because of our product mix.

It's less shipment on 12-inch because of the end customer demand.

Okay, thank you.

Thank you.

Michael Chou, Deutsche Bank.

Hi, good afternoon.

I have several questions.

The first one is regarding your currency.

Could you give us what is the expectation for Q4 and what is your currency in Q3?

Thank you.

Q3 was around 29.7 or 29.65 something like that.

And up to date it's already appreciated more than 1.6%.

And we cannot predict the future in terms of currency so in our guidance we have excluded the impact from the currency movement.

Thank you.

Okay, the second question is do you still maintain the guidance of the -- some 28-nanometer high-K metal gate revenue from Q2 next year?

We didn't recall we have said that.

We said 28-nanometer high-K metal gate will be maybe two, three quarters after poly/SiON.

And we are seeing the take-off of preproduction of 28 poly/SiON by the end of this quarter.

And so we didn't recall we gave a Q1 (multiple speakers).

Oh yes, because you mentioned before two to three quarters.

So can we expect you have some revenue from Q2 next year?

Yes, we expect that our high-K metal gate will be in the second half.

Yes, it will be second half.

Okay.

In that case would you shift your capacity from poly/SiON to high-K metal gate next year?

Or you will increase the total 28-nanometer capacity as well?

Mainly we will increase our total 28-nanometer capacity.

And of course we focus on high-K metal gate version.

Okay.

Given that the challenging high-K metal gate is more, it should be more difficult than poly/SiON, right, so if your poly/SiON revenue could be just low single digit in Q1 how can we -- how do we think about high-K metal gate revenue trend going forward?

Can you give us some color?

Thank you.

I would say it's -- actually our high-K metal gate we have two versions.

And one of the high-K metal gate version is actually doing quite well in terms of yield achievement.

So it's quite depends on the customers, the end customers' demand.

We already -- our high-K metal gate versions, one of the versions we already achieved more than 80% yield.

So this one we are -- we expect that there is some business in the second half next year.

Okay, thank you.

My final question is is it possible you give us some guidance for 28-nanometer capacity by the end of next year?

Michael, as I mentioned we don't have the number for CapEx next year as yet.

It will be available next quarter.

And even you use our base line NTD1.5b you can do a rough calculation according to how much it costs per 1,000 wafer capacity.

And also I've mentioned we have other things we need to spend money as well including specialty 40-nanometer of our P5, P6 [shell] in Tainan.

Thank you.

I have no more questions.

Thank you.

Thank you.

Steven Pelayo, HSBC.

Great, just a couple of follow ups here.

With shipments down 8% to 10% in the fourth quarter and R&D relatively stable, how much flexibility do you have in your SG&A?

Can that also come down by 8% to 10% quarter on quarter or are some of that kind of sticky if you will?

SG&A to come down 8% to 10% is rather difficult.

And we have some as always cost control scheme mostly for fabs, including parts for repairment.

And so altogether we are continuously bringing down our production costs.

But for SG&A with the increased utility costs it's very difficult.

Okay, so I won't assume that's significant.

And then I know in the third quarter you do expect or you receive a lot of these non-operating incomes from your investment portfolio.

But it makes it very unpredictable quarter to quarter.

Do you have any initial thoughts on the non-operating income contribution for the fourth quarter?

Q4 you can put in that as zero for now.

Zero?

Okay.

And last question, SMIC is talking a lot about specialty as well.

I think it's also about 30% of revenues for them.

It really seems like a strategy for them they want to drive and grow that significantly as a percentage of their mix as well.

Is that something you envision UMC doing as well?

A year from today will we be talking that 30% of revenue will be 40% or more?

Help me understand is specialty going to be a bigger growth driver in the overall top line, so it's kind of a barbell if you will 28-nanometer on the leading edge and specialty kind of on the trailing edge?

Can you help me understand the midterm thinking on specialty growth?

I would say it will be more steady, steadier.

Okay, so it maybe stays steady as a percentage?

Okay.

And if I just sneak in one more just since I am.

SMIC also seems to be finally starting to ramp up 40-nanometer as well.

I'm curious any increases or any changes in pricing pressures that's potentially driving some of your guidance?

I know you talked about it being primarily mix related.

But I'm curious on a like-for-like basis given more time has gone by and more competitors are offering 40-nanometer now are you seeing any more intense pressure there?

Yes.

We have -- we haven't felt too much pressure on the 40-nanometer business.

So, yes.

Okay, then I'll just ask one last question.

You talked about five products at 28-nanometer in preproduction.

I'm curious is that from a wide range of segments or they're primarily communications related.

Maybe you can talk in a little bit more detail on what types of products they are.

But help us understand who are -- these initial five products what type of areas they come from?

It's a wide range.

It's not only from communications.

They are from some others.

Okay, quite a bit of breadth there.

Excellent, thank you very much.

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.

We appreciate you following us and we are going to work hard on your behalf.

Thank you very much.

Thank you and thank you for joining us for third quarter earnings conference call today.

We will speak with you on the next quarter.

As always if you have any additional follow-up questions please feel free to contact UMC at ir@umc.com.

Thank you again and have a good day.

Thank you, ladies and gentlemen.

That concludes our conference for the third quarter 2013.

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 disconnect now.

Goodbye.