United Microelectronics Corp

TWSE:2303

Earnings Call Transcript 2018-Q1

[Operator]

Welcome everyone, to UMC's 2018 First Quarter Earnings Conference Call. [Operator Instructions] And 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, Investors, Events section. And now I would like to introduce Mr. Michael Lin, Head of Investor Relations at UMC. Mr. Lin, you may begin.

[Michael Lin, executive]

Thank you, and welcome to the UMC's conference call for the first quarter of 2018.

I'm joined here by Mr. Jason Wang, the President of UMC; and Mr. Chitung Liu, the CFO of UMC. In a moment, we will hear our CFO present the first quarter financial results, followed by our President's key message to address UMC's forecast and the second quarter 2018 guidance. Once our President and the CFO complete their remarks, there will be a Q&A session. UMC's quarterly financial reports are available at our website, www.umc.com, under the Investors Financials section. During this conference, we will 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 ROC security authorities. Now I would like to introduce UMC's CFO, Mr. Chitung Liu, to discuss our first quarter 2018 financials.

[Chi-Tung Liu, executive]

Thank you, Michael. I'd like to go through the first quarter 2018 investor conference presentation material, which can be downloaded from our website. Starting on Page 3, the first quarter of 2018, consolidated revenue was NT$ 37.5 billion, with gross margin at 12.4%. The net income attributable to stockholder of the parent was NT$ 3.4 billion, and the earnings per ordinary shares were NT$ 0.28. On Page 3 you can also see our capacity utilization rate in quarter 1 of 2018 was 94% up from 90% in the Q4 2017, but similar level at same quarter of 2017. And on Page 4, our revenue grew up 2.4% quarter-over-quarter to NT$ 37.5 billion, mainly coming from our volume growth. Gross margin was 12.4% as reported and operating income margin is 2.1% or NT$ 769 million. Our total non-operating income was about NT$ 1.088 billion and therefore the net income attributable to stockholder of the parent was NT$ 3.4 billion or NT$ 0.28 per share. On Page 5, the year-over-year comparison, revenue was almost identical around NT$ 37.5 billion with earnings, net income attributable to stockholder of the parent grew 48.7% year-over-year to NT$ 3.4 billion. And on Page 6, our cash and cash equivalents at the end of first quarter was NT$ 77 billion with stockholder equity around NT$ 216 billion. Page 7, our blended ASP in quarter 1 was relatively flat compared to the previous quarter. So on Page 8, our sales breakdown by countries, by locations, Asia, still the largest component of 47% with North America around 42%. So IDM, on Page 9, account for about 8% of our total revenue, similar to that of Q4 last year. And in terms of segment breakdown on Page 10, Communication, although declined as a percentage of revenue in Q1, 2018, still represent the largest pie of 47% and Consumer is 29%. On Page 11, for technology breakdown, 14 nanometer continue to be around 2% of total revenue. And as we mentioned earlier 28 will be at trough of our run rate, which in Q1 was around 12% of total revenue versus 15% in the previous quarter. And 14 nanometer continue to expand to now 30% of total revenue in Q1 2018. And 2018 on Page 12, in terms of quarterly capacity was the lower point due to the shorter working days and annual maintenance and quarter 2 we see a back to normal condition with further expansion -- capacity expansion at our 12X fab in Xiamen, China. And on Page 13, our total CapEx budget for 2018 remained unchanged at U.S.$1.1 billion. The above is summary of UMC's result for Q1 2018. More details are available in the report which has been posted on our website. I will now turn the call over to our President, UMC, Mr. Wang.

[Jason Wang, executive]

Thank you, Chitung. Good evening, everyone. Here I would like to update the first quarter operating result of UMC. In the first quarter of 2018, despite unfavorable movement of the NT dollar, our foundry revenue increased 2.5% quarter-over-quarter to NT$ 37.44 billion. Stable loading across 8 inch and mature 12 inch technologies resulted in a overall utilization rate of 94%. That brings our wafer shipment to 1.75 million 8-inch equivalent. Softening demand in smartphone and other wireless device was more than offset by the strength in computer and the consumer segment.

Now we looking into the second quarter, we anticipate our wafer shipment to increase, mainly due to a growth business opportunity from wireless communication as well as computer peripheral related chip demand. As we secured new product tape outs across the advanced and mature technology, including 28 nanometers, we will leverage our collaborative efforts with customers, supported by UMC's manufacturing excellence to enhance our market share and financial result. While, we focus on revenue and product growth, we also seek to maximize the shareholder returns. Subject to the shareholders' approval the Board of Director recently proposed a cash dividend distribution of approximately NT$ 0.70 per share, which constitutes around an 88% cash dividend payout ratio. In addition, we begin our 18th share buyback program in March 8, 2018. And we will complete the purchase of shares for cancellation by May 7, 2018. Moving forward, we will continue to improve UMC's profitability and maintaining business growth to maximize benefits to our shareholders. That was our first quarter results. Now let me go over the second quarter's guidance. For wafer shipments, we will show an increase of 2% to 4%. ASP in U.S. dollars is expected to remain flat. The gross profit margin will be in the mid-teens percentage range and capacity utilization rate will be mid-90% range. For foundry CapEx of the 2018 year, it will be remain at U.S. $1.1 billion. That concludes my comments. Thank you all for your attention, and now we are ready for questions.

[Operator]

[Operator Instructions] Your first question is coming from Randy Abrams from Crédit Suisse.

[Randy Abrams, analyst]

First question, I wanted to ask about the 8 inch and mature 12 inch, where you mentioned tight utilization and good demand. I'm curious, if you could talk about potential to lift pricing and profitability on some of these older nodes. And second part is, how you're thinking about capacity there with the tight utilization. If you have clean room space or plan to add capacity? Or if you'd go so far as looking consider fab -- like new fab space like we've seen with Hua Hong and Vanguard considering.

[Jason Wang, executive]

Okay. Randy, good afternoon. For your first question and in terms of the 8 inch and 12 inch of mature nodes price increase. Well the price adjustment, we have not increased pricing simply because there's strong demand. In the past few quarters, we did have discussion with customers on pricing adjustment because of raw material. In the past years the raw material has continue to increase, especially in 8 inch area. So we're making some adjustment there. Our pricing strategy will reflect the raw material cost increase and we'll continue monitor that, but for the other -- demand reason, we don't easily just the increase our pricing. But we are focused more on product mix improvement because the loading.

[Randy Abrams, analyst]

And just a second part of that mix improvement and passing on raw material, I guess do you see potential to get margin improvement from some of those activists. Say, if you can pass on some of the raw material costs, you could see like structural profitability such to recover some of that structural profitability on some of these nodes?

[Jason Wang, executive]

Absolutely. We do expect some improvement on that. That is, we try to offset some of the annual erosion curve and so we hope that will help and for the product mix improvement also help that as well.

[Randy Abrams, analyst]

And do have room for...

[Jason Wang, executive]

And that was for your first question. And then your second question related to, if there is any room for capacity increase? For the 8 inch there is very little room. And we actually in our 2018 CapEx plan for cash base, we are having some CapEx allocated to 8 inch. But most of those would not happen in 2018. Once we spent it, we'll probably not going to see that in the 2018 yet. For the 12 inch, we are doing some of the [ refractability ] in terms of node to node changes to support different mix of the demand. And so we do have some room to increase the capacity, but we haven't seen the need of that. The -- most of our CapEx decision will follow our strategy we've mentioned about in the past and we want to make sure it is ROI driven and we want to make sure that the justification in place before we make that.

[Randy Abrams, analyst]

The other question I wanted to ask on 28, just if you could give us a sense of the outlook. You mentioned the trough this quarter. So how you see the rebound as we progress through the year? How you're seeing engagements? And if you could also talk about competition, both from a pricing and also competing for designs. Now that there's some excess capacity, it looks like even TSMC and some of the other foundries on 28. So how you see the rebound and how you see the competition pricing?

[Jason Wang, executive]

Okay. For the 28 rebound -- well, obviously, Q1 is a challenging quarter for us. We predict the contribution from 28 is only at about 12%. And for going into Q2, we do see good progress in terms of customers tape out and product diversifications. We expect the 28 nanometer revenue will pick up in Q2 quarter-over-quarter basis. And we actually -- for the 28, we continue to see that progress, okay and -- well, going into Q3 as well. But if I separate the poly-SiON and High-K. The poly-SiON, our business remain pretty solid and -- because first is we have a stable demand. Second is, because the SiON solution is ready, so we haven't seen much of the solutions changes there. So SiON actually remains pretty steady there. For High-K its little bit different. There is few reasons. One, the High-K is intensive due to the evolving technologies. We continue see the High-K moving from HPC to HPC plus to 22. So, however, we do see customer pilot rounds and new tape outs in a pipeline, the number of the tape out on High-K is higher than -- higher in second half than the first half. So I do see that as a progress as well. But the other point in High-K that I want to mention about is the new customer base. These new customer base are more fragmented and small volume. So -- and the customer engineering requirement are taking longer time and effort to support. So we believe the 28 High-K will take a little bit longer to recover. But the poly-SiON remains more solid than the High-K at this point.

[Randy Abrams, analyst]

Okay, appreciate the good color. Is there a way to put maybe percentage -- like you see on percentage how quick it may come back? Knowing it's fragmented and may take time, but if there's a way to think, how you think the contribution may come up from the 12%?

[Jason Wang, executive]

You mean from the revenue contribution point of view, I think Q2 be better than Q1. But then we are not guiding the percentage specifically. Again, based on why it is, right, earlier, the poly-SiON remain more -- a bigger portion of that. Thank you.

[Operator]

And the next one is coming from Michael Chou from Deutsche Bank.

[Michael Chou, analyst]

Couple of questions. One is, what is your currency assumption for Q2?

[Jason Wang, executive]

The currency assumption?

[Michael Chou, analyst]

Yes.

[Chi-Tung Liu, executive]

So our Q1 weighted average ForEx exchange rate was NT$ 29.3. For the quarter 2, we don't -- again, we don't do forecast. We just follow the major prediction average by leading banks. For time being, we expect to see small appreciation in NT dollars. But our guidance doesn't involve NT dollars exchange rate. Our guidance in ASP flat is in U.S. dollars.

[Michael Chou, analyst]

Okay. What would be your tax rate in Q2 and as the rest of the year? For the whole year, what is your view?

[Chi-Tung Liu, executive]

Generally speaking, it's always around 15% to 20%. However, I would like to take this opportunity to explain. In quarter 1, we book a pretty big amount of tax credit because there's a Taiwan tax corporate rate increase from 17% to 20%. And as a result, we have booked a onetime provisional tax credit in Q1, and that's the reason why we have tax credit instead of tax expenses for quarter 1. And that would be a single one-off event in first quarter only. And for the whole year average, again, it should be something like between 15% to 20%.

[Michael Chou, analyst]

So you mean 15% to 20%, that means each quarter or that's a whole year average?

[Chi-Tung Liu, executive]

Whole year average.

[Michael Chou, analyst]

But with the Q1 tax credit, so this year tax rate will be lower than 15%, right?

[Chi-Tung Liu, executive]

No, no, it still will be around 15% to 20%.

[Michael Chou, analyst]

Oh, okay, okay. The other question is regarding the [indiscernible] statement of the 28 nanometer progress. Seems like that High-K/Metal Gate will take a longer time to recover. Is that because of the pricing pressure or you need to catch up with 22-nanometer product introduction in the rest of the year. So what is your schedule for 22-nanometer mesh production?

[Jason Wang, executive]

Okay. So for the year, for the 20-nanometer High-K, the major reason I actually mentioned earlier is because the market has evolving technologies, and we are still under the catch-up mode. So in that regard, the recovery remains challenged. And in terms of volume recovery, because the new customer base more fragment, small volume, so that's adding the more challenges to that recovery. So that the comment we make, it will may take a bit longer than that. So if you're asking specific for 22 nanometers, our 22 nanometer current stage is under the process development and the schedule is on track to be ready for customer tape out in first half of '19. But, meanwhile, for 28 High-K's engagement, we're concentrated at the 28 HPC and 28 HPC plus.

[Michael Chou, analyst]

So for your 22 nanometer tape out would that be fully in High-K/Metal Gate together or High-K only?

[Jason Wang, executive]

It's High-K, yes.

[Michael Chou, analyst]

High K.

[Jason Wang, executive]

Thank you.

[Operator]

And the next question is coming from Bill Lu from UBS.

[William Lu, analyst]

My first question is on capital spending. You are saying NT$ 1.1 billion for the full year, Q1, I think, is roughly NT$ 190 million. So on a run rate basis, Q1 is lower than your full year guide? Is the expectation that it will pick up in the second half of year? Can you just talk little bit more about that?

[Chi-Tung Liu, executive]

I think we're still tracking to this NT$ 1.1 billion CapEx budget and maybe quarter 2 will show some slightly increase on a quarter-over-quarter basis. But this is actually a small amount in absolute dollar term and the whole four quarter is actually not different significantly to each other. So this is actually -- even there's a pickup in second quarter, it will be a small amount.

[William Lu, analyst]

So 28, I think there is a bit of capacity around the world. That doesn't change your plans at all in terms of the China expansion, correct?

[Jason Wang, executive]

No, it doesn't change much right now. Out of the NT$ 1.1 billion, 33% is actually 8 inch and 67% is the 12 inch. And on top of that there is minimum of the general budget spending is estimated about 20% in there, so out of the NT$ 1.1 billion. So it's relatively small in terms of the CapEx spending for the 28, yeah.

[William Lu, analyst]

I mean, just that TSMC still said that 28 is little bit softer right now, but longer term, they're still confident that 28 is going to be a long lived node. Is that the way you are thinking about it as well or what's the give and take on your expansion decision?

[Jason Wang, executive]

Well, we believe the same thing. I mean, that's why we decided and we put the capacity for 28. There are some assumption differences. Originally, we saw the High-K will be more stronger than the poly-SiON and now we see -- because the product -- the technology readiness, we see a much better healthier loading for the poly-SiON than the High-K. And -- but I've mentioned earlier, the High-K mainly because the company is evolving. So that make it bit of a difficult to engaging customers. So -- but in a long run I remain confident in the 28 in general.

[William Lu, analyst]

One last follow up on this topic is. You talked about 28 seeing more of the High-K tape outs in the second half. As you engage these new, smaller customers, do they also want 22 as well or is 28 High-K good enough for these smaller customers?

[Jason Wang, executive]

Is our point there is a natural migration. Once you get into 28, you could consider 22. But, again, subject to the different applications, business condition if the customer want to spend investment into the 22. But it is a natural migration from 28 to 22. For the foundry business, we basically is set of 2 set, so it's more just the different flavor.

[William Lu, analyst]

My second question is on your 2Q guidance. If I look at shipments up 2% to 4%, ASP flat. That brings you back to a pretty similar revenue run rate as last year. Gross margin, you are guiding for mid-teens versus at a similar run rate last year, you were getting 17%, 18%. Can you help me understand what is the difference in terms of FX, in terms of maybe pricing mix, et cetera?

[Jason Wang, executive]

Well, from a pricing wise, that's better with year-over-year erosions. There are some erosions included in that assumption. And other is, in a same year last time -- same time last year, the foreign exchange rate is different. So we see some unfavorable foreign exchange rate in Q2. Then higher material wafers as we all know. The -- other than that, we see some of the 28-nanometer loading improvement, but not as significant yet. So I think that's probably the biggest difference between the 2.

[Operator]

And next one is from Donald Lu from Goldman Sachs.

[Donald Lu, analyst]

May I ask first question is the guidance for depreciation and R&D for this year. And my second question is can we get more color on this pretty big tax credit in Q1?

[Chi-Tung Liu, executive]

Depreciation for 2018 is about the same as 2017, with each quarter it's pretty linear. We expect to see some decline starting from 2019.

[Donald Lu, analyst]

So it's a flat year-over-year?

[Chi-Tung Liu, executive]

Yes.

[Donald Lu, analyst]

Depreciation. Okay.

[Chi-Tung Liu, executive]

And as for tax credit, again, as Taiwan corporate tax rate increased from 17% to 20%. So the theoretical tax credit UMC is entitled to use also increased, because of our capital expenditure investment in R&D, et cetera. So we have to book a onetime provisional tax credit in Q1 this year, and that amount is close to NT$ 800 million.

[Donald Lu, analyst]

How about R&D guidance for this year?

[Chi-Tung Liu, executive]

R&D again, it should be somewhat under control perspective and it shouldn't be more than that of last year.

[Donald Lu, analyst]

Okay. R&D expense should be flat to down year-over-year?

[Chi-Tung Liu, executive]

Flat to down year to -- overall, R&D cost should be flat to down compared to last year. But the magnitude of change won't be significant.

[Operator]

And next we have Charlie Chan from Morgan Stanley for questions.

[Charlie Chan, analyst]

Very good execution in a kind of tough market in first half. So my question is on your 28-nanometer business still. First of all, if you look at those new tape outs, do you think majority of them comes from your 40-nanometer project migration? Or is there any new projects of new applications?

[Jason Wang, executive]

Well, not 100% from migrations. They're coming from communication, consumer. We also see automotive infotainments, the ADAS. So they are new application. And some of it is coming from 40-nanometer, but many are actually also new, not 100% from that, yes.

[Charlie Chan, analyst]

Okay, yes. So because we see your quarterly reports, it shows that 40 nanometer is still like 30% of your first quarter revenue, 28 nanometer is 12% right? So my question is that if half of your 40-nanometer business migrate to your 28-nanometer, do you see any kind of risk to fill up your feeding 40-nanometer capacity?

[Jason Wang, executive]

Well, the customer pipeline management is always very critical to us. For the 40 nanometers some of the product will migrate, but some product also will migrating into the 40 nanometers. We see different application coming into a 40. Given time, in the longer term, I think the 40 will also remain very healthy. But throughout the time it may have some quarter coming -- fluctuate a little bit. But I still feel 40 is healthy.

[Charlie Chan, analyst]

And my second part of question is about your profitability. So I am not sure if you've disclosed it before. But can you comment on your 28 nanometer gross margin in the trend into the year end. And also for your Xiamen fab, at what wafer capacity do you get to the economy of scale, right? So my question is basically, how much more CapEx or capacity you need to expand in for the China fab.

[Chi-Tung Liu, executive]

Xiamen fab currently should reach 17,000 wafers by mid of this year and the target is to ramp up to 25,000 wafers to be -- to reach the economy of scale. It could be by the end of this year or the next year to get there. And for profitability, because it's still up and running in very early stage, so it has been loss making and it actually eat up UMC's gross margin by 4% to 5% at least. So that's part of the reason why on a year-over-year comparison basis, our margins seems to be lower and Xiamen is also a major -- one of the major factor.

[Charlie Chan, analyst]

So for 28, what utilization can you get to the breakeven and the timing for that breakeven, can you give us some color?

[Jason Wang, executive]

For the gross margin percentage we don't actually break down by fab or by particular node. So as Chitung mentioned earlier, the gross margin in 2018 will be -- it's more challenging than 2017, it's because the depreciation will be peaked around 2017 to 2019 and we think that we'll -- actually that we'll actually will start seeing some improvement. The part of the reason Chitung mentioned 12X, we see 12X will be a margin drag and we'll see the light loading of 28 will be a margin drag. And other than that -- so that's also the fact what we've mentioned is the -- 2018 gross margin it will be higher than the -- I mean more challenging than 2017. In general, the -- we are under -- so what we call a transitioning period -- okay, because we do see that as a issue and we defined that as a structural profitability issue, so we would like to pursue that. And one other approach that we take is, what we mentioned in the past, is the proceed with a more disciplined CapEx approach, so all the future investments will be more of a ROI driven to enhance that, okay. However, for the past CapEx we spend in a past few years did increase in our annual depreciation, so we have to get over that hump and that's our current plan. Thank you, Charlie.

[Operator]

And the next on is from Gokul Hariharan from JP Morgan.

[Gokul Hariharan, analyst]

First of all, could you talk a little bit about -- you mentioned that poly-SiON is still a major of the mix. Based on what do you see in terms of tape outs et cetera, when do you expect High-K/Metal Gate or HPC or HPC plus nodes to be majority of your 28 nanometer. Do you have any visibility within second half of this year or it could take into 2019?

[Jason Wang, executive]

Well, for poly-SiONs the tape outs continue, the engagements also continue and we have pretty diverse application in the area. So I think that the demand remain very solid. And for the High-K, because there are more fragments, more form a customer base point of view. We don't have a good visibility for when it's going to cross over, but actually it's not in the 2018.

[Gokul Hariharan, analyst]

Could a talk a little a bit more in terms of the work you are doing with the some of your 8 inch specialty technologies trying to move into more higher end automotive industry from the more consumer driver IC kind of stuff. Can you talk about what other process that's going on there, what kind of price increases ASP increases could you see on a, let's say, 2 to 3 basis from doing that. Just give us some color on that.

[Jason Wang, executive]

Okay. Let me -- first of all the 8 inch fab continue to operate at full capacity as everyone knows and this will continue in -- throughout this year is what we believe. And given this worldwide shortage at 8 inch, our field of focus now is one focusing on long term partnership engagement, okay. So we want to make sure that we're serving the customer right, so that we can establish some of the long term partnership, and so we try to establish some kind of the longer term alignment with the customer so the demand and loading can go beyond 2018. The same time we look at different applications and from our demand portfolios and some of the product mix actually helping in terms of blended ASP and we'll actually try to align with those and we'll try to support and not only from a technology standpoint, also from the capacity standpoint. We haven't really talked about the increase of pricing, because I mentioned earlier, we don't simply increase the pricing because the demand surge. We actually more manage that through the product mix improvement. And at the same time, we actually are having a some of the CapEx spending in 8 inch as well that's being our CapEx projection. Those will help us to increase slight -- some improvement of the 8 inch capacity. And so combination of those, we hope the 8 inch ASP and can be maintained or a giving a better way to offset a erosion. So we have some ASP maintenance in there.

[Gokul Hariharan, analyst]

Just one clarification, Chitung, could you clarify your comments on depreciation? I think last time you talked about maybe 0% to 3% decline in 2018 depreciation, if I heard it right. Are we looking for flat in 2018 and decline in 2019? Is it that for depreciation?

[Chi-Tung Liu, executive]

I think this quarter still flat to -- I mean, this year is still flat to 2017, maybe small decrease. But, I think, flat is probably the better way to describe it. But not just decline really in the 2019.

[Operator]

And the next one is from Sebastian Hou from CLSA.

[Sebastian Hou, analyst]

My first question is on the gross margin guidance for second quarter, well up from low teens to mid-teens. So is it more due to better utilization rate or product mix or pricing adjustment?

[Jason Wang, executive]

First of all we have more working days. In Q1 we actually have a fewer working days in annual production to maintenance and we don't have that in Q2. Secondly, we have a improved 28 nanometer loading. So between the 2, we actually see some improvement on the gross margins. The foreign exchange rate and the raw material cost continues -- remains same, so that doesn't give us much of a benefit, yes.

[Sebastian Hou, analyst]

So there is not much of pricing adjustment impacts on the second quarter margin improvement?

[Jason Wang, executive]

The blended ASP will increase, because if you have higher 28 nanometer loading. So -- but at this point, we see that 28 nanometer improvements actually help us to minimize the quarter-over-quarter price erosion. So we actually guide that ASP is flat.

[Sebastian Hou, analyst]

And the second question is on the OpEx. It dropped about like 7% quarter-on-quarter in first quarter. Do you see this kind of the absolute level kind of like the sustainable level over for each quarter throughout this year or will continue to be -- likely to be lower.

[Jason Wang, executive]

Sebastian, you're referring to the Q1 '18 operating expense?

[Sebastian Hou, analyst]

Yes.

[Jason Wang, executive]

I see. Well, in Q1'18 the operating expense reached actually a sharp decline due to the very challenging Q1 environment when we go into Q1. So we've been control that seriously. But the trend is not sustainable at this point. But quarter operating expense we will continue monitoring that closely and spend all effort to control. But that is not a trend that is sustainable, no.

[Sebastian Hou, analyst]

So which means as the situation turns better from second quarter onwards, particularly you mentioned 28 UTR improved, the OpEx likely to increase sequentially again in second quarter and second half.

[Jason Wang, executive]

Well -- yes, I mean the other way to look at this is, our 2018 annual operating expense budget will be similar to our 2017.

[Sebastian Hou, analyst]

And Jason earlier in your prepared remarks you talk about the growing opportunities in wireless communication and computer peripheral related chip demand. Can you elaborate more little bit in detail on the what specific chips are you seeing more opportunity?

[Jason Wang, executive]

If we break it down by the 3 C, the computer actually is strong -- strongest. And it coming from a tablet, fiber controller followed by the communications and we see some RF switch, AP and baseband. So -- but consumer is actually the weakest.

[Sebastian Hou, analyst]

I remember the company earlier talk about the opportunity in RF SOI. So how do you see this platform going forward into -- particularly when we enter 5G? And then how much of this roughly numbers -- how much of this account for your total revenue?

[Jason Wang, executive]

Well, I do not have a specific percentage of that. But the RF SOI is area we've been focused on quite some time and we have some success in 8 inch and we are continuing expanding our technology from 8 inch to 12 inch. But, of course, the 12 inch -- the current -- the overall ecosystem is not completely ready yet. So the near term focus will continue to be a 8 inch supporting our RF SOI business needs and continued developing the ecosystem for the 12 inch RF SOI and it will be our long term focus.

[Sebastian Hou, analyst]

In terms of the IC components in the RF module, this RF SOI, do you still focus more on the switchboard, do you also see rising opportunities on the poly-amplifier?

[Jason Wang, executive]

Well, I mean we do see that. I mean, we see that as continued growth area, but not all of it is UMC addressable. So we -- it's because the product -- the customer engagement as well as our technology offering, so the -- but if your question is about the market outlook, we actually have pretty optimistic view in this area.

[Sebastian Hou, analyst]

Last from me is 28 nanometers. Last quarter, the company talk about the China being aggressive here and you did see some continued pricing pressure. And how do you see the competition from China at this moment and also the pricing environment throughout the end of this year?

[Jason Wang, executive]

With the China -- well, I have to put this way. With the China, national policy of fostering the local semiconductor -- manufacturing environment, you will see more people coming into this market. But at the UMC, we position ourselves to grow along with this Chinese semi market and with our both 8 inch fab do we have one and the 12 inch in Xiamen, we can certainly entertain all those opportunities. Pricing is a market competitiveness issue. We will remain competitive in this market. I mean not only on pricing, also on the manufacturing capability as well as the yield and the cycle times so on. So, yes, I mean we'll sure we will be competitive in the area. Sure thank you.

[Operator]

And next in line will be Roland Shu from Citigroup.

[Roland Shu, analyst]

First, may I know your view of the overall semiconductor foundry and the TM -- and the UMC's, of course, for this year?

[Jason Wang, executive]

Well, I mean semi growth remain there. I mean, we see the overall semi is the mid-single digit for the year and foundry probably a little bit better. We -- for us, 2018, we will see some growth. But it is not going to be as high as the foundry growth. Yes -- I mean as our strategy changed in mid last year, so we do -- we'll see some growth, but not as high as the foundry growth.

[Roland Shu, analyst]

For TSMC's 28 nanometer, 40 nanometer and above nodes, total revenue is ready to decline this year. And your is better, your total revenue will be increased this year. So are you taking share against TSMC or can we have a better view of where does the growth comes from by technology node?

[Jason Wang, executive]

Let me see if I get. First of all, I think, our growth year-over-year will be smaller than the foundry growth, so in that regard I do not think are getting market share. And as far as the TSMC, I do not comment on competitions. And so it's hard to comment that. So -- but I -- that's couple of reference. And I kind of forgot your third question. I am sorry.

[Roland Shu, analyst]

Yes. Just -- I would like to have your breakdown, the growth breakdown for your -- by technology node. So which node will grow strongest and which node will probably decline.

[Chi-Tung Liu, executive]

So, Roland, we have full capacity utilization rate for almost every node expect for 28 High-K and we are continuing to expand our 8 inch capacity at [ Fujian ] as well as some 12-inch capacity in Xiamen and there is also efficiency improvement across the whole fab we operate. So I think that's where the growth mostly coming from.

[Roland Shu, analyst]

Second question is on your -- just the other side, talk about -- if the worldwide 8 inch capacity shortage and now there are some newcomers you're seeing 12 inch aluminum interconnection capacity to build these application based on 8 inch uses related. So will you consider to restart your 12 inch alumina line or maybe you will upgrade or retrofit your 12 inch ideal capacity to do this aluminum interconnection to capture the 8 inch business opportunity going forward.

[Jason Wang, executive]

Well, first of all -- I mean our capacity vacancy is mainly at 28 High-K and so those capacity is not suitable for the 12 inch aluminum line and the inches in the 12-inch aluminum line is there, we are evaluating it. If follow -- our policy is unless it becomes business justifiable we will not get into that. But -- so this decision will be more of a business oriented as long as ROI we can justify. And, yes, I mean we have the financial feasibility to do so and we have the technology to do that as well. But we want to make sure it is ROI justifiable.

[Roland Shu, analyst]

Last question then, maybe Chitung can give us some more color. In first quarter we have big amount on this net other operating income and also net non-operating income in first quarter.

[Chi-Tung Liu, executive]

Other revenue is mainly from subsidies or grant from government. And then, of course, our Xiamen fab is main beneficiary of that in terms of other revenue. And the non-operating income is mainly foreign currency. Again, this is similar to Q4 of last year that are Xiamen fab -- the function currency is renminbi, but they borrow U.S. dollars as a syndicated loan and the strengths in renminbi contribute to the majority of the non-operating income in Q1.

[Roland Shu, analyst]

So for this operating income are we going to have the similar amount or even bigger amount in second quarter for this [ capacity ]?

[Chi-Tung Liu, executive]

No, other revenue every quarter should be similar, but non-operating income is highly dependent on the currency which we can now predict, so we don't expect to see -- we don't make any prediction for the second quarter.

[Roland Shu, analyst]

So the other operating income would be similar amount every quarter through end up this year and the going forward, is it right?

[Chi-Tung Liu, executive]

No, only end of this year, not going forward.

[Roland Shu, analyst]

Okay, just end up this year. Okay.

[Operator]

And the next one is from of Angus Milne of HSBC.

[Angus Milne, analyst]

Just kind of recap questions on the Q1. So can you talk us about the 45, 65 nanometer strength especially pretty strong in 1Q. Could you give us some colors on maybe end market and product application strength over there. And also 20 nanometer, you had a puke down of almost 20% in 1Q quarter-on-quarter. Can you give some on color like how much is that coming down lower wafer shipment and how much of that is coming from -- actually coming from lower ASP.

[Jason Wang, executive]

I'm sorry I kind of lost the question. Can you repeat that? Thank you.

[Angus Milne, analyst]

Yes, just on the -- maybe just 20 nanometer, you had a puking of first quarter, right, down like 20% quarter-on-quarter. We just want to know about like how much of that decline is coming from lower wafer shipments as opposed to lower pricing?

[Chi-Tung Liu, executive]

Well, if you talk about 20, it's mainly because of High-K as a lower utilization rate, not because of pricing.

[Operator]

And next we have Randy Abrams from Crédit Suisse for questions.

[Randy Abrams, analyst]

Xiamen, I mean, Chitung you mentioned the 4 to 5 basis point or point some gross margin. I mean, could you talk about just what you need to do to get back? Is it -- like how much is it getting to scale to 25k or whether there is under loading in that fab or you need to just bring the performance up to the Taiwan -- like the Taiwan standard yield do you have. So if you could just talk about how you could close the gap and maybe a timeline to reducing that drag in Xiamen.

[Chi-Tung Liu, executive]

For the time being we don't really have exact target as yet in mind in terms of -- maybe if we take our Singapore fab as a example, as a reference. It probably take more than 4, 5, 6 -- 4 or 5 years at least given the difficulties in small scale single standalone fabs. But the subsidies certainly offset the operating losses we encounter right now. So I think it's a multiple dimension in terms of how far we need to bring the capacity up to the economy of scale and at the same time load the fab simultaneously. And so I don't see in the near term we're going to see a profit at Xiamen fab. But the improvement definitely will start kicking once we reach the full capacity.

[Randy Abrams, analyst]

And then the second part was just, is full capacity now 25k? I think when the fab was announced it was 40k or 50k. But will that...

[Chi-Tung Liu, executive]

2 modules, 2 stage. So stage 1 is 25 and stage 2 is another 25.

[Randy Abrams, analyst]

And have you made a plan yet for stage 2. Like is it still dependent on ROIC or is it really the plan depend to scale as like a 2 fab concept. And so ultimately in a few years perhaps 50k there still for logic?

[Chi-Tung Liu, executive]

We will focus on stage 1 only first right now. So we want to -- of course, tried to let the operation -- at least can contribute some of the investment for the future. So it also takes time to see the cash flow to come out. So we will focus on stage 1 for now and try to bring up the operating efficiency first before we come out with fab 2 plan.

[Randy Abrams, analyst]

And then just I think about it then the CapEx would come down even further like. Once you fill the 25k in the next year, do you then have not much even space to add capacity so that you go truly into a low CapEx mode or how do you think -- or do you have places you can keep expanding if you still have growth to go after?

[Chi-Tung Liu, executive]

We have some more space for expansion. We saw opening of the phase 2 node between Xiamen and Taiwan. Right now the -- for CapEx projection we're giving out for year -- for the 2018, and the 2018 -- '19 we'll actually leave that as an option in order to discuss and hopefully we can give you some guidance by end of the year.

[Operator]

[Operator Instructions] And the last question is from Sebastian Hou from CLSA.

[Sebastian Hou, analyst]

Just one question is on the CapEx and cash flow. So is in by the new strategy of the companies that will have a lower CapEx going forward. So theoretically free cash flow may increase quite significantly in the next few years. But as I calculate your -- based on your last year free cash flow and the dividend payout. Although the dividend payout ratio is high divided by the EPS you earned. But the -- based on the free cash flow you generate last year is only account for no more than 50%. So going forward, will you -- your payout policy will be based on the earnings per share or based on the free cash flow per share?

[Chi-Tung Liu, executive]

It's a combination. Of course, we want to maintain certain capability in total of months in terms of cash dividend. But payout ratio is also a ratio -- is an index where you want to make sure we can communicate to our shareholders.

[Sebastian Hou, analyst]

But did you pay -- is it possible that you pay out more than 100% or pay special dividend in any of your consideration?

[Chi-Tung Liu, executive]

We don't consider special dividend at the moment. And payout to -- up to 100% is possible and we did that before.

[Sebastian Hou, analyst]

And 100% above?

[Chi-Tung Liu, executive]

Yes. Thank you.

[Operator]

We thank you for all your questions. This concludes today's Q&A session. I'll turn things over to UMC Head of IR for closing remarks.

[Michael Lin, executive]

Thank you for attending this conference today. We appreciate your questions. As always, if you have any additional follow up questions, please feel free to contact UMC at [e-mail]. Have a good day. Thank you.

[Operator]

Thank you, Mr. Lin. And ladies and gentlemen, that concludes our conference for first quarter 2018. Thank you for your participation in UMC's conference. There would be a webcast replay within an hour. Please visit www.umc.com under the Investors Events section. You may now disconnect. Good bye.