聯華電子 (UMC) 2016 Q1 法說會逐字稿

Welcome, everyone, to UMC's 2016 first-quarter earnings conference call. (Operator Instructions). For your information, this conference call is now being broadcast live over the Internet, and webcast replay will be available within an hour after the conference is finished, and 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 Investor Relations at UMC. And, Mr. Huang, you may begin.

Thank you, and welcome to UMC's conference call for the first 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 first quarter's financial results, followed by our CEO's key message to address UMC's forecast and second-quarter guidance. Once our CEO and CFO completed their remarks, there will be a Q&A session.

UMC's quarterly financial reports are available at our website, www.umc.com, under the investors financial sections.

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 these risks, please refer to UMC's filing with the SEC in the US and ROC's securities authorities.

I would like to now introduce UMC's CFO, Mr. Chitung Liu, to discuss UMC's first-quarter 2016 business results.

Okay, thank you, Bowen. I would like to go through the first-quarter 2016 investor conference's presentation material, which can be downloaded from our website. Starting on page 3, the first quarter of 2016 consolidated revenue was TWD34.4b with gross margin around 14.6%. The net income attributable to the stockholders of the parent was TWD0.21b, and the earnings per ordinary shares were TWD0.02.

So, loading in first quarter was 82%, slightly declined from 83% in the previous quarter, but down nearly 11 percentage points from the same quarter of 2015.

So on page 4, revenue grew by 1.6% quarter over quarter, but because of the disruption in operation, mainly due to the earthquake on February 6, our gross profit dropped to TWD5b, or 14.6%, in Q1 2016. And the net income attributable to stockholders of the parent was around TWD210m, or 0.6%, and EPS is TWD0.02, or $0.003 per ADS.

For a year-over-year comparison, due to the lower loading as well as the lower ASP, our revenue dropped 8.6% year over year, and also the earnings in gross profit dropped 44% to TWD5b, and for the EPS comparison, in the first quarter of 2015, it was TWD0.32, and in first quarter of 2015, it's TWD0.02.

For our balance sheet in the first quarter, our cash has increased to TWD59.5b, and our total assets has also increased to TWD344b.

For the operating segment breakdown, for wafer fabrication or our foundry business, our gross margin in the first quarter was 15%, and our operating margin for the first quarter for foundry segment was 1.5%.

And new business, although the revenue has come down to insignificant numbers, still caused some TWD200m-plus loss for our overall consolidated performance.

ASP in the first quarter declined by roughly 2%.

In terms of revenue breakdown, Japan, together with Europe, has showed the weakest performance in the first quarter, and at the same time, Asian has -- revenue has bounced to 45% from 37% in the previous quarter.

IDM, for the similar reason, has come down to about 9%, compared to 15% in the previous quarter.

We see the communication still remained the largest share of our revenue contribution. However, the percentage has come down to 48%, from 52% in the previous quarter, and computer has made up the large share, from 11% to 15%.

Because of the earthquake disruption, our high-end specialty 20-nanometer production was impacted to a large degree in the first quarter, and as a result, the revenue contribution has come down -- come to lower than what we expected number of around 8% in first quarter, which is also a decline from the 11% in the fourth quarter of last year.

Our 40 nanometer, because of the production site, is rather more diversified, as well as very high loading, has reached a very high number of 29% of our total revenue.

Capacity in the second quarter is expected to grow about 2%, mainly coming from our 12-inch wafer fab in Tainan, 12A.

And for the full year CapEx budget, so far, it still remains unchanged, around TWD2.2b, with majority for our 12-inch capacity expansion. And now, I would like to turn the call to our CEO, Mr. Yen.

Thank you, Chitung. Hello, everyone. I would like to update to everyone UMC's first quarter operating results. In the first quarter of 2016, our foundry revenue grew 2.1% sequentially to TWD34.31b. Foundry operating margin was 0.5%. Overall capacity utilization was 82%, bringing wafer shipments to 1.43m 8-inch equivalent wafers.

Towards the end of 2015, we saw signs that the inventory cycle was hitting bottom. Since then, the inventory correction has completed its course, with normal seasonal patterns resuming for first half 2016.

On February 6, just as Chinese New Year was beginning, southern Taiwan experienced a severe earthquake that impacted UMC's 300-millimeter Fab 12A, in Tainan Science Park. However, due to UMC's stringent safety protocols, efficient execution of our earthquake recovery SOP and dedicated personnel, we returned to normal production in just a few working days to sustain 1Q 2016 revenues within our original guidance.

In compliance with our ISO 22301 certification, UMC's business continuity practices ensure accountability on prosperity damage -- property damage, as well as the safety and health of our employees. We will continue to ensure a state of readiness by implementing disaster protocols that introduce business continuity elements within our foundry operations.

UMC has recently announced an agreement with ARM to develop multiple physical IP platforms to accelerate customers' implementation of ARM Artisan IP on our 55-nanometer ultra-low-power and 40-nanometer and 28-nanometer technologies for mobile embedded and IoT chip designs. The partnership will further strengthen UMC's IP portfolio to deliver optimized design support solutions across a broad range of applications.

Looking into 2Q 2016, new product launches in wireless devices will serve as a positive catalyst, leading to stronger end-market demand. As a result, we foresee a significant increase in 28-nanometer shipments for the second quarter. We also expect to receive many new 28-nanometer tapeouts during 2016 that include mobile and consumer applications, which will further diversify our 28-nanometer product pipeline and contribute to 28-nanometer revenue growth.

While we continue to work towards business growth and market share expansion, the Board of Directors recently proposed to distribute a TWD0.55 per share cash dividend to achieve a balance between shareholders' interests and Company expansion. We continue to believe that our efforts in advanced and specialty processes will serve as a long-term catalyst to bring new prosperity and profitability for UMC employees, customers and shareholders.

Now, please allow me some time to summarize the recent highlights in Chinese. (Spoken in Chinese).

I finished my remarks, and now let me go over the second-quarter 2016 guidance. Our wafer shipment in the second quarter to show an increase of approximately 5%. The ASP in NT dollars to increase by 1% to 2% UMC gross profit margin will be in the low 20-percentage range. Capacity utilization rate will be in the high 80-percentage range. Foundry CapEx for 2016 remained as $2.2b.

That concludes my comments. We are now ready for questions.

Operator, please open the lines up. Thanks.

(Operator Instructions). Randy Abrams, Credit Suisse.

Okay, yes, thank you. I wanted to ask for this quarter -- or first quarter, since there were some pretty big mix changes, if you could talk a bit more about that.

And in that context, geography, Europe and Japan and IDMs have a sharp decline, and end markets look like a shift, PC ramping and communications coming back. So if you could give a little more on what happened on the product mix, and then for second quarter, if you expect normalization, what areas you expect to rebound in second quarter.

I'll start. As our CFO, Chitung, just explained, the first quarter, by application, the communication segment is the weakest one, and so that's why we declined in the revenue share from 52% to 49% this quarter -- the first quarter. And computer is a little bit better, and consumer is flat in the first quarter.

And for the second quarter, we expect our competition will bounce back, to become a strongest one.

Okay, can you elaborate that Europe and Japan and IDMs, it looked like a pretty sharp falloff, if that's a one-quarter correction or product change or anything that may be more meaningful there?

Yes, the -- yes. The first quarter, the demand from communication, especially from Japan and Europe, is in the bottom and is quite weak. However, we do see it getting better for the coming quarters.

Okay, and then for the second question, wanted to ask on the 28, if you could go back to you original target was a good rebound to 15% to 20% of sales and getting pretty full with the 20,000 capacity, if you could give an update. And then following on that, with these new tapeouts for second half, where you'll expand from 29,000 to 30,000, if you'll have that capacity available and you expect also to grow further in second half into that capacity.

Yes, looking beyond the February 6 earthquake impact, our 28-nanometer ramp is still on track, so our 28-nanometer production is approaching to world-class [dividending]. The yields of the these products for both 28 poly/SiON and 28 High-k/metal gate versions are at 90 percentage range. Furthermore, we believe we are delivering the highest performance and lowest power consumption of 28-nanometer technology in the world. So we expect [ready] growth customers (technical difficulty) throughout this year.

Okay.

Our capacity expansion will support our 28-nanometer revenue contribution to be getting higher quarter by quarter over quarter. So we expect to -- that this quarter, second quarter, will reach 15 to 18 percentage range, and 3Q will be close to 20 or even better to 20 percentage revenue share.

Okay, and final question, I wanted to ask on two expansions. One, if you could give an update on the China pilot line, when that capacity would start to ramp. And there were some press stories on potential venture in memory, moving it to NAND flash. If you could provide color, if that's a strategic interest or area you're pursuing.

For our Xiamen fab, it's ongoing, and it's actually pretty well, and we do expect the equipment moving to happen right now, and the wafer pilot line, as you mentioned, should be in the third quarter, with volume production in the fourth quarter. Volume-wise should be around our original projection, which is around 6,000 at the end of 2015. So Xiamen fab is on track.

And as for the newspaper report, first of all, we have already declined at UMC, relocating to any memory fab production in China, so we don't have any plans for now, at least, to -- of getting to memory fab (technical difficulty).

However, having said that, we've been approached by various parties in China to [build] semiconductor-related projects, including memory, as well. So whenever there is a concrete plan, we will certainly disclose and make it (technical difficulty).

Okay, and to clarify that, you would consider logic, but not likely to look at the memory as an option.

Well, first of all, Xiamen is a pure foundry, so it's totally separate. Other than Xiamen project, we've been approached by third parties, which include memory suppliers, as well. For the time being, UMC has no plans to get into memory production (technical difficulty). And if there's any other plan, we will disclose accordingly when it's (technical difficulty).

Okay. Thank you, Chitung.

Bill Liu, UBS.

Yes, hi. It's good to see that 28 is finally ramping, so congratulations but a couple of questions on that. If demand on 28 remains healthy into the second half of the year, is there a scenario where you expect capacity beyond (technical difficulty) or that's basically the maximum?

Yes, we expect to complete the 30,000 expansion plan by the end of this year, and of course, we -- based on our customer engagement, it is possible for UMC to build additional 10,000 to 20,000 wafer per month capacity in the coming years.

In the coming years, but probably not this year. Is that what you mean?

It's normally the capacity deployment will take one year.

I see. Okay. Secondly, can you comment on the profitability of 28 nanometers? What is the breakeven utilization and where can profitability get to, say, by the end of the year?

Yes, the -- we expect our 28-nanometer profitability will gradually increase quarter over quarter, and we will try [demand loading] by expanding our 28-nanometer capacity plus those yield-enhancement activities, so that will continue to help UMC improve our 28-nanometer profitability. So in the 4Q 2016, we expect our 28-nanometer profitability will be much closer to 12-inch corporate average.

Okay, but your 12-inch corporate average is a little bit lower than your 8-inch corporate average, right?

Yes, a few percentage points lower.

Got it, got it. That brings me to my next question, which is if you look at second-quarter guidance, if 28 grows from, say, 8% of sales to 18%, and your overall shipment is up 5%, that implies that the rest of the businesses are down, right? So is it fair to assume that maybe 8-inch is staying in a little bit weaker for right now, and what's your outlook for the rest of the year?

Yes, our 8-inch -- yes, in the second quarter is relatively weaker than the previous quarters.

I'm just wondering if you can give me some insights as to what your 8-inch customers might be telling you, because I guess form my vantage point, driver IC, some of it's moving to 12-inch, and I think the overall demand doesn't look that great this year anyways. Is there a -- do you think 8-inch can pick up in the second half of the year, and if so, can you tell me why?

Yes. I would say that the demand in the first quarter and the second quarter is a little bit weaker for the whole industry, including 8-inch business. But, however, UMC's 8-inch foundry business, it has been quite steady. And so the -- our 2Q, second quarter 8-inch loading will be in the mid 80% range. And we believe through our -- all those efforts to improve the loading rate, within the second half of this year the 8-inch loading will be improved.

I guess if I could ask it a different way, if 8-inch loading is to go up in the second half of the year, what kind of applications do you think will drive that?

There are several, including the driver and some, yes, specialty technology in the smart car and --

Okay, great. Thank you very much.

Thank you.

Szeho Ng, BNP.

Hi, good evening, gentlemen. Could you quantify the earthquake impact in Q1, at a cost of goods sold level?

Yes. Yes, we -- as you all might have expected, UMC worked very closely with our customers to minimize the impact of the earthquake, February 6, in Tainan. So first of all, we had no reported (inaudible) issue or personnel injuries.

So, yes. So of course, our Tainan fab suffered a damage from earthquake, that including our process wafers, (inaudible) and facilities. However we received an interim settlement related to the earthquake in the first quarter. And we expect to receive the remaining settlements in 2016, this year. So if we include the remaining settlements into the first-quarter 2016 results, the pro forma gross margin still in line with our original guidance.

All right.

And some customers requested to receive early April shipment in late March. So, as a result, our first-quarter revenue is in line with our original ASP and shipment guidance.

So Szeho, just to add on that, so revenue-wise, probably because of the early shipment from April to late March, very minimal impact on the revenue side. From gross margin, as our CEO has mentioned, we received roughly 2% of gross margin for the early interim settlement from insurance company. And we expect to receive another maybe roughly 3% throughout the rest of the year. So if you add back the 3 percentage point to our gross margin, in the first quarter it would go back to around 18% in foundry gross margin.

Okay, got you. All right. And, yes. And then what's the FX you are using for the Q2 guidance and also for the Q1 historical number.

Q1 it was average around NTD33.3. For the time being we're using NTD32.7 for quarter two.

Okay, all right. Okay, thank you very much.

Steven Pelayo, HSBC.

Yes, you were talking about your cash balance going up quarter on quarter, but that was actually not a function of -- I guess operating cash flow was down and CapEx was up. So could you talk a little bit about the increase in the cash, the sources for that? Did you [commit] a deposit from the joint venture, or was this settlement, insurance payments, just help me understand that.

And then maybe if you could talk just a little bit about free cash flow growth for 2016.

Yes. The cash increase from cash flow statement is mainly coming from the capital injection from our partner in our Xiamen fab which, as a result, increased our liabilities, because we do have buyback options on those investments. So the liability increased and as well as the share -- cash positions also increased. So it's a virtual cash coming from our investors for the Xiamen JV.

Okay. Do you expect any more injections throughout 2016? And what do you think about free cash flow goals for this year in 2016?

I think this part is more accounting issues. And we do expect this part will continue in second quarter. But UMC will start to inject our own capital into Xiamen fab next year. So we probably will see that in the second quarter only.

And for the cash flow forecast, this year, as I mentioned, we have about $2.2b forecast as our CapEx for the whole Group, including Xiamen JV. And for our depreciation alone we expect to see about 20% increase or around NTD50b plus. So for the time being we are about to have equilibrium. If you add our free cash from depreciation plus earnings, versus our dividend and UMC part of the CapEx, it's about equal. And that's the current plan.

Okay. That's all. I wanted to verify that, did you think we could net those out and be flat on the cash flow. Thank you.

Thanks.

Sebastian Hou, CLSA.

Thank you for taking my questions. The first one is regarding 28 nanometer. Earlier in CEO prepared remark you mentioned about you expect to receive more tapeout throughout this year. So I just wonder when your revenue -- you start to -- more products, more products and customers are in mass production in 28 nanometers. And utilization rates go up. And you mentioned about your yield rate right now for both high-K and poly SiON as 90% yield rates. So I just wonder how do you see the stability of the yield rate of the 28? So maybe for one or two products it could be at 90%, but I just wonder, if there is a 15, or 20, or more than 20 products in production, can your yield rate still sustain that 90%, that high yield level? Thank you.

Thank you for your question. Of course the -- what I mentioned just now is for our lead products. And for the new tapeout, it will take a while for the yield learning starting from say 15% gap to the 90%. And it will take a couple of months to get there, our target DD. So it's -- so not -- I'm not saying that all our products of 28 nanometer is at 90%.

Okay. So -- okay, thank you. So, well, then to follow on that is that because, right now, maybe you have fewer products in production so far, and maybe there have some products that have a higher yield. But when you have more tapeouts, and maybe more new products which just have a lower -- would that lower the -- the question I want to ask is would that lower your blended yield rate on the 28 nanometer, so that would have a negative impact on your overall margin.

I don't see that because the -- for those lead products the yield learning experience will be incremental to -- for the upcoming products. So it will shorten the yield learning rate for the future tapeouts. So we see the DD it's getting improved (multiple speakers).

Okay. Thank you. My second question is on the -- because earlier I -- I'm not sure, I just want to double check the numbers. You mentioned about the 8-inch wafer utilization rate in second quarter will be mid 80%. And okay, so how about the 12-inch in second quarter?

12-inch the average is at 90 percentage point.

90%.

Yes.

Okay. And the last question is on the industry outlook. I recall that in the January conference call, CEO gave the guidance or the outlook for the overall industry. So can you -- do you have any update numbers on the semi or foundry and UMC growth in this year?

Yes. We don't have that for this year. We still keep our guidance on the first quarter I guess, yes.

So semi is still 2%, foundry is 4%?

Yes. Our number is the semi will be 1% to 2% growth and foundry will be 4% to 5% growth.

Okay. Thank you, that's all from me.

Roland Shu, Citigroup.

Hi, good evening. Thanks for taking my question. First question is I just try to clarify the exchange rate you used in first quarter. So, Chitung, you said it was NTD33.13 or --

NTD33.3.

NTD33.3, but I just see from your presentation material on page 4, page 5, actually you used NTD32.22.

Yes that's the year-end -- quarter-end conversion rate. So the NTD33.3 is the weighted average number we calculate the revenue, profit, etc. So quarter end, in terms of converting NT dollars to US dollars, we're using the quarter-end exchange rate.

Okay. So the NTD32.7 exchange rate we use for second quarter, that is also for the average, right?

Weighted average, yes, --

Okay, yes.

-- forecast for now.

And just a look at your guidance for second quarter for the blended ASP, you said it will increase by 1% to 2% NT-dollar based. So for this one, (inaudible) for the exchange rate, that will probably appreciate about 3% to 4% and with this 1% to 2% ASP increase. So the total I think the US-dollar based is probably, the ASP is increased about 5% to 6%.

So, however, if you look at your 28 nanometer, I think revenue is going to up 8%, to more than 15%, probably 18%. So for that one (spoken in Chinese) was this -- is this blended ASP increase mainly because of the product mix change on the 28 nanometer, or you have other like-to-like price erosion which help to decrease a little bit for the blended ASP increase from 28 nanometer, right.

It's mostly coming from 28 nanometer product mix increase. But if you divide NTD32.7 by NTD33.3, it's actually 2%. So it's not 3% or 4%. So it's 2%; add 1% to 2%, should be 3% to 4% US dollars.

Yes, but I think that for 28 nanometer actually it increased much more than this, yes. So I think is there any like-to-like price erosion to average down the overall blended ASP increase? How about the pricing pressure for other node?

Not so much. As you can tell our 28 is nearly over 90% loadings in quarter, two so not much pricing pressures for the second quarter. Our 8-inch is in the catching-up mode. And we also expect to see pretty high loading in the second half of this year. So, for the time being, we don't see much pricing pressures for any major node.

Okay, thank you. And second question is on your -- according to your 2015 20-F, customers in Taiwan and China revenue contribution decreased by a total about 5 percentage point. So was it due to foundry competition in China, or there was other reasons for the change? Yes, how do you think about 2016 on the customer contribution from Taiwan and China, will it continue decreasing or will it be increased this year? Thank you.

China, for the time being, only accounts for about 8% of our total revenue.

Yes, but for Taiwan -- yes, it's at 31%, 32% --

The decline will be mainly coming from Taiwan. China is still growing.

Okay. So this is the case in this year, right? So you said for Taiwan customer will be continue declining and China will be increased?

We are sure that China will grow but we are not so sure that Taiwan will decline. And for longer time span, it's also become integrating for Taiwan and China. So it will be difficult to differentiate this, but we will certainly sure our revenue for China will continue to grow, especially after the Xiamen fab starts production.

Okay. So how about the competition? Do you see any competition or strong competition from China or from Taiwan?

Yes. It depends on the technology by technology, application by application. So, yes, in the ballpark competition is everywhere. It's all the time and --

Okay, understood. Yes. Also you said on your 20-F, your top 10 customers last year total account for 57.3% of the total revenue. So is there any customer or any customers with more than 10% revenue contribution last year?

Yes. There are one to two customers more than 10% revenue contribution of UMC total.

And how about this year, are you going to add more customers on your -- to have more than 10% revenue contribution to this year?

It's still about the same, still about the same.

One to two customer, from the same customer or it will be changed?

So far we see the same customers, yes.

Okay, understood. Okay, thank you. That's all my questions. Thank you.

Thank you.

Rick Hsu, Daiwa Securities.

Yes, hi. Good afternoon. Okay, my first question, I think I probably missed the number here, Chitung, could you give me again your first quarter foundry gross margin.

Gross margin foundry is 15%. And OP margin, operating margin was 0.5%. So 15% versus 0.5%.

Okay 15%, so that's for gross margin and OPM is about 0.5%, right?

Yes.

Okay. Thank you so much. And my second question is regarding your 8-inch business. Do you see any risk of market-share loss to your counterparts in China, especially for display driver? And would that be a structural issue, or is it just the seasonal?

Yes. Our 8-inch business has a very wide, diversified base. So I would say this is short-term, the market dynamics. My feeling is not a structural losing the market share. So we are still doing our efforts to improve our market -- our loading 8-inch, through our specialty technology offering or our -- strengthen our customer engagement in this 8-inch business.

Okay, thank you. Maybe one addition to this question is, I mention you talk about your development and business ramp up from the automobile and automotive business, right, industrial applications especially for the smart automobile. Can you give us more update about that? Would be that one of the main contributor to your 8-inch specialty technology that helped offset any business weakness on display drivers?

Yes, that's pretty correct to our efforts.

And --

It's still a small part to UMC's revenue share. But this, the automotive segments, the growth rate and potential growth is the largest one in our forecast.

Any ballpark number in terms of revenue contribution from auto this year, or maybe year-on-year growth, just a ballpark number.

We don't have the number on hand right now.

Okay, thank you so much. Just one last question. Can you update us a little bit about your next-generation development? I think you said 14 nanometer, right?

Yes. Our 14 FinFET technology development is still on track. And we expect to begin the customer tapeout later in this year. And we're still targeting to enter the 14 FinFET production in the second half next year.

Okay, so mass production in second half next year.

Yes.

Right. Can you just elaborate -- sorry, elaborate more color about the applications in terms of this tapeout?

Yes. The first application will be on the communication and consumer.

Alright. Thank you so much.

Thank you.

Steven Pelayo, HSBC.

Great. A few quick follow ups here, accounting ones. Will 40 nanometer increase in dollars for you in the second quarter as well, or is it a shift over to 28?

Yes, we are expecting overall revenue to grow 5%, 6%. And 40 nanometer as percentage of revenue will at least be flat. So, yes.

So it will grow in line as well. A quick couple of accounting ones here, tax rate thoughts for second quarter and the rest of the year?

Tax rate?

Yes.

It's -- I can only give you the full year. It should be between 10% to 15%. But quarter over quarter it's a bit difficult to predict as in the second quarter, when we pay out more than 50% of our earnings, the remaining unpaid earnings will be levied about 10%, so-called the retained earnings tax. So that will happen in quarter two. But we're expecting our tax rate, effective tax rate in-between 10% to 15%.

Alright. And two more quick ones here. I guess with the China operations and the non-controlling interest line, I assume, as that fab ramps towards the end of the year, I guess there's probably loss sharing that goes on. So how will that impact the P&L as we exit the year?

Again, we own 30% for now, but we do have controlling over this JV. So we will consolidate 100% on the -- for the time being it's on the operating expenses side. When we start production, all the costs will be allocated to its appropriate item. And the remaining 70% will be given back to the other shareholders, regardless it's loss or profit at the bottom line.

So that's what I mean. I'm assuming when you start a fab, you usually start it at losses. You are sharing essentially 70% of those losses with the partner, so that would be an add-back to your P&L.

Yes (multiple speakers).

Can you quantify at all how --

That's already the case for the second quarter, although it's a minor numbers. (Multiple speakers).

How much of an impact do you think it could be this year?

How much would be the impact? I cannot give you a forecast for Xiamen JV as it's a wide range. But, of course, since you assume -- it's reasonable to assume it will be a loss-making entity.

Okay. And you gave full-year depreciation guidance. On a quarter-on-quarter basis, does it step up significantly with 28 nanometer, going to the high-teens percentage of revenue as well?

For the full year it's 20% year over year. And this quarter, Q1, over the previous quarter, was around 6.6%. And quarter two over quarter one is around 5%; 4% or 5%.

4% to 5%. And then my last question; I'm sorry, so many. In your 20-F you did actually reveal the largest customer and it was up 23% year on year, so a very big growth considering the overall Company I think was up what 3% or so? So I'm just curious, is this market share gains that are going on, is there M&A activity with that customer that is opening up opportunities for you? Or do you expect that kind of momentum to sustain? When you think about your largest customer, do you think it will grow in line this year in 2016, or continue to outperform like it did in 2015?

I cannot predict single customers. But we will try to maintain a diversified client base. But the top few customers, of course, will remain very important to our overall revenue contribution. As our CEO just pointed out, we do expect to see one to two customers continue to contribute more than 10% of our revenue base. And they are exactly the same customers as in 2015.

Okay, thank you very much.

Thank you.

Thank you for all your questions. That concludes today's Q&A session. And now 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.

Thank you. And ladies and gentlemen, that concludes our conference for first quarter 2016. 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 Investor Relations, Investor Events section. You may now disconnect. Good bye.