Welcome everyone to UMC's 2020 third quarter earnings conference call. [Operator Instructions] For your information, this conference call is now being broadcasted live over the Internet. Webcast replay will be available within an hour after the conference is finished. Please visit our website www.umc.com under the Investors events section. And now, I would like to introduce Mr. Michael Lin, Head of Investor Relations at UMC. Mr. Lin. Please begin.

Thank you, and welcome to the UMC's conference call for the third quarter of 2020. I am joined 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 third quarter financial results, followed by our President's key message to address UMC's focus and the fourth quarter 2020 guidance. Once our President and the CFO complete their remarks, there will be a Q&A section. UMC's quarterly financial reports are available at our website, www.umc.com, under the Investors, Financials section. During this conference, we may make forward-looking statements based on management's current expectations and beliefs. These forward-looking statements are subject to a number of risks and uncertainties that could cause actual results to differ materially, including the risks that may be beyond the Company's control. For these risks, please refer to UMC's filing with the SEC in the US and the ROC security authorities. Now, I would like to introduce UMC's CFO, Mr. Chitung Liu, to discuss UMC's third quarter 2020 financial results.

Thank you, Michael. I would like to go through the third quarter 2020 investor conference presentation material, which can be downloaded from our website. Starting on Page 3, the third quarter of 2020 consolidated revenue was TWD44.87 billion, with a gross margin at 21.8%. The net income attributable to the stockholder of the parent was TWD9.11 billion and the earnings per ordinary shares was TWD0.75. On Page 4, the income statement quarter-over-quarter comparison. Revenue grew 1.1% sequentially to TWD44.87 billion, based on the combination of shipment increase, as well as ASP increase. However, the strengthened NT dollars offsetting most of the gains from shipment increase, as well as ASP increase. Gross margin as a result after the stronger NT, as well as the higher summer utility costs declined 4.8% quarter-over-quarter to TWD9.77 billion, a 21.8% gross margin. Operating expenses stayed relatively flat to last quarter. However, this is two factors affect the operating expenses. The positive one comes from, we have a reversal of expected credit loss. We collect the payables from one of our customers, which is around TWD500 million also. But at the same time, given that we have to retain our Thailand employees issuing certain equity-based compensations to the key employees and management. So, the compensation cost is also increased by around TWD400 million plus. And the collect of expected credit loss is a one quarter event. However, the increased compensation based upon equity-based scheme is going to lingering for multiple quarters. For the net other operating income and expenses, also we see two one-off factors. So, you can see that most of our net other income and expenses per quarter is roughly about TWD1 billion. But in the third quarter, it went up to TWD2.8 billion, based upon: one, is the once a year annual interest…

Thank you, Chitung. Good evening, everyone. Here, I would like to update the third quarter operating result of UMC. During the third quarter, consolidated operating margin reached 15.9%, while utilization rate remained firm at 97%. Wafer shipments reached 2.25 million 8-inch equivalent wafers. During Q3, work-from-home and home schooling trends continued to contribute to stable end market demand for applications in wireless connectivity, power management IC used in smartphones, as well as high-speed interface I/O controllers found in the computing devices. In addition to demand stability across various end markets, our 28-nanometer revenue grew quarter-over-quarter as customer product tape-outs continued throughout the quarter. Moving forward, we expect to see a sustained increase in the number of 28-nanometer tape-outs, which will further diversify our 28-nanometer exposure to end markets and customers. Looking into the fourth quarter, demand from consumer and computer-related applications will lead to a minor increase in wafer shipments, propelled by ongoing work-from-home initiatives and home schooling. Furthermore, we have seen an uptick in semiconductor demand due to more silicon content in particular applications such as newly adopted -- deployed 5G smartphones, IoT devices and other consumer products. Therefore, current industry landscape appears to show favorable supply and demand dynamics toward foundry. Hence, UMC will pursue a delicate balance in strengthening our customer relationships, while securing interest for our shareholders to ensure our long-term growth. Let's move on to the fourth quarter 2020 guidance. Our wafer shipment will increase by 1% to 2%. ASP in US dollar is expect to increase by 1%. However, surging NT dollars to offset all that benefits from Q4 shipments and ASP growth. Gross profit margin to remain flat, despite adverse currency impact. Capacity utilization rate will be in the mid-90% range. Our 2020 CapEx budget will be $1 billion. That conclude my comments. Thank you all for your attention. Now, we are ready for questions.

Yes. Thank you, President Wang. [Operator Instructions] Our first question is coming from Randy Abrams, Credit Suisse. Go ahead, please.

Okay. Yes. Thank you and congratulations on the good results. I wanted to ask the first question, a two-part on the pricing outlook. If you could give the outlook for ASPs over the next couple of quarters, factoring in the rising foundry pricing on some mature nodes? And also, you talked about further 28 ramp-up. And then the second part of it is, I think the last line in your remarks were about balancing the -- strengthening relationships, while also securing shareholder value. So, I'm curious just -- the risk you're seeing in terms of, if you are lifting pricing versus other foundries doing a stable pricing model, if any risk on market share or pricing, if we do head into a downturn.

Okay. Sure. First of all, thank you, Randy. And as far as the pricing question, despite we mentioned this on last quarter and just earlier, that despite our UMC's pricing power, we believe it has been much improved, especially on the 8-inch business. However, we always have to keep our commitment to our long-term customers. We have been taking advantage of this opportunity to strengthen the customer relationship and enhance the product portfolio pipeline, which will ensure the UMC long-term market position. And at this time, we have settled the 2020's pricing and allocation with our customer. The outlook for 8-inch business appears still favorable for foundry, but the 12-inch ASP will remain firm based on the normal pricing scheme. The overall blended ASP improvement is still subject to actual product mix. For the near-term, the Q3 and Q4 '20, the pricing adjustment was only applicable to customers that require incremental capacity support. So that will probably give you a flavor of the near-term Q3, Q4, as well as some sort of the outlook for 2021. You talk about the -- you mentioned the effects on the 28-nanometer contributions. In Q3, the higher 28-nanometer ASP contribution was mainly offset by the changing in the product mix. We actually see some benefits -- I mean, the benefit of the product mix on 28, but it still got offset by some other product mix. An example is that the decline in our 90-nanometer product, which were designed into the high-end market. That led the ASP [indiscernible] as a parent even with some of the improvement on our 28 revenue contribution. So, that's for the pricing answer. The -- when we talk about the -- we tried to seeking the balance between the shareholder inches, as well as the long-term customer relationship strengthen the customer relation, we're mainly talking about that a suite going into this few quarter, as well as for the outlook for 2021. We continue seeing the business outlook remained pretty firm as we expect the business traction will continue into the 2021. So, the overall business outlook is healthy. So that means the utilization situation will remain very full. And in that sense, we are carefully evaluate our CapEx decision and then consider whether how should we expanding our capacity in supporting customer. But in the same time, we need to watch the balance for the shareholder interest. So that's what we are referring to.

Okay. Thanks. Appreciate the color. Maybe a couple of follow-ups to that. First, for the pricing where you mentioned this year it's only on the incremental business. If there is a way to get a feel, how it's looking at this stage for the pricing for next year? And the second part of that would be on gross margins. Just on your view that you'll be near full capacity, good pricing, 28 rising and also maybe depreciation down, how you feel the gross margin maybe trending?

First of all, when we look at the gross margin, we would -- bit of a disappointment for Q3, because the Q3 was largely affected by the appreciation of NT dollars. Right? So the NT dollar -- the foreign exchange rate that we have some impact to our gross margin and we will expecting it will continue trending up. Now, despite the NT dollar impact is our expectation and goal to continue see the gross margin trend going up and following the better product mix and also the continuous utilization, high utilization.

Okay. And then if I can ask a question, then you've cut-off a 90 node where it's -- I think down a good bit from what's a year ago. So that pocket, I assume is the one with underutilization. Is there kind of a new application to backfill or with capacity tight you can migrate some of the tools to other nodes?

Okay. The major reason for the 90-nanometer decline or impact was due to a decrease in wafer demand found in our Japan region. And particular, this impact affects our 12M, the Japan fab. And one of the important customers are -- is experiencing some deteriorating business condition beginning in Q3. You asked whether we'd be able to backfill with other devices. No, and we already started. We pick up the -- we started the mitigation programs this quarter and we expect is 12M loading will gradually improve. We can really improve immediately, because we had to bring product over and we have to make some modification to it. So the loading will gradually improve each quarter. In Q4 2020, the next quarter, the 12M loading will be the trough of UMC has begun to leverage our other customer base. And so, we expect the duration of this recovery mode will probably left around three quarters, counting from current quarter. So we should start seeing some signs of a recovery from Q1 and continue recovering that to a much normal level.

Okay. And just one other follow-up, because you mentioned the CapEx, you're doing a consideration now that you're full. Should we view base case, it's been around $1 billion, that's the level we're looking at when you're expanding or potential at high utilization could start to advance? And how are you thinking about like a build versus buy on some of the mature nodes, where capacity is tighter?

Both are important. Both are options to us. The -- for the CapEx decision, we are confident and discussing that in [Sony], but since we're going into the Q4 2020 and we're going to quickly finalizing our 2021 CapEx plan, so the 2021 CapEx will be disclosed during the next quarter conference call. And just a note, each of our CapEx investment will still extended carefully through our stringent ROE justification. So, as we promised our shareholders, as our internal policy, so we'll continue managing under that principle.

Okay. Great. Thanks a lot, Jason.

Sure. Thanks, Randy.

And the next question is coming from Gokul Hariharan of J.P. Morgan. Go ahead, please.

Yeah. Thanks and congrats on the good results. First question I had is, how should we think about margins, especially gross margins over the next year or so? Obviously, as Chitung had pointed out in the past calls, 2022, we are expecting a meaningful drop-off in depreciation. But between now and then, given that we are already running at about 95% to 97% utilization, should we think that gross margin is likely to stay in this early 20% level and then we kind of get to a meaningful drop-off in depreciation and improvement in gross margin in 2022? That was my first question.

Okay. I think it's difficult for us to predict the gross margin rate. As Jason just pointed out earlier, it is set in our goal, continue to find opportunities and optimal position -- process to enhance our gross margin. And, of course, we are having -- we have some headwinds from the currency for this quarter and also likely next quarter. So they kind of eat up about at least 1.5% to 2% of our gross margin. So, currency certainly play a quite important role in the near-term. And longer-term, we do expect to see a better ASP coming from our product mix enhancement, also our 28-nanometer ramp. So, we do see quite a few factors will help our overall gross margins and certainly, it is our goal. But just we cannot really predict what's that number for our gross margin for the longer-term.

Understood. If we think about going into next year, given the very tight capacity that you're seeing in -- especially in mature nodes, do you plan to go down the acquisition route again on some of these mature nodes? What is the thinking regarding capacity expansion? I think you talked about some selective expansion in 28-nanometer, but could you talk a little bit about what are your plans on the mature node side as well?

Well, I mean, we also touched that earlier. We continue exploring both options, both organically and inorganically. So the answer is yes, we are always open to exploring new opportunity that will help the Company growth, okay, on both ways. So, if a buy versus made, we will do both. Okay? We think that this year's CapEx, we are putting additional capacity in our 12X, the Xiamen facility, we are maxing up to a 25K per month run rate. And we have a -- we feel comfortable about that capacity and we have confidence that we'll have a good momentum to build that fab. And so, that outlook remain pretty firm to us at this point. And at the same time, we are continually exploring others. And for that additional capacity are mainly for the 28-nanometer product.

Okay. Understood. One last question on OpEx. How should we think about OpEx? I think OpEx ratio has been coming down in the last year or so. But in terms of OpEx and R&D expenses, are we basically going to stay at around these levels over the next couple of years or are there some increases for some of your specialty products in 28-nanometer expansion?

Yeah. Other than the one-off impacts in the third quarter of 2020, which I mentioned, one is the reversal of our expected credit loss. So there was a gain of around TWD500 million. At the same time, we also increased our equity-based compensation schemes to our employees. So that also increase our OpEx by another TWD400 million plus. That's going to last for multiple quarters. So, in absolute dollar term, we expect the OpEx to increase starting from Q4 of 2020, but then, of course, for the longer term, as a percentage of revenue, we hope it is our current duty to keep the ratio somewhat flat or even come down slightly.

Okay. That's very clear. Thank you very much. Thank you.

Thank you.

Thank you.

And the next question is coming from Bruce of Goldman Sachs. Go ahead, please.

Hi. Good afternoon and thanks for the great result. So I want to ask more about the 28. So what is the current utilization for 28-nanometer in third quarter and fourth quarter?

Let me take a look at the -- the 28-nanometer right now is mid -- about mid-90s for the quarter. And we expect 28-nanometer loading will actually continue the increase quarter-over-quarter.

So 28-nanometer is already at 90%-plus utilization rate?

Yes. In Q4 '20, our current projection is at mid-90s. Yes, it's over 90% utilization rate.

What was it in third quarter?

Third quarter is also over 90%, but is not reaching the mid-90s yet.

I see. So given the high utilization rates, so what is the capacity expansion plan for the 28-nanometers? Because since we are already at a very, very high utilization already.

The capacity will be -- the 28 will increase in -- starting from Q2 2021 and the number will reach above -- by the end of the 2021, we'll probably see about 42K from the end of the Q4. 2020 is a 34K. So there is approximately 8K increase on year-to-year basis. And the number with ramping in Q2 '21.

I see. Okay. Because the CapEx, I just saw that you only spend about like TWD500 million year-to-date. So we didn't really see the spike in terms of CapEx compared to last year, given the high utilization rate. So we -- I'm just surprised that we do not see a more aggressive CapEx or capacity expansion plan for the 28.

I mean, we have been leveraging out the existing pool and we've been cautiously deploy our CapEx plan. And the following today's full resolution, I think you will see announcements that, along with today's CapEx approval, the number actually going up quite a bit today.

I see. Understood. So another question is for the gross margin. So what is the forex impact for the gross margin in third quarter? And what would be the gross margin guidance, assuming the same forex? I mean, I just want to know what is the real gross margin improvement excluding the forex factors?

Back to our Q4 guidance, we are guiding 1% to 2% shipment growth and 1% ASP growth, that's all going to be wiped out by forex. So that's the magnitude of the forex impact for gross margin in Q4. And for the magnitude in the third quarter is a little bit less than that of Q4.

A little bit less than what's the impact in third quarter, is that right?

A little less than that in Q4.

I see. Understand. Thank you.

And next question is coming from Szeho Ng, China Renaissance. Go ahead, please.

Hi. Hello, gentlemen. My first question is regarding the non-controlling stake. In Q3, the number has actually came down quite a bit. I just want to know if it is one-off or any specific reason for the drop?

You mean, non-controlling interests?

Right. The minority interest.

Yeah. That's because the key operation is our Xiamen fab and their loss in the third quarter is much less because of the one-off annual -- once a year interest expenses subsidies received for the third quarter.

Okay. Got you. And interest subsidy would continue for next couple of years in Q3 every year?

In Q2 or Q3, we'll continue for a couple of years. But the debt outstanding should gradually come down as we gradually pay off the debt.

Okay. And second question, back to the gross margin, is it possible to talk about your gross margin difference between 8-inch and 12-inch? And I believe that 12-inch gross margin will be lower, right?

No, we don't really do that kind of breakdown. And -- but as you can imagine, the overall depreciation unit cost for 8-inch is much lower than that of 12-inch. So, accounting margin-wise, yes, 12-inch is based upon our full loading condition right now is certainly higher than that of 12-inch.

Okay. Okay. Yeah. And last question on R&D, how should we model the figures going forward? Should it be flat steady from absolute dollar level?

Sorry. Say that again.

The R&D, how should we model it going forward? Should -- is it going to be kind of stable on absolute dollar basis?

Yeah. We believe it will remain stable for the foreseeable future.

Okay. All right. Okay. Thank you very much and congratulations.

Thank you.

And next we'll have Charlie Chan of Morgan Stanley for questions. Go ahead, please.

Hi. Good afternoon, and thanks for taking my question and congratulations for a great results. So I've got a couple of questions. First of all, it's more about the short-term, because there seems to be some trends some customers still want to transfer the project to your foundry, given concern about the SMIC's continuity. So does the management also see a similar behavior from customers? And how long you think they can sustain? Thanks.

So, well, first of all, it's our policy that we're not commenting on our competitor. And as far as the question you're referring to is more of a geopolitical tensions that is our belief that did have some impact to the foundry landscape, particular to the customers sourcing strategy. And we are -- also found the customers have decided to diversify their foundry sourcing strategy, and started the business explorations, maybe not only with UMC, but the -- as it is we see activity is being associated with that. In due course, we will strive to capture those opportunities, if we could. And as long as it will be subject to our capacity availability, as well as our customers alignment based on their wafer requirements. So, the current priority for ours at this time is still serving our existing customers first, because as you know we are running very full. And it's our job to supporting our current customer and strengthen that partnership, and continue enhancing our market position based on available technology and capacity is also our goal, but our current priority is serving our existing customers first.

Okay. Thanks. Yeah. And also related to that, I know your fab is very busy and I understand that your pricing for 2020 is already set, except for some incremental demand. But we have kind of a new start over for the 2021 business discussion, do you plan to kind of revise of your wafer price to all the customers, so there is no discrimination just that because the -- it's all about demand/supply and you want to kind of hike the price to all your customers seem to in 2021.

Well, the -- yes, I mean -- we -- I mentioned earlier we are seeing, and we have a -- the landscape has changed -- the landscape has changed it. And we particularly in 8-inch advanced node area, so the answer is yes. The -- for the 8-inch business, we are settling those 2021 pricing now and there is a baseline adjustment, OK.

Okay. Okay.

Yes, there is. And the -- but we're still limiting at the 8-inch business. The 12-inch ASP will remain firm, based on the normal pricing skin, so.

Okay.

Yeah. So we haven't making any adjustments on the 12-inch standpoint, but mainly on the 8-inch side.

Okay. Sorry, I missed that. So, can you quantify the magnitude of the baseline pricing change for 8-inch?

The message here is because the -- we always look at it on a blended ASP standpoint, so it's very hard to quantify that. Even we quantify that, you can't imply what's the end result. The overall blended ASP improvement was still subject to the actual product mix.

Right.

The -- based on the loading. For example, if we have better 8-inch productivity improvement, I mean, if we are shipping higher percentage of 8-inch wafers versus the 12-inch the -- despite the baseline 8-inch prices increase that we're still going to drag down the overall blended ASP. So, it's difficult for us to quantify that. It's better to referring to on the quarter-to-quarter guidance because we will provide the ASP guidance on the quarter-to-quarter basis.

Okay. And lastly -- thanks for that. Lastly, I think it's a really good move that you settled with the US Department of the Justice. So, can I confirm this is already settled and there is no longer relative compensation expense afterwards? And also, between you and Micron, do you think there is also a lawsuit to settle or are you think this also already complete along with this US DOJ settlements? Thank you.

The criminal case with DOJ is settled and $50 million fine is reflected in the third quarter financial statement. And for the civil case with Micron, that's still ongoing. And we don't speculate the result, but we will try our very best to defend our shareholder, our Company's right and shareholders' rights.

Okay. Thanks. I think there is a great strategy that big overhang is now almost removed. So, congratulations about the trends. Those were my questions. I will get back to the queue. Thank you.

Thank you.

And the next question is coming from Roland Shu, Citigroup. Go ahead, please.

Yes. Congrats for the very good results. And my first question is a follow-up 28-nanometer. So, I look at your 28-nanometer, I believe, in 3Q your capacity for 28-nanometer should be much bigger than your capacity in 4Q '16, which was the peak of your 28-nanometer revenue. And also you comment in 3Q, you loaded your 28-nanometer capacity at above 90% utilization. So, I think the 28-nanometer wafer shipment in 3Q this year should be much bigger than 4Q 2016. However, I look at the revenue point of view, in 3Q, your 28-nanometer revenue probably was on the 83% of same period in 4Q '15. So, is this on your -- the more wafer shipment? However, with based on your much lower wafer revenue are coming from the product mix change? And can you give us more color on what exactly have changed in the product mix on 28-nanometer? Thank you.

Well, first of all, five years is a very long time for pricing trend in the foundry business. So if we compare to all the electronics products, five years ago, I think, it will be very, very different. So it's not only foundry. We just don't compare these kind of issues with a five-year interval. And certainly, the products are also different compared to five years ago. So, I think it's difficult way to even try to compare to 2015. I think all we're trying to do here is to fully utilize our 28-nanometer capacity and moving to a high-end as much as we can. And we're also diversifying different clientele for our 28-nanometer and we are moving our 28 to 22 as well to increase and enhance competitiveness for both ourselves and our customers. So, we're pretty happy with the progress we made for our 28, and it has been within our plan. And for the next coming quarter, we do expect 28-nanometer as a percentage of revenue to show meaningful increase quarter-over-quarter.

Okay. And then you said that -- and obviously the 28 capacity will be reached 34,000 wafer per month and this actually is above the economic scale for 28-nanometer. So, are you expecting your 28-nanometer gross margin to reach corporate average in 3Q or in 4Q, any timeframe?

Again, we don't really have detailed breakdown by geometries. But, yes, when 28-nanometer is at nearly 95% capacity utilization rate, we do see a similar corporate gross margin compared to the rest of the 12-inch node. But we also are spending, we have put a new CapEx in our Xiamen fab moving from 18,000 wafers right now to 25,000 by mid of 2021. All the newly increased CapEx and also depreciation will certainly put pressures on the gross margin for our Xiamen fab. So, our Xiamen fab, even though we have seen performance being improved compared to year 2019 by a good margin, still they are in red in terms of margins. So, overall, if you plan with the new CapEx and new depreciation, 12-inch -- 28-nanometer is now going to be at the corporate average, because of the Xiamen fab.

Understood. Thank you. And my last question, as for you lawsuit settlement, you -- I think that in three months ago, before you believe there was no round two for UMC for this case, but now the meet is a laser one, secret breach. So can you elaborate for the change put in? Why previously you saw laser -- there was no wrongdoings and -- but now you think there is one set of secret breach? Thank you.

It's very clearly stated in our news release that under US law, Company has to be responsible for employee's behavior and employee did break the law by -- did violate the Company policy and also performed without the knowledge of top management. So that's where the one come from and that's different scenario under the Taiwan law.

Understood. So this is a mainly you pay this $50 million mainly for employees both or in place around the way?

It's very clearly stated in our news release.

So, it's our position to pleading guilty means UMC recognize and accepting the liability of the action of the certain employees.

Okay. Understood. Okay. Thank you. Yeah.

And the next question is coming from Nicolas Baratte, Macquarie. Go ahead, please.

Yes. Hello. Boring questions somehow for Chitung. So you've explained in lot of detail, thank you very much, the other operating income and expenses, which go in the TWD2.8 billion, right? On the non-operating side, you also have a pretty, pretty big number there a TWD2 billion in 3Q. And sorry, maybe I didn't hear what you said about tax, but you also have a very low tax rate in the third quarter.

So for the non-operated -- non -- other operating income, we sell off the solar power subsidiary Nexpower facility for a gain of about TWD1 billion, so that's included in the TWD2.8 billion number for the other operating income. But the non-operating income is mainly the valuation gain from a strong capital market, which UMC recognized about TWD3.9 billion gain from the financial margin -- capital market gains, but offset by about $50 million fine we paid to DOJ.

So the $60 million fine is in non-operating costs?

Yes, that's right. Non-operating.

Okay.

Yeah.

Okay. Understood. Thank you very much. And the tax rate, Chitung?

Thank you.

Tax rate?

And what about the tax rate?

Tax rate is a little bit complicated. We do enjoy certain tax rate for some of our overseas operations. So blended tax rate is probably less than 5% to 10% this year and right now it's -- actually, it's lower for the third quarter, but overall for the whole year I do expect to see about 5% tax rate for 2020. But it will eventually be somewhere between 10% to 15%, should the tax rate or [tax rate] gradually mature.

Okay. Another two questions, if you don't mind. Number one is, what do you expect is the impact of the Japan fab on margins and OpEx for next year? Does it tend to lower margin or increase margin, increase OpEx, lower OpEx, what should we think about here?

Japan fab in terms of impact for OpEx will be stable for the next few quarters.

In percentage?

However, as our President just mentioned, one of the major customers is having a more deteriorating business outlook for the one of the high end products. So they are in the recovery mode right now. So Q4 will be the trough of their loading, also their margin will be below corporate average in Q4 and gradually to pick up over the next couple of quarters.

Understood. If you don't mind, a last question about 28 nanometers looks like everybody is very interested in 28-nanometer. Revenue has been increasing, the 28 node revenue in U.S. dollar has been increasing. In the first half of the year it was 25% and in 3Q it's 75% year-on-year, right? And Q-on-Q, it's also very big increase of 40%, right? So how many new products or new customer or new stuff do you have in 28-nanometer from 3Q '20?

There is a no specific numbers here, but yeah, in Q3, our 28-nanometer from the revenue standpoint, we are experiencing many new products following wireless applications as well as the computing-related products and the -- we see more and more diversification on our product line in the 28 nanometers, in which we are glad to see, and this momentum actually will continue. It will continue into Q4. We will see the 28-nanometer contribution will continue to grow -- the quarter earlier as well. So yeah, I mean, we don't have a specific product count, but there are many products ramping starting from Q3 already.

Thank you.

Sure.

And the next question is coming from Sunny Lin, UBS. Go ahead please.

Hi, Jason, Chitung, Michael. Thank you for taking my questions, and congrats on the very good result. So I have two questions. Number one is on your 8-inch. Management has been reiterating to focus on improving the technology capability for mature process and therefore the product mix. So now that your 8-inch is basically full and I think you could be a bit more selective, so wondering how would your mix evolve in the next two to three years? And my second question is regarding row wafers. I think the expectation by the industry is that 2021 should be a better year with end market recovery. And considering that row wafer had been under supply in the past, I just wonder how would you do differently this time around in buying row wafers? That's all from me. Thank you.

Sure. So in terms of the 8-inch, from the product mix and the mix improvement standpoint, that is the continuous effort, especially with the 8-inch advanced node area, 8-inch advanced node anything under 0.18 micron. And so anything under that we'll consider the advance. And we see many applications coming into the 8-inch advanced area. So we see product mix improvement effort is continued. For anything that's greater than 0.18 is -- behave a little bit different. There are different applications coming into that, but I won't consider that actually as a better mix, because of many of the mature 8-inch actually migrating to the advanced 8-inch. So the key focus on the mature 8-inch is two-fold. One is we try to upgrading our facility. We can convert sound the mature technology capacity into the advanced node and to serving the high demand of the advanced 8-inch area. The second is by developing a new technology to adapting the new applications, okay? So from a mix standpoint, I think the mature side will stay pretty flat and goal is try to continue fully blow in that area versus the advanced 8-inch will continue to improve the mix and to also enhance the blended ASP. So that's sort of respond to your 8-inch question. For the wave substrate materials, after the loss experience, we have examined our relationship with our suppliers. So now that we have a long-term contract arrangement in place now and for the old suppliers, and we continue to have an ongoing discussion with those suppliers as the same time that we tried to qualifying multiple source on different product lines. So, by doing that, we'll actually be able to secure our supply with more of a competitive solutions. So that is quite -- adapt after last experience that we have. And so far we feel comfortable about that, but the result that we'll have, remain to be seen.

Got it. A very quick follow-up. So would it be fair to assume that for 2021 modes of your long-term contract should have been finalized and with maybe reasonable pricing?

I will say so. The founder the institute's founder, there are some already settled and there are some still under negotiation, but yes, I think it's interesting, [indiscernible] it meets our expectation at this time, yes.

Got it. Thank you very much. That's very helpful.

And the next question is coming from Sebastian Hou, CLSA. Go ahead please.

Thanks for taking my questions. The first is to follow on the 28 nanometers. So I think Chitung early mentioned about we should expect the 28 nanometers could be -- could drive the margin higher from here. So can you explain in further details about what could drive the 28 nanometers profit higher from here given that it's really run in the mid-90s or 90% plus utilization rate? Thank you.

Mainly revenues from our Xiamen fab currently had rather big loss and the loss has been reduced year-over-year and we do expect to see the structure profitability for Xiamen fab to be meaningfully improve after they reached the 25,000 economy of scale operations, which you saw by the mid of 2021. So that's a smaller loss coming from Xiamen will be a way of 28-nanometer margin enhancement. And of course, we will continue to see upgrade from 28 to 22 and also better product mix for 28, that will also help.

Got it. So your Xiamen fab right now is 17,000, is that right?

About 18 now already.

18. So for 18 to 25, it can make a big difference in terms of economy of scale?

After we reach 25, the full size, we will see potential to see even further cut in terms of loss.

Okay, great. Thank you. Second question is on the inventory. I think some of your peers are producing given the inventory situation on your customers' side, although there has been sales above average for multiple quarters already. I'm curious about how you review on the inventory situation to customers? And how do you see -- what's your view on this everlasting higher inventory situation for -- could be the new norm? Thank you.

Well, I mean, first of all, you're right. I mean -- and also from other observation that you say the inventory was at a high level. We are what we think we have seen quarter-over-quarter decline in inventory given the end market has started to recover. But the data -- yes, the data this show the inventory level at this time still at a higher than the normal level. So despite that we see some quarter declining, but the current inventory level remains at a higher level. Now whether if this situation is -- will become a new norm, that remains to be seen. We believe in the higher level, but we see because the higher content of some of the higher silicon compounds on the devices and because the pandemic situation whether this high level of inventory can be a new norm. That is a possibility. It's still too early to judge at this time, but we will continue monitoring that progress.

Okay. Thank you. One follow-up on that is given the capacity is pretty tight right now everywhere, leading edge foundry and also showing edge and everywhere. So when you communicate with your customers, do you see any gap between your customers' ordering behavior versus your internal forecast on the demand, i.e., whether there could be some overbooking behavior because your customers are just afraid of not getting enough capacity?

Yeah. I mean, this conversation -- this type of the conversation is happening on a daily basis. The interesting thing is, if you look at the inventory, it may not be within our customers' space. It could be in the southern -- it could lay -- somewhere in the supply chain. It could be at a system level or electronics level, but not necessarily on the silicon level. So we have been constantly checking with our customer and they're checking with their customer and this type of conversation is ongoing. People are concerned about this, whether there is a double booking situation. But at this point, I would say, people that we've been talking to as well as ourselves, we -- know one can judge whether this is a new norm or this is going to be an inventory correction coming up soon because just -- we have to continue seeing those things through. Now there are some precaution measures that we can do. Hopefully we're putting more resources on the area that really are in shortage. And so maybe have priority business than others. And again it's a delicate process, it's not as a straightforward. But I mean, we -- I think, not just us, the entire supply chain find it best, try to mitigate the situation here. And we just have to -- we have to sit and see what's going to happen. But I mean, right now, I mean the demand coming up on customer remains very solid and very strong. And we're doing our due diligence and hopefully we're making a good decision here.

Got it. Got it. Very fair comments, and thank you for that, Jason. That's all from me.

Thank you.

Ladies and gentlemen, we're running out of time. So we're taking the last one. And the last question is coming from Bruce of Goldman Sachs. Go ahead please.

Hi. Thank you for taking my question again. I want to double-check my understanding about the gross margin for the forex impact. So for the fourth quarter, the ASP is going up by 1% to 2%, instrument is 1% or 2%. So the total revenue is like 2% to 3%. But the forex eat it up on the gross margin side. So does that mean that gross margin impact is at 2% to 3%? Is my understanding correct?

2% to 3% on the top-line. And every 1% increase of NT dollar against U.S. dollar eat up about 0.5 percentage point of our gross margin.

So that's about 1% to 1.5% gross margin impact in the fourth quarter?

I'm glad you said that.

And by the way, the ASP increase was called at 1% not 1% to 2%.

Okay. Thank you. The follow-up -- a quick follow-up is that for management keep on saying that the CapEx will consider the ROE or ROIC for the shareholder, which is great for the investors. However, can we have like a longer term capital intensity guidance such as like, for example, like 20%, which is a reasonable target. Can we expect that or our CapEx were maintained at the lower than the depreciation, which is around about $1.5 billion a year, something like that?

No, we don't commit something like that. I mean, we really don't see how much money we can afford to spend. We don't look CapEx at that way. It's really on what kind of return we can achieve and how much CapEx is needed in order to trigger that kind of project. And also, our President just mentioned, we are looking at opportunities both organically and inorganically. So it's more complicated for UMC's case compared to the other foundry, which probably only relies on mostly on the organic growth. So we will look at our combined basis and to see what kind of capacity expansion is the best solution for our customer, as well as to our shareholders.

Maybe I can add a couple here. Three years ago, we actually adopt the strategy that first, our first priority is reshaping our financial structure. And so our goal is tied to improve our financial structure first. And then the second strategy was focused on cost effective capacity expansion, that is the combination of the made and buy, which is what we're doing right now. I think we at a level that we were going through the first phase and we feel comfortable with where we at right now and we're looking into the second phase. And the ultimate goal is we try to striking a balance between the shareholder interest, as well as the growth of the company. So your answers are legit, but the question is we are looking at all this on a holistic view and maybe at the time that mature and we'd be able to put you maybe there is a reference in terms of the CapEx budget or affordability level. So -- but we're still going through that so-called Phase 1 and we at -- we're reaching probably a late stage of the Phase 1 and we continue accelerating our second strategy. Hope this helped too.

Yeah. that helps. I think the investors fully appreciate that the return profile improvement for the last two, three years. I think the one of the major concerns from investor, which I got is that people worry about that UMC might trigger a major -- massive CapEx for the 28, because if you want to increase your 28 by 10 or 20,000 whether per month, that's like TWD2 billion to TWD3 billion. So I mean that's one-off or one major CapEx hurdle which is the concern right now. So I want to see whether we have some upper limit or some kind of like guidance?

At this point, you don't need to worry about that and it's been -- we have been pretty good disciplines here tied to manage cash flow. We're going to stay that way. If there is any changeover, we should talk, but yeah, not at this point.

Thank you.

Sure.

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

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 ir@umc.com. Have a good day. Thank you.

Thank you, and ladies and gentlemen that concludes our conference for third quarter 2020. We thank you for your participation in UMC’s conference. There will be a webcast replay within an hour, please visit www.umc.com or the Investors event section. You may now disconnect. Goodbye.