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

Randy Abrams - Credit Suisse Michael Chou - Deutsche Bank Donald Lu - Goldman Sachs Roland Shu - Citigroup Rick Hsu - Daiwa Securities Steven Pelayo - HSBC Charlie Chan - Morgan Stanley Gokul Hariharan - JP Morgan Sebastian Hou - CASA

Welcome everyone to UMC's 2016 Fourth Quarter Earnings Conference Call. All lines have been placed on mute to prevent background noise. After a presentation, there will be a question-and-answer session. Please follow the instructions even at that time, if you would like to ask the question. For your information, this conference call is now being broadcasted live over the internet. Webcast replay will be available within an hour, after the conference is finished. Please visit our website www.umc.com under the Investor Relations, Investors, Events section. And now, I would like to introduce Mr. Bowen Huang, Head of Investors Relations at UMC. Mr. Huang, you may begin.

Thank you and welcome to UMC’s conference call for the fourth quarter of 2016. I am joined by Mr. Po Wen Yen, the CEO of UMC; and Mr. Chitung Liu, the CFO of UMC. In a moment, we will hear our CFO present the fourth quarter financial results, followed by our CEO's key message to address UMC's focus and the first quarter 2017 guidance. Once our CFO and CEO complete their remarks, there will be a Q&A session. UMC's quarterly financial reports are available at our website at www.umc.com under the Investor's financial section. During this conference, we may make forward-looking statements based on management's current expectation 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 these risks, please refer to UMC's filing with the SEC in the U.S. and the ROC securities authorities. I would now like to introduce UMC's CFO, Mr. Chitung Liu, to discuss our fourth quarter 2016 business results.

Thank you, Bowen. I would like to go through the fourth quarter '16 investor conference presentation material which can be downloaded from our website. Starting on Page 3, the fourth quarter of 2016 consolidated revenue was NT$38.31 billion with gross margin at 22.9%. The net income attributable to the stockholder of the parent was NT$2.55 billion and the earnings per share for this quarter were NT$0.21. And the operating result summary, our capacity dilution rate in Q4 last year was 94%, about 5 percentage point improvement from the previous quarter and for cash on hand is around NT$57.5 billion. And because of the higher shipment, revenue grew by about 0.4% compared to third quarter ’16 to NT$38.3 billion. Gross margin was 22.9% about 5.5 improvement from third quarter of ’16, and net income is NT$449 million due to higher non-operating losses which we will go into detail later, and the net income attributable to stockholder of the parent was NT$2.54 billion, EPS was 0.1. For the full year, revenues show about 2.1% year-over-year gross, a NT$147.87 billion, and gross margin percentage was 20.5 or NT$30.3 billion. And operating income is about 4.2% or NT$6.2 billion, and net income attributable to stockholder of the parent was NT$8.3 billion or 38% on a year-over-year decline. EPS for the full year is 0.68. And other than our $57.5 billion are in cash, our total asset also reached 386 billion. For operating segment, both of our profit comes from wafers application in Q4 2016. The losses comes from new business has come down and become quite insignificant on consolidate basis. ASP showed a 2% decline in Q4. And for revenue breakdown, Asia grew up to about 45% in Q4 and North America declined by 4 percentage point to 48, and Europe and Japan remain unchanged. For the full year, we also see a decline in Europe and Japan business units. And Asia continued to grow to 44% and North America remained our largest market for 49% of our revenue. IDM remained unchanged for 7% for Q4 over Q3 in 2016, but our full year has showed a decline from 12% in 2015 to 8% in 2016. And in terms of segment breakdown, consumer is unchanged, and communication still remained around 53% unsold, and computer is 13%. And for the full year, it doesn't really change much compare to 2015 with communication account for 53% of our total revenue. And 28-nanometer set the record in Q4 2016 with 22% of our total revenue, and for 40-nanometer revenue in below account for 48% of our total Q4 revenue. For the full year, 28-nanometer showed quite an increase from 10% in 2015 to 17% in 2016. For 40-nanometer, the percentage has also increased from 24% to 27%. We will see some capacity decline due to annual maintenance in Q1 2017, but we continue to invest in our leading edge capacity for both 12A, 12i as well as 12X in Xiamen China. We budget around 2 billion CapEx for 2017, majority will be our leading edge capacity. So that above is a summary of UMC's results for Q4 2016, 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.

Thank you, Chitung. Hello, everyone. I'd like to update to everyone UMC’s fourth quarter operating results. In the fourth quarter of 2016, UMC’s revenue from foundry operations was NT$38.22 billion. Our overall capacity utilization reached 94%, bringing wafer shipments to 1.66 million to 8-inch equivalent wafers. Operating margin was 6.3%. During the quarter, our 28-nanometer and 40-nanometer utilization rate continued to exceed 90%, while strength in 8-inch consumer and communication demand raised 8-inch Fab utilization to nearly 100%. We also realized a noteworthy milestone in November with the grand opening of our 300-millimeter Fab 12X in Xiamen, China, which began shipping 40-nanometer customer wafers just 20 months after the fab’s March 2015 groundbreaking. This site will ideally position UMC to capitalize on the vast business opportunities within China’s semiconductor market while bringing us closer to our Chinese customers, where our team can provide superior technical and manufacturing services and more efficiently bring new tape outs into production. With regard to our advanced 14-nanometer technology, we have recently made substantial progress for this advanced node. Following intensive engineering activities with our customer, UMC’s 14-nanometer transistor performance has delivered speed and leakage results which are comparable with the industry’s 14-nanometer standards. Our yields have fulfilled customer requirements, and we anticipate 14-nanometer wafer shipments to commence in first quarter 2017, highlighting our determined efforts to reach this important milestone. Looking into the first quarter of 2017, as we enter early year seasonality, we expect a sequential decrease in our foundry revenue. For full year 2017, UMC will continue to work towards a year of growth and prosperity. Our fundamental of 28-nanometer process technology knowhow will enable our team to develop new manufacturing solutions on logic and specialty technology platforms. We will also expand 300-millimeter capacity at Fab 12X to address growing customer wafer demand, increase our foundry market share and elevate the competiveness of our foundry services. We believe these efforts will position UMC to capture the next wave of growth opportunities. Now, allow me sometime to summarize the recent highlights in Chinese. [Foreign Language] Now, I have finished my remarks and let me go over to the first quarter 2017 guidance. Our shipments will show a decrease of approximately 1%. The ASP in NT dollar will show a decrease by approximately 3%. UMC's gross profit margin will be in the mid-teens of percentage point range. The capacity utilization rate will be approximately 90%. Our foundry CapEx for 2017 will be $2.0 billion. That concludes my comments. We are now ready for questions. Operator, please open the lines up. Thanks.

Thank you, Mr. Yen. And ladies and gentlemen, we will now begin our question-and-answer session. [Operator Instructions] The first question is from Randy Abrams from Credit Suisse. Please ask your question.

I wanted to ask the first question about the CapEx 2 billion budget. If you could talk about how much capacity you plan to ramp at your China the Fab 12X? And then also within that -- for total how much you expect 28-nanometer capacity, if you can give where that 28 capacity is now, and then your plans to ramp additional 28 capacity through the year?

Yes, our CapEx for the China is about 50% of total CapEx for 2017. And our China 12X Fab capacity will increase from 3,000 wafers per month up to 11,000 wafers per month. And we have CapEx allocation on the 28-nanometer expansion at our 12A Tainan, Taiwan, and we increase from 29,000 wafers per month to 35,000 wafers per month.

And the second question, I wanted to ask about the first quarter sales and gross margins. First, if you could talk about the sales that’s holding up shipments relative to your peers. Could you talk maybe by application? How the applications are performing in 8-inch versus 12-inch? But then also on gross margin, it seems like the decline a bit more than normal to connect to mid-teens, so that factors -- if it's the ASP reset beginning of the year or another factor on the gross margin?

I'll look on the quarter '17 by application the communication and computer, we have a few percentage drop compared to Q4 last year. And consumer will increase 13 percentage points.

On the consumer, what's driving the -- is there certain applications driving consumer to grow where first quarter is normal seasonally down or fewer working days?

Yes, we see that the microcontroller units have a stronger demand on 1Q '17.

And then, if you could talk about the gross margins, what was the factor to guide down to mid-teens whether it's ASP mix or another factor?

Yes, ASP definitely is a factor. We have lower utilization rate for 28-nanometer in quarter one, which result in a lower rate blended ASP for first quarter. Also for Q4 last year, we have some insurance claim to help our overall gross margin. The insurance claim in Q4 last year was over a NT$1 billion, so that also -- and raised the gross margin in Q4 last year.

And last question on non-operating loss, if you could go through, I think you disclosed part of it was a onetime. Just the couple of non-operating charges that you're talking, if you expect are back to kind of normal breakeven as we go into first quarter?

Yes, for Q4 there were a few one-off items and also some short-term items, other than the one we've disclosed which is a solar firm invested by our investment company, that's one-time impairment loss around NT$600 million. Also we have higher interest expenses coming from our CapEx, which in the beginning is highly geared before more capital will be pulled in. They also encounter some ForEx losses because of the weakness in the RMB, and they also have no U.S. denominated revenue to offset the weakness in RMB. So, the situation should improve along with small invested capital in Xiamen as well as increasing U.S. denominated revenue.

And you expect to be -- so we could still have some losses or we get back to normal or near breakeven looking forward?

I think our goal is try to manage to a more breakeven level, but it's also highly depends upon the level of stock market which gives us some room for disposal gains.

And the next question is from Michael Chou of Deutsche Bank. Please ask your question.

Two questions. One is when would you see your Xiamen Fab to enter 28-nanometer mass production?

Currently, they're only planned for 14-nanometer production. And we will apply to Taiwan regulators for approval to initiate 28-nanometer production once our 14-nanometer start shipments in Taiwan.

Okay, so do you have a rough schedule? It could be the first half of 2018, or even later?

Of course, to our interest to have to try to expedite the whole process, but again it's really up to the regulator in Taiwan.

Okay. Second question, do you have any color for your 28-nanometer sales portion by end of this year?

You are asking for the full year average?

Full year averages or end of this year?

No, we don’t have the exact number for now.

Can we say your sales portion will still be higher than last year, or you think that the sales portion will drop this year?

We believe in the second half. The second half especially the Q4 this year will be more than 20%. That means you're coming from 28-nanometer.

A follow-up question is, can you give some color for the sales portion between poly/SiON and High-K/Metal Gate in 28-nanometer this year and last year?

In the last year, we have about equivalent share between the 28 High-K/Metal Gate and a poly/SiON. And this year your tenure quarter, you're very-very much quarter-by-quarter and we're still pushing to have a higher revenue contribution from High-K/Metal Gate.

So, can we say UMC has a higher High-K/Metal Gate sales portion versus poly/SiON this year? I mean compare to 2016 -- in 2016, you mentioned it's a 50:50. So, can we say UMC has a higher High-K/Metal Gate sales portion by end of this year or the whole year as well?

No, we don’t have exact number for now. But I'd say that for this first half, poly/SiON, 28 poly/SiON will have higher revenue than our High-K/Metal Gate, but the second half expecting High-K/Metal Gate will become the strongest revenue contributor.

Okay. So, can we say second half this year which you see High-K/Metal Gate to exceed poly/SiON sales portion?

Now, we are targeting on that.

And the next one is from Donald Lu from Goldman Sachs.

My first question is on 14-nanometer. I see the percentage of 14-nanometer revenue has not increased, but you are investing in Xiamen. So, what's the projection for the Xiamen 14-nanometer revenue for this year? That’s my first question. Second question is on ASP, in Q1 I think Chitung commented 28-nanometer revenue might decline a little bit, but why it's declining 3%? That’s quite significant. Is there more than expected pricing pressure on 8-inch or 12-inch? Thank you.

Allow me to take the first question. You're mentioning about the Xiamen, yes, we are now ramping at the 14-nanometer products, customers' wafers in that Xiamen Fab. And as Chitung just explained, we'll follow Taiwan government and management technology restriction. So, after we have 14-nanometer revenue coming from customers wafers and then we will ramp our 28-nanometer in different Xiamen Fab.

But why those -- in Q4 we see the 28, -- 14-nanometer revenue actually has been flat almost in Q3?

So, the Xiamen Fab only turned mass production officially starting from the month of December in 2016. So, they only have like nearly almost no contribution from wafer out point of view, and most of wafer outs would only start coming in Q1 this year. And also very minor, we are talking about 3,000 wafers only, so very minor.

Okay.

For the ASP, you are referring to about 3% decline. I didn’t say the 28-nanometer would decline a little bit. I said 28-nanometer would decline in Q1 as a percentage of revenue. And I think that probably the more main reason for the whole 3% blended ASP decline.

Okay, so the 3% is the cost 28-nanometer revenue percentage in the mix would decline in Q1?

Mostly, of course, there is some price adjustment as well for beginning of the year as always.

By this although compete pricing pressure become a little bit more than last year or is that similar?

Similar like Q4 ASP, if you were aware we already showed a few percentage decline in Q4 alone last year, and that’s mainly attributed to the normal turnout pricing competition. And for Q1, this year is mainly due to the just next of 28-nanomter revenue decline.

And the next one is from Roland Shu from Citigroup. Go ahead please.

First question is for your first quarter Xiamen guidance. You said on your first quarter utilization is about 90% which is lower than 94% in 4Q last year. And the capacity in first quarter this year will be also lower due to annual maintenance. So by these two factors, how can you give us some more color, how will be the first quarter Xiamen guidance only declined by 1%?

So, Roland, it's always in wafer pipeline with all different definitions. So, when we calculate loading, it's based upon wafers in the pipeline and -- so some of the wafer produced last quarter will be shipped this quarter. I think that's a main reason.

So, at least more 12-inch wafer or 8-inch wafer?

There's no specific trend. This is just more like across the Board random like business as usual type of wafer shipped with all type of flat.

Second question is that looking at your 8-inch wafer utilization increased from 90% in 3Q to almost fully loaded in 4Q, but looking at your 8-inch wafer revenue only increased by 6%. So, was it because of ASP declined or product mix change? And how about all the 8-inch wafer revenue outlook in this year? Thank you.

So, as I mention Q4, we saw some pricing decline mainly due to pricing competition in maturing of larger in the 8-inch space. So, yes, you're correct, pricing was quite competitive in Q4 last year for 8-inch.

So, how about all the 8-inch pricing competition in this year?

Q1 overall blend has been 3%, but the pricing is pretty much done, has reached a level of plateau, but we don't have full year outlook yet.

But how about overall 8-inch wafer revenue for this year, since last year actually your all the 8-inch wafer revenues have been declined. So, how do you think about this year?

I'd say that declined a little bit because of price erosion year-over-year, and however, we're -- the full loading on the 8-inch fab will continue to at least for the first half of this year. So we -- that would allow UMC to improve our product mix depending on ASP, average ASP.

Last question is for your insurance claim in last quarter or it's all in cost of goods sold line or is there any claim in other line? Thank you.

It all cost of goods sold.

So, more than 1 billion was all in the cost of goods sold?

Yes.

And the next one is Rick Hsu from Daiwa Securities. Go ahead please.

So my first question is, can Chitung elaborate your non-op, the net investment loss again. I think the total is up 1.1 billion and you mentioned above some loss on your solar firm. Can you elaborate more that in details?

Yes, there are three or four major components. The one-off part is the impairment loss coming from our solar firm investment company. I that’s around NT$600 million in Q4, and we also have higher interest expenses because of our Xiamen Fab is highly geared in the early stage of operation. At the same time, they also encounter following currency losses due to weaker RMB and because they don’t have U.S. denominated revenue to offset that yet, and they are currently suffering from higher ForEx losses. And the situation for both interest expenses as well as ForEx losses should gradually improve in 2017 because of the U.S dollar revenue as well as of higher injected capital.

Then can you give us your depreciation guidance for this year linearity quarter basis?

For last year in 2016, depreciation expense increased by a little bit more than 15% year-over-year with Q4 is the highest quarter. And for 2017, we expect to see another 5% to 10% year-over-year increase with each quarter is quite average, so quite similar. So, we see maybe 1% or 2% increase for Q1 over Q4 last year.

Okay, so the 1% to 2% sequential increase in Q1 this year will be depreciation. Alright, the last question from me is, how do you see your customers inventory levels? Do you see any inventory issue in the first half or you think it's okay and pretty healthy?

Our inventory in 1Q is pretty much flattish. We don’t see -- pretty much flat.

Okay, when you say flattish inventory that’s your own inventory or your customers' inventory?

Our own inventory.

What about your customers' inventory? Do you have any idea about your customer inventory?

Also customers, they've been some uncertainty and they do not provide the clear numbers for us.

And the next question is from Steven Pelayo from HSBC. Go ahead please.

Why don’t we start on 14-nanometer, you talked about some wafer shipments commencing in the first question. Maybe you could talk a little bit about the number design, capacity plan and maybe revenue targets let’s say about the end of the year for 14-nanometer?

Yes, we are beginning shipping 14-nanometer revenues in the first half especially in this quarter, and there are more in the second quarter coming. And our CapEx for 14 will be just around 2,000 wafers per months. And so that’s our status.

Okay, do you have a target maybe by doing end of the year for that capacity and maybe the revenue that potentially you could get by at the end of the year on 14-nanometer?

We don’t have -- we will give more color in the coming next in the April's conference call. So far we really have increasing 24 customers here test chip for now and that’s our focus to engage more 14 potential adapters.

Okay. And I wondered if you could just talk a little bit I guess your fairly free cash flow negative in 2016? Do you think we can be above breakeven on free cash flow in 2017?

It depends on how you look at the breakdown. For UMC Taiwan, yes, we’re going to see positive net free cash flow for the first time in recent years, but our JV in Xiamen still will encounter heavy investment, which the negative cash flow expected.

Okay and just one final question. I know you've talked about 28-nanometer declining as a percentage of revenue in the first quarter. Can you'll be a little bit more specific there where you're talking 5% decline quarter-on-quarter in dollar? Or is something more significant, can you be a little bit more clear there?

Yes, our first quarter, the revenue contribution for 28 will be at around mid-teens percentage point.

Mid-teen?

Yes.

Okay, fairly significant decline. Is there any more details behind that, I think in the fourth quarter it was around 22% of revenues?

Yes, there are some market dynamics. The first way customers, they're migrating into advanced nodes earlier than our and expected. And the second way of customers, we're heavily engaged, are coming later in our expectation. So that comes off our first quarter, 28-nanometer is under loading.

Okay, well, I am still impressed though with your offsetting it elsewhere as 28-nanometer falls from 22% of revenues to mid-teens. What node is growing and for you to offset that in the fourth quarter while in the first quarter -- pardon me -- in the first quarter?

The first quarter has been -- again what's your question?

You've just suggested that in the first quarter of 2017, 28-nanometer revenue would fall to the mid-teens, and that is from the 22% of revenue level in the fourth quarter. So, given your guidance for shipments, it would suggest that something else is some other node is definitely up sequentially. So what node is strongest for you I guess in the first quarter?

Actually, our 14-nanometer is still very strong. Our 40 is very full and our 8-inch is also full. So, that supports our, the overall utilization is around 90%.

And the next one is from Charlie Chan from Morgan Stanley. Go ahead please.

So, could you please elaborate what kind of a consumer product is strong in the first quarter? I mean are those TV set-top box or anything related or we can assume it to have IC?

Yes, it's a microcontroller unit. It's the strongest part at least in the semiconductor I mean the consumer segment.

So, are those MCUs for what kind of consumer, is it home appliances or toys or what kind of consumers?

Mainly for home appliance.

Home appliances. Okay. Thanks. May I get a sense about your OpEx trend in terms of absolute dollar or percentage of revenue in 2017?

It's our goal to maintain a stable percentage of revenue in the OpEx, but there will be also positive patterns as well. Normally, in Q1, it's always the lowest, and Q4 it's always the highest.

I thought there was some one-time OpEx last year when you ramped up the Xiamen Fab. So is that kind recurrent OpEx or that part of OpEx will experience in 2017?

The early start-up costs associated with Xiamen Fab is counted under OpEx until the end of November last year. And starting from the mass production in December, all the costs item will go to the appropriate lines.

Meaning cost of goods sold will increase a little bit OpEx or decline slightly.

That's right.

And lastly, I want to make sure for your CapEx for Xiamen Fab, will there be any contribution from Xiamen government? How do you share the CapEx?

We don’t share CapEx. We share capital injection, the paid-in capital according to the investment agreement. So, you're going to see after this recent, this year 2017, we're budgeted to inject $415 million after that UMC will control over 50% of the JV. Currently, we only control about 32%.

Lastly, I guess all investors would be curious, so you didn't give explicit full-year revenue guidance but just in the sense of outgrowing or under-growing industry, let's say foundry industry growth is high single digit, what do you think your growth rate will be?

Yes, our numbers for the semiconductor industry is on the year-over-year gross rate will be around 5%, and for the foundry segment, excluding memory, our numbers is 7% to 8%. And UMC will have a mild growth in 2017.

So means low single digit growth is, am I right?

We will be in line with the semiconductor industry’s growth.

Okay.

2017.

So lastly, it seems like your utilization rate is like a full percentage lower than fourth quarter and do you expect or does that mean second quarter revenue will see a sequential decline? Because as Roland just asked, your 1Q shipment declined only 1% so I would assume the wafer outs for second quarter could be even lower.

We don’t guide on that so far and we’ll give more color in the coming analyst conference call.

Okay, so under pricing pressure part do you think those pricing pressures or competition in edge comes from Taiwanese industry peers or is that coming from Taiwanese foundry?

It’s a cross support, but we are more concerned on loading. So, once we have, as I just explained once we improve the utilization rate, we will have a better position to optimizing our product mix to improve our ASP.

And the next one is from Gokul Hariharan from JP Morgan. Go ahead please.

My first question, could you comment a little bit more on the 28-nanometer supply/demand situation? You talked about some of the first movers, first stream of customers moving onto advanced nodes, there being a time lag in terms of bringing on second wave customers. At the same time, it seems like there is a fair bit of capacity expansion in Taiwan as well as in China on 28-nanometer. So could you talk a little bit about where you think the 28-nanometer supply/demand balance is achieved, is it going to take until second half for you to ramp some of the new customers, or do we see this as a one quarter thing and we see a pick-up in Q2?

Yes, our 28-namometer are process solution, we will believe is positioned very well especially on our High-K/Metal Gate. UMC is our one of the only two gate-last High-K/Metal Gate versions providers. And because especially our High-K/Metal Gate, we improve our process of performance, we are now outperforming in the rest of the industries and factories especially our High-K/Metal Gate is -- we have much smaller leakage, which is also representing the lower power consumption, or we can enjoy our higher speed at the same energy level. But so that makes UMC can more attractions from our customers on UMC's High-K/Metal Gate process. So though there are some bumpy because of the macro dynamics than demand on adjustment, but we believe the 28 will be the very strong and last long. And UMC is in a very good position to gain more demand on that.

Okay, so this short-term situation in terms of bringing on your second wave customers, do you expect 28 to start growing in Q2 again or is it still going to be relatively sluggish in the second quarter as well?

As I explained, we don’t guide the second quarter but we do believe that based on our customer engagement and demand, we believe that they're very strong bounce back starting from the mid of this year, the mid of this year in only 28-nanometers demand.

My second question is on the gross margin side, is there some impact -- a meaningful impact from Xiamen ramp up in the gross margins in Q1? How should we think about gross margins as we go through the year? Are we getting back up to the 20%-plus territory in the second half of the year or is it going to take a little bit longer as you have a lot of capacity coming online in China?

Chinese fab will have impact about 4 percentage points for overall gross margin; however, of course we will try very hard. The major factors are pretty much up to our loading ASP trends. So, that’s still a must be the factor.

So, the 4% impact, Chitung, is in Q1 or is it something that will stay through the year as you bring up the Xiamen site?

That's our forecast figures throughout the year. But we're also expecting the 50% ownership of minority interest as well.

And the next one is from Sebastian Hou from CASA. Go ahead please.

So my first one is I need some clarification from you guys on the first quarter gross profit margin guidance. So if we take a look back on the fourth quarter, so if we take out the NT$1 billion insurance claim which means your fourth quarter gross profit margin could have been about 20%, which is nice and better I think than earlier guidance without insurance claim. But if we compare that with your first quarter margin guidance, which is about mid-teens, which means that your margin is going to decline by 5 percentage points, and if we consider this along with your lower revenue contribution from 28-nanometers, which presumably should be carrying lower than corporate average margin. So from the product mix perspective it seems like your margins should trend to a better territory. So I'm just a little bit confused here. Can you explain more why we see such a large margin decline in first quarter?

Your assumption is not 100% correct I mean 28-nanometer was below our corporate average in terms of gross margin; however, because of the underutilization the cost of 28-nanometer manufacturing is also becoming higher. And the contribution or earnings strength if you will is actually bigger than Q4. So, overall there is still dragged down our gross margin because of the lower capacity utilization rate in 28-nanometer capacity.

Okay, yes, I think there's a lot of factors impacting your margins, so I think the problem is with probably just one sector. Well, thanks for explaining that. The follow on that is that the -- what is your OpEx ratio in first quarter?

It should be still similar to that of Q4 last year but a little bit improvement.

So your Q4 last year is about 14% OpEx ratio, so a little bit improvement, probably a similar level, but your gross margin guidance of mid-teens which means that your first quarter operating level pretty much just breakeven. I'm not sure if that assumption?

Yes, looks like case for Q1, yes, but we will try to stay in back.

My second part of the question is 28-nanometer, so I think the CEO mentioned about the capacity for 28-nanometer this year, additional capacity will be 6,000 wafer per month. I remember in the past two calls you mentioned about you were going to -- you were planning to add about 10K to 15K by -- in 2017, so your total 28 capacity will reach 40,000 to 45,000 by -- well, in 2017. So it seems like your capacity expansion on 28 is more conservative than earlier, I just wonder what has changed.

Actually, it's not changed. I mean we're having additional 5K in Taiwan and plan five additional 5K in Xiamen pending upon subject to Taiwan government approval. So, once we get the approval of that, the 5,000 capacity will be putting to the picture.

Okay, so which means your CapEx for this year will probably be raised in the second half or later depending on the approval from government.

Partially -- part of the equipment, we've already been planned, but not executed yet. So, it's a small portion compared to our overall CapEx. But still there is a range for the CapEx like last year our budget was 2.2 billion, but we end up spending about 2.8 billion because of the higher or faster schedule in terms of Xiamen expansion.

Okay, so you're still on track to build like 10 probably minimum 10K of the new capacity on 28, just pending approval?

Okay.

My two-part question is on the minority, on the P&L, so it seems your minority has increased quite a bit in fourth quarter and along the way in the past few quarters. I believe that -- assume that's mostly from your 12X Fab in Xiamen. So I just wonder can you give us some guidance or hints on how should we model this in first quarter 2017 and also for 2017?

Once, we inject our capital for another 450 million, UMC holding will increase from the 30% to 50%. The minority interest will also come down from 70% to 50% after we complete our capital induction in Q1 this year.

But your CapEx will -- you're also adding more capacity, so the number still probably will go higher?

You mean?

The minority.

Minority interest, it depends on the result of our Xiamen Fab. Right now, it's focused to be a loss but we don’t have detail numbers for the full year.

Okay, and I have a follow up on your Xiamen -- sorry, the 12X Fab. So CEO mentioned about 11k of the capacity planned by the end of this year, so the first part of the question on that is, is that all for 40-nanometers?

The first 6,000 wafers is obviously 40 nanometers, the future expansion which is ongoing were at the pending on Taiwan government’s approval.

Okay. So that extra 5,000 wafer is 28-nanometer which we discussed earlier. It is good to here.

That's on the government approval, yes.

The last question is on the -- I noticed your IDM revenue, the customer -- so your revenue from IDM customers have declined from a total contribution from 12% to 8% and putting it in dollar terms is declining by over 30% year over year last year. So I wonder, is there any change in terms of commitment or what kind of a change you see from -- I mean what happened?

I will say the part of the reason is because that the industry's consolidation. Some IDM companies will acquire by the fabless design companies, so that we calculate into that fabless design companies.

Okay. So there is not much change in terms of the customer commitments, but mostly due to because of some customer they acquired?

Yes, it is a very significant of industry’s consolidation in 2015.

Ladies and gentlemen, we are running out of time. So, there would be the end of today's Q&A session. And I’ll turn things over to UMC, Head of IR 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.

Thank you ladies and gentlemen. That concludes our conference for 4Q 16. And 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 Investors, Events section. You may now disconnect. Good bye.