Welcome everyone to UMC’s 2022 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 2 hours after the conference has finished. Please visit our website www.umc.com under the Investor Relations, Investors Events section. Now I would like to introduce Mr. Michael Lin, Head of Investor Relations at UMC. Mr. Lin, please begin.

Thank you, and welcome to UMC’s Conference Call for the Third Quarter of 2022. I’m joined by Mr. Jason Wang, the President of UMC; and Mr. Chi-Tung 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 fourth quarter 2022 guidance. Once our President and 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 a more detailed discussion of these risks and uncertainties, please refer to our recent and subsequent filings with the SEC and the ROC security authorities. During this conference you may view our financial presentation material, which is being broadcast live through the Internet. Now I would like to introduce UMC’s CFO, Mr. Chi-Tung Liu, to discuss UMC’s third quarter 2022 financial results.

Thank you, Michael. I would like to go through the 3Q 2022 investor conference presentation material, which can be downloaded for view in real time from our website. Starting on Page 4. The third quarter of 2022, consolidated revenue was TWD 75.39 billion, with gross margin at 47.3%. Net income attributable to the stockholder of the parent was TWD 27 billion and the earnings per ordinary shares were TWD 2.19. Utilization rate in the third quarter remained at 100% plus. And our overall wafer shipment and also ASP are pretty much in line with our previous quarter’s guidance. Next page, please. So for the third quarter, revenue grew by 4.6% sequentially to TWD 75.39 billion. Gross margin rate of 47% or in absolute dollar -- NT dollar terms is TWD 35.66 billion. And operating income margin reached 40% compared to 39.1% in the previous quarter. And net income attributable to the shareholder of the parent was TWD 26.99 billion or equivalent of EPS TWD 2.19. This is a meaningful growth compared to TWD 1.74 EPS in the previous quarter. For the first 3 quarters, the cumulative revenue grow by 37%, thanks to the help from shipment increase as well as ASP enhancement and also the NT dollar depreciation, all 3 factors play important roles. And for the gross margin rate for the first 3 quarters was around 45.8% or TWD 96.6 billion. Operating income surpassed TWD 80 billion for the first 3 quarters, which is a 38.3% operating income margin rate compared to 22.1% in the same period of 2021. For the net income for the shareholder of the parent is TWD 68.131 or 32.3%, and that’s a 71% year-over-year growth and EPS for the first 3 quarters reached TWD 5.54 per shares. So we pay out $3 per share…

Thank you, Chi-Tung. Good evening, everyone. Here, I would like to share UMC’s third quarter highlights. In Q3, our results benefited from product mix optimization and a more favorable exchange rate, while fab capacity remains fully utilized. Despite softening demand in consumer end markets, strength in certain wireless communication area drove further expansion in our 22/28 nanometer business, which accounted for 25% of overall third quarter revenue and lift wafer average selling price. We believe our industry-leading position in OLED, OLED display driver IC will continue to drive growth for our 22/28 nanometer technologies amid growing adoption of OLED panels in smartphones and other end devices. We also saw sustained momentum in our automotive business during the quarter, and we intend to pursue more collaboration opportunities with existing and potential automotive customers. Moving into the fourth quarter. We expect to face headwinds amid demand weakness impacted by factors including the inflationary environment and Ukraine war. While UMC will not be immune to the inventory correction affecting the industry, we will work closely with our customers as they adjust elevated inventories to align with current market conditions. At the same time, we will continue to deliver differentiated technology processes to enable customers’ product pipeline. We have revised the company’s 2022 capital expenditures down to USD 3 billion. Our capacity expansion in Tainan and Singapore are still progressing as planned in order to meet long-term supply commitments. Despite near-term turbulence, the structural story of increasing silicon content driven by the rise of 5G, AIoT, and ED remains intact. With our comprehensive technology offering focused on manufacturing excellence and resilient financial structure, UMC will further increase our exposure to strong growth markets and consolidate our specialty technology leadership. Now let’s move on to the fourth quarter 2022 guidance. Our wafer shipment will decrease by approximately 10%. ASP in U.S. dollar will remain flat. Gross profit margin will be in the low 40% range. Capacity utilization rate will be at 90%. Our 2022 cash-based CapEx will be revised to USD 3 billion. That concludes my comments. Thank you all for your attention. Now we are ready for questions.

[Operator Instructions] And our first question is coming from Randy Abrams, Credit Suisse.

I wanted to ask the first question just about the -- you’re guiding a somewhat mild or moderate correction in the outlook fourth quarter. Could you discuss if it’s still largely these consumer areas like PC, Android and the digital TV? Or is that extending now if you’re seeing any broadening out to auto industrial networking or also the IDMs, which looked like they had a good quarter in third quarter?

Okay. Well, for the Q4, the auto and industrial segment remains stable for now. While the inventory level is still healthy for the auto and industrial segment in Q4, we have observed that PC and smartphone are still undergoing a prolonged inventory correction compared to the previous quarter’s estimate, which indicates there’s a no tangible sign of a recovery in the near term. We do foresee that downward trend for the consumer segment will continue to linger into first half 2023 due to a slow inventory digestion. And it’s our belief the bottom of this down cycle will mainly determine when the end system market starts to resume its momentum. And at this point, the visibility is still low. But in the Q4, we are able to mitigate our 8-inch fab loading to remain 100% in Q3 after swapping some allocation from notebook, TV, smartphone to auto and AIoT. However, the lighter 8-inch utilization will happen in Q4 and mainly because impacted by the deteriorating market condition in the smartphone and PC for -- because of the inventory correction. Randy?

I’m here. Sorry, they removed me from the queue. Sorry, no, I’m back in. Okay. Yes. No, I wanted to ask my second question. Actually, taking that into consideration, is there a rough way to think, given that what you were saying on consumer and you could see a little bit of broadening where you see utilization trending toward first half. And I’m curious at this stage, 2023 now I think originally you were thinking a growth industry and UMC would grow. If you have an initial view for ‘23 industry and the UMC outlook?

Sure. Well, I mean, we -- after the quarter from last quarter, given the rising macro uncertainties and ongoing inventory correction, like I said, in the PC and the smartphone space and as well as the high inflationary cost pressure and we do foresee the semi and foundry industry will decline in 2023 now. But that’s off a very high base for 2022. And for UMC, the -- we -- while the visibility in 2023 is low, we still expect our 12 ASP [indiscernible] will come on stream by mid-2023. And with our ASP outlook is expected to remain firm as the management’s goal is to grow in line or outperform with UMC’s addressable market. I kind of want to emphasize that it’s within our addressable area. Yes.

Okay. Just -- so do you expect your addressable market -- I’m just trying to understand that versus the industry down, do you think your addressable market is growing next year?

I think given the current condition, it will be very challenging to see that up. I think it will probably decline in 2023.

Okay. Understood.

And I think we should be able to give you much more clarity for another quarter, yes, in January, yes.

Okay. No, that’s helpful. And if I could ask on the -- 2 parts on the CapEx. One is that -- if there was a timing the $3.6 billion coming to $3 billion, as it was tracking a bit behind through the year, were you pushing out certain projects? And if there’s a view then, would that be additive or a view on next year? And the other part of the question, just on that P6. Is that still -- it had LTAs tied to it. Are those LTAs still covered? Or are you seeing customers trying to reschedule or some change where some of that capacity in this environment may not be utilized?

Okay. So for the first -- the 2022 CapEx cut, there are 2 reasons that led to the cut for the cash-based CapEx. First, it was primarily due to the equipment vendor delayed, all right? So -- and the second is given the current downturn, we are reprioritize some of the projects to respond to the current cyclicality. Meanwhile, we’re still keeping our commitment to our P6 LTA customers. And the question for the -- is there any change in the LTA customer for P6? Right now the -- both customers and us are taking serious and commitment to this relationship, so commitments and partnerships. So we have not seen any significant changes in this space. So the LTA has some impact to us, but is more in the near term, now with the P6.

Okay. Just clarify that. So some impact on the near term, not P6, what’s happening to the other LTAs? And then I’ll get back in the queue.

Sure. The -- I mean, we do see -- while this going through this market turbulence and obviously, not all our customers immune from it. And while this inventory correction continued to worsen and our customers are trading this LTA, while they’re trading it seriously, but they are reprioritized on their product mix and reflect the end markets as well as minimize the -- in order to minimize some of the obligation to avoid some of the LTA penalties. And so -- but as a result, there are some LTA penalties already happening, but it’s very insignificant. And we don’t want to get into specific on the LTAs, but at this point, all subject to business are still operating compliant to the LTA terms.

Next question, Gokul Hariharan, JP Morgan.

First of all, could you talk a little bit about what we are seeing, I think you did call out 8-inch as one area where there is some utilization slack. Could we talk about 28-nanometer and the rest of the 12-inch portfolio looks like 20-nanometer still remains pretty solid in terms of utilization? Are we seeing slack in other parts like 40, 65, et cetera, in Q4 and as we look into Q1? And how do we expect utilization to trend in first half? Obviously, not very clear at this point, but are we looking at utilization going all the way back to like 80% or even below 80% as we have seen in the past cycles? That’s my first question.

Okay. Well, actually, this is a big question. Let me see if I can -- I got older. But first of all, as we touched the 8-inch and your question is about 28. If I may, I’d like to kind of go back to about the segment first and before I get to that. In Q3, we actually see the automotive revenue growth into double digits. And the consumer grew low single digits. Communication grew low single digit, while the computing has declined about high single digit. And while we’re swapping those, like I mentioned earlier, swapping between the auto, industrial and PC and smartphone, I mentioned earlier, we’ll be able to manage that full-loading. Now going into the Q4, we see all segments are declining, except automotive. So on a sequential point of view, the automotive, we do see a continuous double-digit growth and while the others start seeing a declining. And I think that that happened to have impact to our 8-inch loading because PC and smartphone is majorly impacting the 8-inch. We -- however, the lighter 8-inch utilization in Q4, I mentioned, it was impacted by the continuous mobile and TV inventory correction and our 12-inch product portfolio come with a more higher differentiation with the leading specialty technology as well as some of the diversified market segment and sub-segment. And therefore, it’s less vulnerable at this point compared to the 8-inch in Q4. Okay. While we see this correction lingering into first half and we have to -- we probably have to report the Q1 when the time comes. Now we know the 12-inch is actually much healthier than the 8-inch in Q4. For the 28 specifically, is part of that. And with our resilience on 28 mainly come from the robust demand from our targeted segment such as the OLED driver and automotive space. So at this point, the 28 remains healthy at least for Q4. And if we look at the 28 outlook, for the long term, we still believe 28 and 22 will be a long-lasting note, driven by the application such as ISP, OLED driver IC, where this demand will continue to grow. And we believe we’re well-positioned with those diversified product portfolio to capture this market opportunity. And so that’s why we actually remain confident with our 28 and 22 business, mainly because they continuously strengthening our solutions and customer profile. And I guess I answered the 28, so the color between the 8-inch and 28, I’m not sure if I captured all your questions.

Yes. That’s very clear. One part is on utilization. I think you talked about 90% utilization in Q4. Should we expect utilizations to go all the way back down to 80% or below like previous downturns? Or you have some offsets in first half, which could kind of protect some of the utilization?

Yes. At this point, we won’t be able to answer a question like this. I mean, as we mentioned earlier, we’ll still not be able to pinpoint where is the trough of the cycle or where is the end of the inventory correction. So we will only be able to give you guidance on quarter-on-quarter basis.

Got it. My second question is on pricing. I think if I go back and look at pricing, blended pricing, especially, in every downturn, we see some degree of price compression. Magnitude is different based on each downturn. But if you look at the last 5, 6 downturns, we’ve already seen price compression on a blended basis. Now this time around, you have been a lot more resilient in your communication about pricing to the market as well as customers. Could you talk a little bit about what is your pricing strategy as we go into the downturn, especially as you mentioned, we still don’t see the trough in terms of where this inventory cycle could drop out? So if your pricing strategy is going to be dependent on how utilization shapes up or you would basically take a pricing strategy such that even if utilization is much weaker because the downturn is longer, pricing is something you would not be very variable on?

Okay. Well, I mean the foundry is a cyclical industry. So there is a couple of layers answer to that. One is the ASP position. Our strategy and position on ASP is the ASP will reflect both product mix and pricing adjustments. While we continue to improve our product mix, our pricing strategy is to reflect our value proposition on technology leadership and differentiation. We do value longer-term partnership over the near-term cyclical factors. We believe the pricing reflect our market value and proposition. And so we would like to continue to strengthen our relationship with the customer on a long-term basis and mutually growing with our customers. So that’s our view in terms of ASP. However, if for the short term, because the cyclicality reasons between the ASP business and loading or utilization, we do examine that. Under the recent market conditions, we believe the trade-off between the loading and price will end up with very limited benefits in demand creation because the weakness in sell-through and demand. So we -- unless we see a significant benefit that I don’t think that will be much change on that. At this current point, for our pricing in 2023, we expect it to be firm, but we closely monitor it, yes, if there is a benefit to it.

Next question, Laura Chen of Citigroup.

My question is about the recent intensified geopolitical risk. I’m just wondering what the implication to UMC, do you potentially see more opportunity to further gain share at the expense of your Chinese peers? Or you are actually seeing more intensified competition in your Chinese domestic market? That’s my first question.

Sure. Well, with the recent U.S. export control and regulation, the saving this month and the impact on UMC has been limited as the restriction mainly target more advanced nodes that are not part of the UMC’s addressable market. We are in close communication with the suppliers and customers as they navigate the new rule of schools announced earlier this month. Our current focus is focused on compliance as we believe the compliance is one of our key business principle and we will continue to monitor the development of the U.S. export control policies closely and take risk management measure as necessary or even evaluate if there’s any opportunity. But at this time, I think it’s still premature and we’re closely monitoring the progress. Yes.

Okay. So you haven’t really seen any movement in terms of the clients’ order flow because of this event?

Well, I mean, they certainly are talk and -- but I don’t think there’s at a stage of a movement yet. No.

Okay. Great. And also, my second question is about the IDM outsourcing trends. You also mentioned that the auto demand is still quite strong at least for this quarter. But we know that like Texas Instruments just reported this morning, they also see some weakness happening aside from the auto application. Since those IDMs, they also build up their own capacity. So I’m just wondering what’s your view on those IDM outsourcing trend? Is that sustainable in particular, we see a lot of uncertainties ahead into next year?

Well, first of all, the key word is adjust. You mentioned that the U.S. customer has just announced their view in terms of the market. And obviously, is just happening. But in our view, the auto and industrial segment remains stable for now that reflected in our Q3 because they’d be able to offset some of the weakness in other segments. And while we’re going into the end of the year as well as Q1, we remain cautious and we were definitely working closely with our customer in terms of outlook. Meanwhile, for the auto and industrial, is a -- for the longer-term perspective, is our target segment and the focus. And we will continue expanding our auto business from both IDM customer and the fabless customer through both technology innovation and capacity expansion alignment. Because IDM customers tend to have their own capacity plan, so it’s important that we put in the capacity expansion alignment as a part of our ongoing discussion with them.

Next question is Sunny Lin of UBS.

My first question is on 28-nanometer. I understand your expansions are all based on the steady long-term demand outlook. But I wonder what kind of factors could potentially trigger you to reconsider the extension timing and scale? And perhaps because of LTAs, do you still have some flexibilities in terms of the adjustments?

Well, I mean like I said, for the longer term outlook, we do believe this 28 or 22 -- and 22 will be a long-lasting node. And so we actually feel comfortable with our perspective on this. And our -- the reason market dynamics does not change our long-term view and as well our long-term relevance in this -- with the customers. And we will continue pursue this. And as we believe it is a sweet spot for many applications. And while this demand will continue to grow. And we are confident with our solution offering that our 28 business can be maintained at a healthy level in 2023 and beyond. So for the LTA, I kind of touched that earlier already for the P6. But maybe I can give a bit of our view on the LTA in general. The LTA may not be able to mitigating all shortfall because the market volatility. But with LTA, the customer will typically reprioritize their product mix, even reallocating their multiple sorting product to UMC to fulfill their obligations. And we actually still believe this LTA help. And at this current downturn cycle, we not just look at the P6 or 28 LTA. We’re actually taking this opportunity to re-exam our LTA selection criteria to make sure that we align with the megatrends and other growing markets as we believe the LTA contract could help to mitigate downturn cycle and demonstrate the mutual commitment for -- from both customer and UMC. And so that’s kind of -- I’d like to share that with you in fact going to so-called P6 LTA situation. Yes.

Got it. That’s very helpful. So a quick follow-up on your P6 expansion. I think in July, you expected 2023 capacity increase for the total company could be about 5%. Now with lower CapEx reduction, I wonder what’s your latest guidance for next year’s capacity increase?

Well, at this point, the number is slightly below 5%. So it’s still very close to 5%. And the initial ramp was kind of still started from mid-2023, but you will reach to 12,000 a month by the end of 2023. So the ramp profile got changed a little bit.

Got it. So 12,000 per month by end of 2023, so maybe approaching full scale by late 2024?

Well, I mean, for 2024, we still have to -- that’s still in progress right now. We react to the current market downturn and start adjusting our 2022 spending and which reflects the 2023 ramp. And at the same time, we will look into 2023 CapEx as well. But at this point, it’s kind of too early to report the 2023 CapEx as well as the end profile, yes.

Got it. My second question is to follow up on LTAs. And so one part is that if a customer has LTA at both existing capacities and also your new capacities, would there be a case that they trim the production at existing capacity, but remain committed to the new supply, in that case, how would you mitigate the impact on your existing capacity?

Well, I mean our commitment to the LTA customers impact decide the ramp profile change on the new capacity, but our commitment to it is still there. So we have to manage the base capacity to support them, just to make sure that we fulfill our obligations. And at the same time, we will continue to align with all customers with their outlook to make sure that we don’t cause any surprises for both LTA customers as well as the non-LTA customer.

And also there is a pricing differences between the existing capacity and newly built capacity, even though they both LTA or that the pricing are different.

That’s right. And so the other part of my LTA question is that I think in recent weeks, some fabless to talk about allowing customers to extend the duration of the LTAs. Would that be a practice that you also consider?

Well, I mean we always entertain customers’ requests. And -- but in principle is we have to treat those LTA contracts seriously. And so we will provide certain flexibility, but not to the extent you mentioned about it. So at this point, as I mentioned, the customer is still supporting us with try to reprioritize their product mix, reallocating some of their multiple source products to UMC. If they are able to fulfill the 100% of their obligation and then we will comply to the LTA terms.

Next question, Bruce Lu of Goldman Sachs.

I think you just mentioned that your pricing [indiscernible] and you do the exercise to exam the utilization in the ASP erosion. So which means that given the current 90% utilization rate, you have no plan to change your pricing or for the first quarter next year, most likely is going further south and you have no plan to change your pricing assumption as well. So can you tell us what is the special or help us to go through the exam you did and make sure we have the right understanding if we are more negative or more positive for the outlook for 2023?

I mean, right now from a end market standpoint, our view of 2023 is the market will decline. So it will be a challenging year. But from a ASP standpoint, I think there’s a limited trade-off between the loading and pricing. So -- and mainly because the weakness in the sell-through, the end market. If there is no elasticity of the ASP and with the volume and so the benefit will be -- is hard to capture. So that’s why we think the first criteria is we have to understand the end market momentum. And then we’ll talk about the market share consideration, so on and so forth. So they are different condition and criteria to exam. And -- but at this point, we’re kind of -- we’re finding difficulty even we pass the first layers because we haven’t seen any benefit on that. And we do believe our pricing reflects our relevance with the entire supply chain. And so we just have to continue look at the longer-term perspective and make sure that we position ourselves to capture the opportunity, not just looking at the short-term tactics. And -- but if the short-term tactics doesn’t help, I mean, it’s a option we can take, but not at this current point.

So can we assume that even the capacity [indiscernible] going down to 80% or even 75%, the pricing strategy will remain unchanged?

I mean, I think it’s not gauged by the utilization. It’s gauged by if the pricing will trigger the higher utilization results.

I see. Understand. Okay. The second thing is that I think that for your 20-nanometer, the revenue is much more resilient than your other nodes given your product and customers’ choice or new strategy or better execution. However, your driver IC revenue exposure is still high, your consumer product exposure is still high, which creates a lot of fluctuation for your business. So what’s your plan and strategy to change that? And how can we -- how or when we can see some improvement like your other node can be as resilient as 28?

Well, I mean, it’s our goal to continue with the customer and product portfolio enhancement activities. And we’re still in the middle of the process that. And this one is we have to continue to deliver our technology innovation. And in order to -- with the target market and so we can improve the portfolios and we will diversify our customer and market segments. So this effort is continuous. We have been doing that for the past few years, a couple of years, and we see some through of that. But we’re not there yet. I think we’re still in the process to continue with the product and the portfolio enhancement. And we will continue to do that. And by doing that, we believe we will continue to help with that. At the end of the day, it’s hard to immune from the entire market long-term, but with a better product mix and with a more compelling solution to our customer, we do think we can be resilient or even better than the industry’s performance.

Well, can you elaborate a bit more because for 28, it somehow has been more differentiated stuff for like [I4] or other stuff or OLED. But it’s not easy for investors to visualize the like 8-inch, what can you do for the differentiated 8-inch business? Or what can you do for your 65 or 90 nanometer? Can you elaborate a little bit more?

For 8-inch, for instance, on 8-inch, there are structural demands coming from EV power management, even for the migration of DDR5 power management and joint development program with IDM to support their growth in their respective market segment. So there are still some structural demand in some of the areas, and we’re going to continue to pursue those target market space. And for the mature or legacy 12-inch, for instance, 35-40 the MCU, the [indiscernible] memory just come in to 40 now and with the automotive market. And by expanding our capacity into the nonvolatile memory, we think that will actually help the resilient as well. And so there are certain market segments that we believe that has a structural demand going forward. And so we define that as so-called a targeted market. And so the goal is you have to deliver your solution -- compounding solution and meeting the customer demand as well as the capacity support. And hopefully, you can be less vulnerable. I don’t know if we are resilient enough, but I think our goal is try to be less vulnerable, yes.

Next question Szeho Ng, China Renaissance.

I have two questions from my side. First one regarding the revenue breakdown, right? I believe for the others category at 14% in Q3, I think majority is related to automotive. Is it a fair assumption?

That’s correct assumption, yes.

Oh, okay. All right. And maybe down to, let’s say, 3-5 years from now is it fair to assume that percentage would double? Is it a [indiscernible] assumption?

Well, I mean, with -- like the -- Bruce just asked earlier, is our target to reaching greater than 40% by 2025 with that revenue can come from the mega trend. That includes the automotive, yes.

I see. All right. Okay. Second one maybe for Chi-Tung. Regarding the recent approval for us to buy back the remaining 30% of UMC Xiamen, so when would that transaction be closed? And would there be any -- how should we envision regarding the P&L implication on the company?

Yes. UMC or UMC Xiamen is already making profit. And as you can tell from our financial report, there’s a minority interest of positive profit actually that for the 30%-ish outside shareholder of UMC. Once we complete 100%, then we will be able to phase, say, for this minority interest. And our goal, the mandate is to allow UMC to complete the transaction, we see a window of 3 years, but we certainly intend to do so in a faster expedite manner.

Okay. So that means between 3 years or you haven’t decided exact timing, right?

Yes, 3 years the window of the timing, but we won’t wait for 3 years.

Oh, I see. I see. But would that happen next year realistically?

There is still some technical issues and we won’t be able to tell you the exact year, but our intention is work hard on this execution of this buyback.

I see. Got you. And once the transaction is completed, would there be a one-off P&L impact?

No, there won’t be any one-off impact. Just we don’t need to post this minority interest for the outside shareholders.

I see. Anyway a small number right, right now the minority interest to the group. Okay. All right. Okay.

Next question, Charlie Chan of Morgan Stanley.

And still very good results. I mean the market challenging. So my first question is about your 14-nanometer. So Jason, maybe a kind of a long-term question about how much of the 28-nanometer demand today will start to transfer to 14-nanometer domain coming 3 to 5 years? And I think as said before, whether UMC has any advantage for the 14-nanometer FinFET? That’s my first question.

So the question is about 14, right?

Yes, nano.

Yes, 14. Yes.

Well, we currently -- we are investing in growing our 28/22 technologies, like you said and along with the capacity expansion because we believe -- we expect the increase of market potential with increase in customer adoption. The 14 FinFET from technology standpoint, we have entered for a few years ago. But on the capacity expansion plan, our current approach is still aligned to the existing 28-nanometer customer with their product transition. And so we’ll probably will gradually address that. But at this time, we still focus on the 28 and 22 expansion. Expansion, yes.

Got it. Yes. So because it’s also linked to that geopolitical risk issue, right, because now China’s fabrication for 14-nanometer FinFET is restricted, right? So as a company so do you think you should accelerate a little bit your 14-nanometer in preparation for potential demand from China customers if they need a 14-nanometer FinFET foundry service 3 to 5 years later?

Believe me, the 14 is on very top of this. We do have -- we do pay attention to the 14 and it’s a important topic for us. We’re not going to abandon the 14. So -- but we are examining a few things. One is from the priority standpoint and the timing of that. Right now from the -- we founded by certain boundaries for instance, on these ROI-driven policies. We want to make sure the -- when we examine the opportunity for us to invest and where will we invest first? So at this point, we’re putting the 28 ahead of the 14, but it doesn’t mean that will change -- I mean that doesn’t change. So it does come out to -- because the market dynamic changes, 14 become much more attractive than 28, yes, the priority will shift. Yes.

Okay. That makes sense. And a couple of very quick questions. Do you see a similar change that [TSN] is saying that your customers’ inventory are declined this quarter? Or do you think the inventory continues to go up at your customers?

Well, on the PC and the smartphone space, like we report earlier, we see the digestion actually slower than we expected from us previous estimated. So we don’t see that tie up, but we see a very slow digestion because the -- yes, because the inventory buildup can be controlled by its new wafer starts. And -- but for the -- the inventory to be deplete, that requires a sell-through. So I think the challenges that we have seen is we -- is the end market sell-through situation. And we just have to closely watch that. If the sell-to government, I think that the timing of that will probably accelerate. But that remains to be seen at this point, yes, because the overall macroeconomic environment, yes.

Okay. And then switch gears to a financial question to Chi-Tung. The first question is your CapEx cuts. May I know which nodes you’re cutting the demand and anything related to your Singapore new set projects?

No. Mainly as Jason just mentioned, there are 2 reasons. First of all is the P6 equipment delivery, the delivery lead time got long. So it’s a delay for our P6 China profile, but we are trying our commitment to our own LTA P6 customers. And for the current cycle, of course, we are getting more nimble and cautious. And we reprioritized several projects and that’s part of the contribution to $3 billion, I mean $3.6 billion to $3 billion CapEx revision.

Yes. So may I know any specific nodes that you kind of changed the CapEx spending?

This is primarily 28-nanometer, so the ramp profile got the long. So the impact will be mainly on 28-nanometer.

Okay. That’s fine. So Szeho Ng also asked about some key assumptions for gross margin. So for this year, I think you already provided fourth quarter guidance, right? So I think I believe you can comment on the full year base. So on a full year base, in terms of the gross margin impact, how much of that comes from the so-called the cost inflation? And how much benefit coming from the NT dollar depreciation and also your kind of equipment depreciation trend?

Every 1% of NT dollar depreciation may contribute about 0.4% of our absolute gross margin point. And for Q4, we are guiding for a low 40% gross margins. That will be mainly coming from a lower -- 10% lower gross margin. And in terms of depreciation, for 2023, we are seeing about 5%, 6% increase for overall depreciation.

Excuse me, you mean 2022 or 2023?

2023, about 5% increase for absolute depreciation number. For 2022, we are talking about minus -- a decline of 5% to 6% of depreciation expenses.

I see. I see. Yes. So actually, that’s a kind of a basis for next year assumption. So for next year, do you see any kind of inflated cost besides the current depreciation because your industry peers seems to suggest that there are kind of inflated cost in the [indiscernible], wages. I’m wondering how that is going to impact your next year gross margin?

If we set aside the utilities, which we pay higher summer utility price, which is about to over in October. The general inflation-led cost increase, we are probably talking about low single digit, maybe 1% or 2%. But it varies, depends on which component you are talking about. And we are working with our customers constantly try to cope with this issue.

Okay. So to summarize, next year in terms of cost of goods sold, I guess, still some inflation in terms of variable cost and your equipment depreciation will go up by 5% to 6%. But on the currency side, on a year-on-year base, I guess, probably at least a 6% NT dollar depreciation year-on-year. So that’s going to offset those cost increase. Is that how you budget for next year cost of goods sold?

Yes, we don’t do it that way. We do it from a [part and up way]. But I think that your model buildup, so I think most of the parameters I already shared with you.

Next question, Gokul Hariharan, JP Morgan.

I had a quick follow-up. So given that we still have a Singapore fab coming up, how should we think about CapEx direction next year? Now it’s too early to give absolute guidance. But are we still thinking in the same range of $3 billion kind of spending? Or do we see that the spending could come down next year given a bit of a downturn?

Yes. First of all, you’re right. It is too early to disclose the 2023 CapEx number. The full year 2023 CapEx, the cash-based CapEx budget we will update to you in January, the January -- month of January. But it’s very subject to the market outlook. We’re closely monitoring it, and then we’ll report that next time, yes.

Okay. But any directional comments? Is it going to be flat, down, up?

Really difficult right now but we are keeping our commitment to our LTA customers for P6 as well as P2 in Singapore. So the ballpark schedule won’t change, but the equipment, total delay is something out of our control and we are working closely with our customers to -- in order to keep our commitment.

Got it. And one more follow-up on 14-nanometer FinFET. Do any of your LTA commitments for the future fabs include 14-nanometer capacity as well that you need to ramp up? Or there is nothing that covers 14-nanometer in your future LTAs?

The current LTA covers the no technology no migration options. And -- but with the current visibility, that migration is from 28 to 22, and -- but the option is embedded with the LTA program, yes.

Ladies and gentlemen, we are taking the last question. And the last question is from Randy Abrams, Credit Suisse.

Just a couple of clarifications. One for the fourth quarter, following up on Charlie’s question, the low 40%s. If you net the utilization decline with the currency benefit, it seems to get to only a couple of points, like about a 2-point sequential. So are there other swing factors like the utilization falling further into first quarter? Or is it just a bit of conservatism baked in? I’m just trying to understand the moving parts.

Yes. I think low 40%s still with a range and we are 47% in third quarter, fueled by the magnitude of NT dollar depreciation. So it also depends on how you assume the currency movement in Q4 versus U.S. dollars. You mentioned Q1 loading, and certainly another factor as well.

Is a factor. Okay. Hey, which currency are you assuming right now for the one -- for that...

We don’t assume. We are taking the numbers out of the Bloomberg.

Okay. Understood. Okay. And then a follow-up on this payout. With the high earnings this year, but heading into slowdown, what’s the way to think about, you had $3 last year, but earnings went up a lot or $3 paid this year. How should we think about the payout level or payout ratio into off of this year’s earnings?

Yes, we certainly commit to pay at least 50% of our earnings in cash to our shareholders. And we also want to have a steady and hopefully a high level of dividend. So I think that were the basic assumptions for cash dividend payout.

Okay. And the final question I have is just 2 parts. I just wanted to clarify the push-out was tool-related. We only have about 12,000 end of year. Is that push-out any bit demand-related or it’s completely a tool push-out? And then the second part for Singapore, where you’re keeping that on track, factoring the geopolitical, are you seeing rising customer interest for the non-Taiwan side to actually move ahead with Singapore a bit more aggressively?

Well, for the first question is the larger portion of that push-out is because the equipment vendor delays. And there are some social with the demand reason, but larger portion is because they’re too delayed. The Singapore’s interest, yes, they are interesting in Singapore various customers. They are sort of benefit to us in addition to the geographic, the regional risk because there are other factors in there in play. So I won’t say that’s 100% because of the geographical reason -- geopolitical reason, but we definitely see some of the interest rise, yes.

And that capacity is 2025 is when we’d see that capacity?

Yes, late 2024, but mostly because of the 2 delays now, it’s about early 2025.

Ladies and gentlemen, we thank you for all your questions. That concludes today’s Q&A session. And I’ll turn things over to UMC Head of IR for closing remarks.

Thank you, everyone, 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. And ladies and gentlemen, that concludes our conference for third quarter 2022. Thank you for your participation in UMC’s conference. There will be a webcast replay within 2 hours. Please visit www.umc.com under the Investors Events section. You may now disconnect. Goodbye.