UMC

Welcome everyone to UMC's 2026 first quarter earnings conference call. All lines have been placed on mute to prevent background noise. After the presentation, there will be a question-and-answer session. Please follow the instructions given at the time if you would like to ask the question. For your information, this conference call is now being broadcasted live over the internet. Webcast replay will be available within two 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. David Wong, Investor Relations Manager of UMC. Mr. Wong, please begin.

Thank you. Welcome to UMC's conference call for the first quarter of 2026. I'm joined by Mr. Chitung Liu, CFO of UMC, and Mr. Michael Lin, Senior Director of Finance. In a moment, we will hear our CFO present the first quarter financial results, followed by our key message to address UMC's focus in second quarter 2026 guidance. Once our CFO complete the 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 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 description of these risks and uncertainties, please refer to our recent and subsequent filings with the SEC and the ROC securities authorities. During this conference, you may view our financial presentation material, which is being broadcasted live through the internet. I would now like to introduce UMC CFO, Mr. Chitung Liu, to discuss UMC's first quarter 2026 financial results.

Thank you, David. I'd like to go through the first quarter 2026 investor conference presentation material, which can be downloaded or viewed in real time from our website. Starting on page four, the first quarter of 2026, consolidated revenue was TWD 61.04 billion, with gross margin at 29.2%. Net income attributable to the stockholders of the parent was TWD 16.17 billion, and the earnings per ordinary share were TWD 1.29, which show pretty good growth compared to both last quarter as well as the same quarter of last year. On page five, first starting from the sequential comparison. Revenue was basically flat or down 1.2% sequentially to TWD 61.4 billion.

Gross margin at 29.2%, slightly declined from the previous quarter of 30.7%. Net income attributable to shareholder of the parent though, has increased 50% sequentially to TWD 16.17 billion, partially due to the strength of the stock market performance and the non-operating income, grow 50% to TWD 5.3 billion in the first quarter of 2026. EPS as a result reached TWD 1.29, EPS per ADS is TWD 0.204 in the first quarter of 2026. On page six, year-over-year comparison. Revenue grow by 5.5% year-over-year, mainly due to shipment increase.

In gross margin also show 2.5 percentage point improvement to 29.2%, to TWD 17.8 billion in the first quarter of 2026. EPS also show nearly more than 100% growth in the net income compared to TWD 7.7 billion in the first quarter of last year. On page seven, balance sheet highlights. Total equity reach TWD 406 billion, and cash on hand still over TWD 100 billion at the end of first quarter of 2026. On page eight, our ASP declined slightly in the first quarter of 2026, mainly due to a better-than-expected 8 in wafer shipment, which bring down the blended ASP. On page nine, our revenue breakdown by different geography.

The changes are very minor. We see some have declined in Europe region from 11% in the previous quarter to 9% in this quarter. The other region pretty stay relatively similar compared to the Q4 2025. On page 10, IDM show a bigger decline from 20% in the previous quarter to now 14% of the total revenue. On page 11, Communication also declined 3% quarter-over-quarter to 39%, when consumer increased by 4% to 32% in first Q 2026. For technology breakdown, our revenue below 40 nm still remain over 50% of the total shipment. 28 nm, 22 nm is around 34%, slightly declined from the previous quarter.

On page 13, there's some annual maintenance schedule or maintenance in the first quarter of 2026, resulting slight decline in available capacity in the first quarter of 2026. We will see the total available capacity to go back to the previous level in the second quarter of 2026. On page 14 is our overall budget annual CapEx, which for the time being, still stay around $1.5 billion. The above is a summary of UMC's results for first quarter of 2026. Next, I would like to go to share our key messages. In the first quarter, our wafer shipment increased 2.7% sequentially on a relatively strong growth in the consumer segment, lifting overall utilization rate to 79%, which is a continued improvement.

Despite decline in blended ASP during the quarter, which I explained earlier, this is partially reflected higher 8 in wafer shipment. Gross margin held firm at 29.2%. Demand for our 22 nm logic and specialty process continued to gain momentum, with 22 nm revenue now reach another record high and accounting for about 14% of total first quarter revenue. At the end of this year, over 50 customers will have complete tape-out on our 22 nm platform for a very diverse range of applications, including display driver IC, network chips, and microcontrollers. We continue to invest in the next generation technology beyond 22 nm. Our 12 nm collaboration with our partner will provide customers with technology continuity as well as a U.S.-based manufacturing option.

UMC also recently announced important development in the emerging business, including a strategic partnership to deploy thin-film lithium niobate TFLN photonics for AI infrastructures. Going to second quarter, we expect strong wafer shipment growth across both 8 in and 12 in portfolios, supported by a strong rebound in the communication segment, as well as healthy demand across computer, consumer, and industrial markets. When the current memory supply shortage and ongoing conflict in the Middle East are creating certain headwinds and market volatilities, UMC continue to foresee resilient market demand. UMC will continue to monitor industry and macroeconomic development closely when prudently managing our business to cope with market dynamic amid evolving semiconductor landscape change. Let's move on to second quarter 2026 guidance. Our wafer shipment will increase by high single digits, and ASP in U.S. dollar terms will increase by low single digit.

Gross margin will be approximately 30%, and capacity utilization rate will be in the low 80% range. Our 2026 cash base CapEx, as I mentioned earlier, so far will maintain around TWD 1.5 billion budget. That concludes my comments, and thank you all for your attention. Now we are ready for questions.

Yes, thank you, Mr. Liu. Ladies and gentlemen, we will now begin the question-and-answer session. If you have a question for any of today's speakers, please press star one on your telephone keypad and you will enter the queue. After you are announced, please ask your question. If you find that your question has been answered before it is your turn to speak, then please press star two to cancel the question. Thank you. Now please press star one on your keypad if you would like to ask a question. Thank you. Our first question will be coming from Gokul Hariharan of JPMorgan. Go ahead, please.

Yeah. Hi, Chitung. Thanks for taking my question. First of all, on the pricing environment, I think last time, you guys have talked about pricing environment being more favorable. Any more improvement that you're seeing on the pricing front right now, in terms of your discussions with customers? Is it mainly to reflect the higher operating costs, or are you able to kind of recognize some price increases even beyond the operating cost improvement? When I look at Q2 low single-digit QoQ price increase, is that mainly a blended price increase because 12 in is growing faster or is there a like prie increase included here as well?

Yes. We recently sent out a letter to our customers, talking about the price increase to happen in the second half of 2026. The blended ASP in second quarter increase is, yes, mainly from the mix improvement. I would say 22 nm and 28 nm will be the main help for the blended ASP increase in second quarter. When the first half of 2026 were underway, we are seeing resilient demand across a broad range of fabrication, including communication, industrial and consumer, and even the AI-related segment for us. This momentum is contributing to a sustained and increasing tight capacity environment across UMC's portfolio. To support such demand, UMC continues to enhance manufacturing efficiency and invest in technology and capacity to ensure reliable, high quality wafer supply.

This ongoing investment, together with increasing key cost drivers, including raw materials, energy and logistics, are essential to sustain our long-term operational excellence and service commitment. In light of these factors, we will implement a wafer price adjustment in the second half of 2026, which will set up a more favorable position for the upcoming 2027. The pricing reflects both the evolving supply and demand environment and the continuing investment required to support our customers' growth. Pricing adjustment will based on the factor, including UMC's product mix, strategy, capacity agreement, and also the long-term partnership. This pricing adjustment will be implemented. Hopefully, we will do our best in a very disciplined and sustainable manner to ensure our operational health and continuing to support our customer growth and long-term success.

Got it. Thanks, Chitung. Just to follow up on that, I think, I think if I remember right, I think your letter was like 8%-10% for 8 in and 12 in. Are we seeing further potential for increasing price, like? What is the reception you're hearing from the customers that have that you've been in consultation with? I think, is there broad acceptance of this pricing freeze or you see some degree of pushback given some of the customers, the demand seems to be still pretty sluggish.

Yeah. One thing we don't want to do, or we never really did, is being opportunist to take advantage of customers. That's something we will never do. Our pricing strategy has always been anchored in the value where we deliver our differentiated technology, diversified manufacturing footprint, and hopefully, again, work at operational excellence. These trends continue to enhance our customers' product competitiveness and strengthen their supply chain resilience. As semi supply chain evolves, we are seeing market recognition on UMC value, and this is driving structurally higher and more sustainable demand. We remain committed to investing in this core strength to reinforce a virtuous cycle of value creation. In parallel, we continue to enhance ASP through structure product mix optimization.

This includes strong momentum in our 22 nm platform and reduction in commoditized segment exposures, and thereby strengthening UMC's long-term ASP profile. I'm pretty sure we didn't really mention anything numerical in our letter to customers. It will be based upon different segments, different technology, and our long-term partnership.

Got it. My second question on gross margin. We had some improvement in utilization from like mid-70%s to low 80%s in Q2 based on your guidance. Gross margin is still roughly hanging around 30% now. With this price increase, like, how should we think about gross margins? Are we likely to get back to like high 30% or 40% kind of levels that we were in back in the 2021, 2022 kind of time frame? That might require a much bigger improvement in utilization.

Unfortunately, we're still in the peak of our depreciation increase cycle. I think in the previous quarter we mentioned our depreciation curve will only peak out starting from next year. We're still seeing quarter-after-quarter depreciation expenses. If you ask me about EBITDA margin, I think I will be more comfortable talking about a better upside. If you talk about gross margin or operating margin, we still under a lot of pressures from depreciation expenses increase. Not to mention the recent geopolitical tension lead to raw material costs and the energy costs and logistic costs to increase. Our utilization rate will be in the low 80% range in Q2.

However, the margin uplift from higher shipments will be largely offset by the higher depreciation and higher utility costs, like I mentioned. For the remainder of 2026, the 12 in in Singapore ramp up will start in meaningful terms in the second half of 2026, which will continue to carry higher depreciation expenses over the next several quarters. Of course, from UMC's side, we will continue to proactively deploy cost reduction efforts, including multi-sourcing, streamlining our operations, managing supply chain pricing and drive automation transformation. All these measures will help UMC to partially offset the cost headwinds and maintain our or hopefully enhance our EBITDA margin.

Got it. Yeah. Thanks, Chitung. Thank you very much.

Thank you.

Next one, Charlie Chan, Morgan Stanley. Go ahead, please.

Thanks for taking my question, thanks, Chitung. Also, congratulations for a very strong guidance and outlook. I do have some questions about the details of it, especially, you comment about the communication segment is very strong. May I know, is that coming from AI-related networking or also the smartphone business you're also seeing, so-called a rebound? Thank you.

First of all, communication was weaker in third quarter, so there will be a meaningful rebound in second quarter. For UMC will be driven by DDI, networking, FPGA and ISP. Those segments will show stronger growth in the communication segment for the second quarter.

Okay. If I may, I think the company also offering lots of new technology, no matter silicon photonics, foundry service, or your advanced packaging, right? Can management talk about your future plan for those advanced packaging capacity expansion and also potential revenue contribution in the coming two years?

These are two questions. One is on packaging, one is on silicon photonics. Maybe David can help to answer the question.

Yeah, sure. As far as advanced packaging, as you're aware, we're seeing more engagements pick up on our advanced packaging solutions. As you know, we're working with more than 10 customers on advanced packaging. Currently, we expect more than 35 new tape outs in 2026. We foresee that revenue for advanced packaging next year will be significantly higher. As of now, we're in production with our for an bridge die solution and discrete DTC, Deep Trench Capacitor, with more products that'll ramp up shortly.

May I know for the bridge die or DTC, are those working with other foundry partners? Is that part of the TSMC supply chain or TSMC supply chain?

Yeah. We don't really comment on our partnership. As you know, as we talked about last quarter, these things are picking up a little bit of steam.

Okay. Okay. On top of that, do you need to further expand your so-called interposer capacity for that bridge die demand? Or, the current capacity is sufficient for that bridge die? I'm assuming that, taking the interposers more wafer capacity, whereas the bridge die, though more advanced, right? It doesn't really come much wafer capacity.

Yeah. As far as capacity planning for all of our new businesses, that will be aligned with our customers as well as market demand. Our ramp-up schedule will be aligned with the market outlook.

Okay. Okay. Thanks, David. Thanks, Chitung.

We're moving on to silicon photonics as well.

Oh, sure. Yes, please.

For silicon photonics, we're now working with industry-leading customers that will help us ramp on silicon photonics. Additionally, preliminary data shows that our silicon photonics performance is on par or better than our peers. This is a result of our manufacturing excellence and also the fact that we use fully automated benches equipment versus peers. In addition, we're on track to deliver our PDK 1.0 in 2027, which is based on the IMEC license.

We are also enabling integration for our customer by evaluating hybrid bonding, TSV, and Chiplet integration.

Okay. Got it. Thanks. I'll be back to the queue. Thanks.

Sure.

Next question, Sunny Lin, UBS. Go ahead, please.

Good afternoon. Thank you for taking my questions, and congrats on the improving outlook. My first question is to follow up on what you guided earlier this year. Three months ago, you did guide better growth from second half of this year, driven by new products and market share expansion. Now, Q2 is indeed pretty solid. I wonder, how should you think about going to second half? Do you still hold the view that for this year, you should see even stronger growth going to second half the year?

We remain optimistic about our overall 2026 business outlook. In the first half, we are seeing resilient demand across a broad range of applications, such as communication, industrial, consumer, and AI-related sectors. This momentum, we think, is continuing to sustain an increasingly tight capacity environment across UMC's portfolio. We expect this momentum to continue into second half, especially our 22 nm logic embedded high-v platform, where we expect to grow in the high-teen percentage range for those segments, second half compared to first half. There's also another driver coming from our image recovery, which is progressing to deliver good growth year-over-year because of somewhat low base last year. The stronger second half outlook support our expectation of a full year performance improvement.

I think we still stick to UMC's going to outperform the growth of our addressable market in 2026. We're definitely, from UMC side, we're committed to deliver a better performance than exist last year's results. Hopefully this will be a turning point for UMC to broaden our addressable market. This ASP uplift we talk about in the second half, together with our strategic investment for the upcoming 12 nm silicon photonic and advanced packaging are all going to support UMC's sustainable long-term growth.

Thank you very much, Chitung. May I follow up on full year outlook? I think several foundries have reported better outlook for 2026. One is for addressable market, do you see some upside for low single digit growth?

I think the UMC addressable market show some incremental improvement, but not really significantly different. Overall market, I think, is definitely better if you include all the AI boom. Overall, the semiconductor industry is projected now grow by mid-teen in 2026. UMC's addressable market may be slightly better, but still grow by the low single digit percentage, which is again, only slightly better than our last quarter's forecast.

No problem. Thank you. My second question is if we look at high level, there are lots of concerns around disconnect between consumer end market and mature foundry improvement. Maybe if you could help us understand why for this year, although smartphone and PC end market are showing some weakness, broader mature foundry space, including UMC, get to see improving demand throughout the year, it seems. For UMC, how should we think about your server exposure? What would be the key products that you get to serve from server? From here, would you be able to benefit from server opportunity as well?

Our technology predominantly support customer addressing high-end market segment. The low-end demand tend to be more resilient. Even with recent memory tightness, supply is typically prioritized for those high-end, high-value devices. As a matter of fact, our value-added technology, including 28 nm, 22 nm, high-voltage and RFSOI, will help customer gain more shares in the high-end smartphone segment in 2026. This will also help UMC navigate the headwinds from the communication slash mobile segment. When we remain attentive to the potential impacts from either memory tightness, market tightness or Our current assessment is that any potential headwinds are manageable. We will continue to monitor the situation closely with our customers. The other factor is really UMC has been defocused on the commoditized segment.

Any so-called generic commodity type of market segment, UMC will certainly try to scale down our exposures. That will help our overall position as a foundry. As for server, we don't really have a breakdown by the server end market, which we can maybe try to do that. This is maybe two layers, three layers away from our market segment.

No problem. Thank you very much.

Thank you. Next question, [Hause Lu], Bank of America, go ahead, please.

Yes, thanks, Chitung and David. Congrats on the very good result and the guidance. I guess two questions from me, starting from utilization. You reported in the first quarter 79% and will be up to 80% ± range in second quarter. Would you be able to provide some of the breakdown between 8 in versus 12 in nodes? Also your expectation for second half this year, judging from your guidance just now that you think second half will be better than first half. Also separately, between respectively for 8 in and also 12 in into second half. Thank you.

It's, it's not around 80%, it's above 80%. It was definitely 80% something and for the second quarter. For Q2, we will see stronger growth coming from 22 nm and 28 nm, relatively speaking. 8 in will continue with some rebound. Our Japanese operation is below corporate average, which is more in the 65 nm, 80 nm technologies. For quarter two, even though 8 in will show some improvement, it's still slightly below corporate average. 12 in as a whole are still slightly above corporate average, but the gap is certainly narrowing.

Okay. Got it. Second question is just regarding the pricing line. You discussed about like for like pricing, blended pricing outlook seems to be tracking better because of improving mix as well as improving utilization across the board. I think two things. One is just on the pricing outlook for full year. I remember last time you mentioned it is going to be firm throughout this year. Would you be able to provide some update on that? Second thing is for the pricing on the like-for-like basis, would you be able to just share with us that's your strategy on pricing? I understand there's a lot of macro factors moving, a lot of moving factors in the macro environment that you might need to pass on the cost to your customers.

Would you be able to share if there's any time in the history that your customers have been willing to or were willing to accept a price hike when your utilization is at around low 80% levels? Two questions here. First one is on the pricing outlook for full year. Any update on the versus the last time frame outlook? Second thing is just on your long-term view that's regarding the pricing environment right now versus your long-term trajectory of the business that 80%. Is this sufficient enough for your customers to accept a price hike across the board? Thank you.

Again, we really don't want to be perceived as opportunists to take advantage of our customers. The pricing adjustment, coming back to our pricing strategy. The ASP improvements are fundamentally anchored in our value proposition, and also the technology and manufacturing service, which is supported by a structured demand rather than short-term pricing tactics. We are seeing customer market share gains along with ongoing structural shift in the foundry landscape, generating durable demand for UMC. We want to emphasize our differentiated technology and global footprint enable customers to strengthen their competitiveness across all segments, including AI, communication, consumer, and industrial automotive segments. Like stated in our letter to customers, starting from the second half of 2026, we will implement disciplined pricing adjustment to mitigate some cost headwinds by maintaining customers' competitiveness.

Meanwhile, hopefully more product mix optimization driven by a strong 22 nm demand will continue to drive ASP expansion. Even though the steady recovery in 8 in loading may partially offset the blended ASP list, our overall 8 in recovery plan is progressing well with several fab running nearly 100%.

That's great. That's very clear. I think just a quick follow-up on that is probably the firm pricing outlook for full year, you guided last time, is it still holding the same statement, or is this actually incrementally better? On the EBITDA margins, you mentioned just now that you have a better target on that. Would you be able to quantify it? Thanks.

If we take out the increased depreciation largely coming from the Singapore fab ramp with better loadings and potentially some pricing adjustment in the second half, we are confident if you take out the depreciation, the improved outlook should be able to offset some of the cost increase, especially energy and logistic, et cetera. Our pricing outlook certainly with the price increase ledger to customers is slightly better than the previous quarter.

Got it. Thank you so much, Chitung and David. I'll be taking the queue.

Thank you. Next one, [Skati Ku], BNP. Go ahead, please.

Hello, management. Congrats on the very good result and the guidance. My question is a follow-up to a previous one. The consumer segment revenue seems very strong with a 4 percentage point increase in the product. I wonder, is it because of the demand recovery or the pricing dynamic changes? Going forward into the second quarter and the full year, how would you see the trend will be like? Thank you.

Consumers, growth in Q1 was mainly driven by Wi-Fi and DTV set-top box, tech. For the second quarter, the growth will continue and will be driven by MCU, LCD controller and power-related products.

Yes. Got it. Thank you very much.

Thank you. Next one, Laura Chen, Citi. Go ahead, please.

Yes. Hi, can you hear me clearly?

Yes.

Yes. Hi, thank you for taking my question and congrats for the good result. I'm just wondering that the progress with Intel's engagement. We already have a good progress per management of previous mention. Just wondering that for next year, if we start to see some like progress and breakthrough, how should we think about that, the potentially again, increasing, maybe R&D force or any impact on our like revenue and also expense?

We cannot give revenue guidance now. I think the timing-wise, we are talking, we are planning by later 2027, we will start to see initial commercial production. In terms of investment, it's already happening, and that's also partially reflect in our increased R&D expenses. Michael can help me to comment more in some details.

Yeah. Yes. This 12 nm project work with Intel continue to go well. We remain on schedule to de-deliver the PDK and associate IP to customer in 2026. We anticipate that the product tape out will commence in 2027, which will making a significant step toward the commercial deployment and future revenue growth. UMC and Intel are working closely to ensure this successful tape out and efficient ramp up to mass production for the 12 nm customer product. So the application that we, for this 12 nm project will be including the DTV, Wi-Fi connectivity and high-speed interface product.

Sure. Thank you. I'm also wondering that since the expansion in the U.S. is probably one of the direction UMC is looking for. I'm just wondering that following the 12 in technology, any plan to further engage with the more advanced node with Intel?

We have to stick to however we have and to make it execute well and solid. We cannot speculate the future. Our focus now is deliver the 12 nm platform to customers. In the future, if anything makes sense for both partners as well as our customers, certainly we will consider to extend our collaboration to other derivative for technologies. For the time being, the only focus is on 12 nm platform.

Okay. Very clear. Thank you.

Next one, Felix Pan, KGI. Go ahead, please.

Hi, thank you for taking my question. I got two question. Recently there's a lot of rumor talking about the UMC in talk with the client about the potential memory foundry business. I know it's a little bit unlikely, and Chitung also mentioned that you guys gonna scale down the commodity business, but I still want to get some clarification how the company see the opportunity for the current strong memory demand, both the NAND or NOR. Is that possible? Would do anything business related to the memory? That's my first question.

Again, we will not be able to comment on, of course, market speculation like this. But our strength is really in the differentiated specialty technology, which elevate our competitiveness to collaborate with customers. We will pursue long-term and sustainable business opportunities, which demonstrate by our current comprehensive technology portfolio, such as embedded high-v, embedded non-volatile memory, BCD, RF SOI, et cetera, et cetera. Again, we will not do short-term opportunity chase. This is just not our way of managing business.

Okay. Okay. Thank you. That's very clear. My second question regarding to the 8 in tightness at the moment. Based on my understanding, this is primarily driven by the global leading foundries. They optimize their capacity, so some they're exiting some business for their 8 in foundry. Primarily I think this is a supply driven, but also we see some incremental demand improvement. Is that possible to break down how you guys see the 8 in tightness is more demand driven or supply tightness driven? If I can may have a follow-up to follow up the [Hause] previous question, what's the 8 in utilization rate in first quarter? Thanks.

We like to see this is really because of our competitiveness. We always prepare ourselves to cope with industry dynamics, and we welcome any opportunity to support our customers. We view this landscape shift as an opportunity rather than a given. We want to work hard to further optimize our product mix and gradually improve our performance. It's very difficult to differentiate the two factors you mentioned. Again, we like to think the only thing we can control is our own competitiveness and our technology portfolio. We will continue to work hard to invest, to broaden our technology portfolio and our service to our customers.

Okay. First quarter's utilization for 8 in, if I may ask?

First quarter company-wide was 79%. As I mentioned, previously it was, 8 in is below corporate average. The situation, the delta, the improvement in the second quarter is higher for 8 in. For second quarter, the 8 in average loading still were below corporate average.

Okay, thanks.

Thank you. Next one, Bruce Lu, Goldman Sachs. Go ahead, please.

Hi. Thank you for taking my question. The question is regarding to the legacy node for 12 in. You know, your competitor is talking about like exiting the market. You know, what's the real situation for UMC is facing right now, you know? How much more business we can expect for the legacy node, you know, overflow or in different ways that do we see the possibility to, you know, kick off another round of CapEx in especially Singapore for like LTA with the customer for the potential new business?

Well, this is a very hypothetical question for us to answer. I mean, it's very difficult. I mean, it's somewhat similar to the fundamental of our 8 in views. The only thing we can control is our own competitiveness and technology portfolio, and we think there still are plenty of upside there, no matter is 8 in or 12 in legacy market segment. Of course, the market dynamic shift help us or have presented the opportunity, but it's really up to us to have the competitive edge to gain those opportunities. Those are the areas we are focusing right now. If you talk about this advanced packaging, it actually going to take some of the legacy part of the 12 in capacity in our Singapore fab.

If the market dynamic continues with the customer demand, certainly there's upside in terms of capacity for those 12 in capacity in Singapore.

Well, I should ask in different ways that earlier, so the previous investment is that, you know, you only take LTA for the new capacity expense, expansion for your 12 in. Is that still the case for the future capacity expansion?

Well, we don't want to limit ourselves to the market opportunities. Back in three, four years ago, when the market present the need and we need the customers to share the investment risk, that's where the LTA comes from. Going forward with all the new technology opportunities, such as the silicon photonics and advanced packaging, we will continue to work closely with our customers, including share the risk of further investment. Will that be in the form of LTA or any other forms, we cannot comment because we are still in the early stage of the technology development. The outlook is promising, but it's still a little bit too early to comment.

Understand. Thank you. My second question is, can you comment a bit about like, you know, 14 nm high-voltage progress? Because I think we asked a question a couple quarters ago when Jason answered that the driver IC might not need to go for 14 nm and beyond. But right now, TSMC is talking about like 14 nm high-voltage process, right? Is that the, like, the technology trend is getting clear that the driver IC will continue to migrate to smaller geometry?

I think to start by first, we don't comment on competitor. Okay.

I'm asking about the driver IC technology trend, right?

We have a proven track record for driver IC. For the current industry lead in 22 nm, 28 nm for late display solution, UMC is always recognized as the global leader. when customer migrating to FinFET, and that's where our FinFET heavy solution will continue to provide better performance, lower leakage and more dies, more die size savings. again, it's all boil down to our own competitiveness, and we do have the upcoming FinFET heavy solution as well.

All right, [Chi]. Thank you.

Thank you. Next one, Gokul Hariharan, JPMorgan. Go ahead, please.

Yeah, hi. On the silicon photonics piece, could you talk a little bit more about the kind of engagements that UMC is making? Are these mostly for pure pluggable silicon photonics, or are you also engaging in some of the CPO-related projects? Given that you also have this PDK for the IMEC version of the technology coming out soon, how should we think about the ramp of the photonics-related revenues over the next couple of years? Should we expect some meaningful progress next year, or do we have to wait for this IMEC related IP to really be out there before we start to see some photonics related revenues really kind of hitting the P&L?

Yeah. As far as the current silicon photonics, the key milestone is for us to release the PDK in 2027. It'll be version 1.0. Obviously, it's based on the IMEC license. As far as the current designs, they're for PICs, they're basically pluggable solutions. At the same token, we're also looking to enabling integration for customers by considering other, you know, hybrid bonds, TSV solutions or Chiplet integrations that will help us be in a better position when CPO kind of takes place further down the road. For now, it's pretty much all the a lot of the PICs discussions and designs that we're under customer engagement.

Okay. That's clear. Secondly, on the mature 12 in nodes, I think 2022 still seems to be pretty strong in terms of utilization. Could you comment a little bit on 40 nm and 65 nm, 55 nm status? Like, how are the utilization there? Especially given you commented there is some slack in the Japan, which I think, if I remember right, was 55 nm and 40 nm. Any forward-looking comments on how that utilization is likely to get filled given that you're also engaging some of the bridge die projects?

Yeah. As far as for the 40 nnm, 55 nm and 65 nm, short term, I think the revenue contribution for Q2 will be healthy. From a longer term perspective, we're confident on the business outlook for UMC's 40 nm and 55 nm and 65 nm technologies. We are seeing longer term, there's gonna be, you know, more designs and that'll hopefully lift some of that long-term utilization rates.

Any products that are that you can call out here that are critical here to lift that utilization rate?

Well, I think, right now they're under discussion on customer engagements, but once we've seen some real material uplift in UTR, we'll be more than happy to share them with you. Okay, cool. Thank you.

Thank you. Ladies and gentlemen, in the interest of time, we're taking the last question. The last one, Charlie Chan, Morgan Stanley. Go ahead, please, Charlie.

Thanks for taking my follow-up questions. First question is follow on the pricing strategy. As a previous caller just mentioned that you did send some letters to customers. I'm wondering what's the customer's reaction. Meaning, are they kind of very happy to accept the price hike because they can also pass through to customers? Given some end market difficulty, some customers have some pushback. It would be much easier if management can give us some like preliminary second half price hike assumption. Thank you.

Well, I cannot speak for our customers. Again, we appreciate their long-term support. Just like our vendors, those raw material supplier and the energy supplier to UMC, it's going to be a win-win for the longer term. We need those to continue to provide efficient manufacturing and continuous investment. I'm pretty sure our customer understand where it is coming from, the key is really how UMC can help them to increase their competitiveness in the longer term and gain more shares. I think that's the key message we want to deliver to our customer, and also we appreciate their long-term support.

Got it. Thanks. Also a follow-up question to Laura's question about the Intel partnership. I want to associate that to my previous question about your advanced stages, especially you said bridge die and BCD, the Deep Trench Capacitor. Is that the right way to think about that, because Intel EMIB also need those bridge die and DTC discrete components? Do you think is it right way to think about UMC would be a very important partner for Intel's EMIB-[E] or Intel's advanced packaging supply chain? Thank you.

There's a lot of speculation here and we cannot do that. We have to respect our important partners. Again, our current focus is on 12 nm platform.

Mm.

No, nothing else. There's this important collaboration for both parties and, we have to make it work. This is too important, especially for UMC. We are putting all the possible resources, try to make sure we deliver.

Yeah. Anyway, it sounds very, it sounds very reasonable because you have all the capability and technology that your key partner may want. We look forward to your next update. Thanks, Chitung.

Thank you.

Thanks.

That concludes today's Q&A session. I'll turn things over to UMC IR Manager for closing remarks. Go ahead, please.

Thank you everyone for joining us today. We appreciate your questions. As always, if you have any additional follow-up questions, please feel free to contact UMC at [email protected]. Have a good day.

Thank you. Ladies and gentlemen, that concludes our conference for 1Q 2026. Thank you for your participation in UMC's conference. There will be a webcast replay within two hours. Please visit www.umc.com under the Investors Event section. You may now disconnect. Thank you again. Goodbye.

Welcome, everyone, to UMC's 2025 Fourth 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 fourth quarter of 2025. I'm joined by Mr. Jason Wang, President of UMC; and Mr. Chi-Tung Liu, the CFO of UMC. In a moment, we will hear our CFO present the fourth quarter financial results followed by our President's key message to address UMC's focus and the first quarter 2026 guidance. Once our President and 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 Financial 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 description 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 fourth quarter 2025 financial results.

Thank you, Michael. I'd like to go through the 4Q '25 investor conference presentation material, which can be downloaded or viewed in real time from our website. Starting on Page 4, the fourth quarter of 2025. Consolidated revenue was TWD 61.81 billion, with a gross margin around 30.7%. The net income attributable to the stockholder of the parent was TWD 10.06 billion and the earnings per ordinary shares were TWD 0.81. Utilization rate in the fourth quarter is stayed the same as the previous one, around 78%. For the sequential comparison, revenue grow 4.5% quarter-over-quarter to TWD 61.8 billion. Gross margin improved to over 30% to now 30.7% or gross margin of TWD 18.95 billion. And the non-operating income remained similar to that of last quarter. And the net income overall contributed to shareholder of the parent is around TWD 10.05 billion or EPS of TWD 0.81 in Q4 of 2025. For year-over-year comparison, on Page 6, revenue grew by 2.3% to reach TWD 237.5 billion for the whole year of 2025. Gross margin rate is around 29% or TWD 68.9 billion. And for the net income attributable to the shareholder of the parent for year 2025, is around TWD 41.7 billion or 17.6% net income rate. EPS for 2025 was TWD 3.34, which is a decline compared to that of TWD 3.8 in 2024. On Page 7, our balance sheet at the end of 2025. Cash amounts still more than TWD 110 billion, with total equity of the company is now TWD 379.8 billion at the end of 2025. For ASP on Page 8, you can tell for the last three quarters or four quarters, it pretty much remained similar level for our blended ASP for throughout the 2025. For revenue breakdown on Page 9. For quarterly comparison, the change is mainly showing in the increase in Asia and Europe with now North America represents about 21% in Q4 of last year. For the full year breakdown on Page 10, the change is similar. We see North America dropped from 25% in 2024 to 22% in 2025. For Page 11, IDM for Q4 revenue still represent about 20%, almost no change. But for the full year number on Page 12, IDM account for 19%, increased by 3 percentage points to 19% in 2025. For quarterly revenue breakdown by application, it remains almost similar quarter-over-quarter on Page 13. For the annual performance on the application breakdown on Page 4 (sic) [ Page 14 ] consumer increased by 3 percentage points to 31% from 28% in the previous year. And we continue to see 22-nanometer to be our key driver of growth for the recent quarters and also forward-looking as well. So 22 and 28 nanometers revenue in Q4 '25 now represent 36% of the total revenue pool. On Page 16. For the full year, the increase of 22 and 28 nanometers revenue is 3 percentage points, and we also show about 2 percentage point increase in 14-nanometer on a year-over-year comparison. Capacity remained flat on a quarter-over-quarter comparison base, but it will decline by roughly 1% due to the annual maintenance schedule. On Page 18, our latest forecast for 2026 CapEx plan is around USD 1.5 billion, which is slightly declined from USD 1.6 billion in the year of 2025. The above is a summary of UMC's results for Q4 2025. More details are available in the report, which has been posted on our website. I will now turn the call over to President of UMC, Mr. Jason Wang.

Thank you, Chi-Tung. Good evening, everyone. Here, I would like to share UMC's fourth quarter results. In the fourth quarter, our results were in line with the guidance with a flattish wafer shipments amid mild demand across most of the markets. The 4.5% revenue increase during the quarter was supported by favorable foreign exchange movements as well as a sequential growth in our 22- and 28-nanometer business, which continues to improve our product mix. With the 22- and 28-nanometer segment, 22-nanometer's revenue increased 31% quarter-on-quarter to a record high, accounting for more than 13% of total fourth quarter revenue. Looking at the full year, UMC delivered solid performance in 2025 with shipment increasing 12.3% and revenue in U.S. dollar up 5.3% year-on-year. Going into the first quarter of 2026, we expect wafer demand to remain firm. UMC is confident that 2026 will be another growth year as a tape-out on our 22-nanometer platform accelerate, and other new solutions continue to gain business traction. We have been working hard to lay the foundation for our next phase of growth, investing for the future in both capacity and technology. In 2025, we completed the new Phase III facility at our Singapore Fab 12i, which is already playing a central role in supporting customers to diversify supply chain. At the same time, we are striving to expand our footprint in the U.S. through an innovative yet cost-effective modes of partnership, such as our 12-nanometer collaboration with Intel and the recently announced MoU with the Polar Semiconductor. The leadership UMC has built over the past few years across specialty technologies, including embedded High Voltage, Non-Volatile Memory, and BCD, has and will continue to sustain stable business growth. Looking ahead to 2026 and beyond, we expect advanced packaging and silicon photonics to serve as a new growth catalysts, positioning UMC to address the evolving needs of a high performance of applications across AI, networking, consumer, automotive and more. Now let's move on to first quarter 2026 guidance. Our wafer shipment will remain flat. ASP in U.S. dollar will remain firm. Gross margin will be approximately in the high 20% range. Capacity utilization rate will be in the mid-70% range. Our 2026 cash-based CapEx budget will be USD 1.5 billion. That concludes my comments. Thank you all for your attention. Now we are ready for questions.

Yes. Thank you, President Wang. [Operator Instructions] Now first question will be coming from Sunny Lin, UBS.

So, I have a few questions. Number one, Jason, may we have your thoughts on overall market outlook for 2026. And then for semi versus foundry? And if UMC can continue to outgrow your adjustable market for this year?

Sure. Well for 2026, we expect AI-related segment remains as the primary growth driver in semi-industry. And furthermore, with the continuous commercial deployment of Edge AI applications, demand for chip using a general purpose server is also expected to rise. In contracts, the adverse effect of the memory supply imbalance could put some pressure on specific consumer electronics. But overall, the semiconductor industry is projected to grow by mid-teens in 2026. The question for foundry market, we believe that AI demand will remain strong and is the main contributor behind the low 20% growth projection in the foundry market this year. On the other hand, although the memory pricing may impact demand of the foundry market, at this time, we -- at UMC, we estimate that our addressable market will grow by low single-digit percentage. And UMC's growth was expected to outperform the average growth of our addressable market.

Got it. So, then my second question is on pricing. Lots of discussions and obviously, Chinese peers are raising pricing. So how should we think about the pricing outlook for mature foundry and for UMC through 2026? Would UMC be able to start to reflect better value? And if yes, which product categories should we expect more upside from here?

Okay. Well, we do anticipate a more favorable ASP environment in 2026 versus 2025. This outlook really reflects our disciplined pricing strategy and the positive impact from multiple reasons, product mix optimization, loading improvement and reduced exposure to more commoditized market segment. As you're referring to China players, we expect the strong growth momentum in our 22-nanometer demand to support our product mix in 2026 as well. Overall, our pricing strategy remains consistent and is anchored to the value where we deliver technology differentiation and manufacturing excellence. So, we do think the 2026 pricing environment is more favorable now. Now the question about which product, and I mean, we don't comment pricing on specific product or any specific node. But in general, we do see the environment is more favorable now.

No probolem. That's very helpful. And then a follow-up would be on the overall industry supply versus demand for the coming few years. TSMC on the recent earnings conference talked about the plan to optimize capacity for mature nodes to better support cloud AI demand in coming few years. So from your perspective, how should we think about the opportunity here? Are you starting to see more client engagement for new products in the coming few years?

We're always excited to see more customer engagement. So -- but more importantly is we need to prepare ourselves to cope with the market dynamics, and we welcome any opportunity to support our customers. So, we view this landscape shift as an opportunity to further optimize our product mix and gradually improve ASP and margin as well.

No, got it. Got it. And then maybe lastly, just on your Singapore expansion. How quickly are you planning to ramp capacity in 2026 and in 2027? And how should we think about the differentiation of products that you have for Singapore versus the Taiwan capacities for 22- and 28-nanometer? And then with that, how should we forecast the depreciation in 2026 and 2027?

Well, first of all, for the year of 2026, the capacity increase will be around 1.2% year-over-year for us. And for our Singapore facility, the expansion will start in the second half of 2026. And with capacity deployment ramp from second half of 2026 will continue into the 2027. In terms of the node available in our Singapore facility, it's our strategy that we have a geographically diverse manufacturing booking between Taiwan, Singapore, Japan, U.S. and from a technology, no coverage standpoint, we would like to coverage most of the nodes. So the customer has a benefit of passing different sets of forecast.

As for the depreciation forecast, we are looking for some like low teen annual increase in the full year depreciation expenses. As for next year, we don't have the exact number here, but it's very likely to be the similar amount for 2026. So in a way, we will see the depreciation curve to peak either this year or next year with a very similar numbers.

Next one, Haas Liu, Bank of America.

Congrats on the results. I would actually like to follow-up on the pricing. If we look at the like-for-like pricing environment, based on your current mid- to high 70 percentage of the utilization, if we strip out any of the consideration of the product mix improvement, are you able to improve or just to pass on your higher manufacturing costs or material costs to your customers at this stage? Or you still receive a meaningful pushback from your customers?

Well, I mean, the pricing discussion is always ongoing. The overall pricing strategy remains consistent, as I mentioned earlier. In 2026, we do see some market dynamic changes, so forth. Certain customers we do have some adjusted pricing upward. And so -- and for a certain customer, we still have some of the pricing -- I mean, the onetime pricing adjustment at the beginning of the year to support their market share expansion, as well as the competitiveness. So net-net, within the environment more favorable now in 2026.

Okay. Yes. So, when you talk about you are supporting your customers to gain market share by strengthening their cost structure. Do you mean you are actually adjusting down your pricing for those customers? Or is it actually up for this year?

We have a mix of that. For certain customers, we have adjusted pricing upward. And for certain customers, we will apply the onetime price adjustment downward, yes.

Okay. Got it. And then just on the near term, a couple of your Fabless customers recently talked about earlier and also stronger inventory restocking because of the memory price hike. I was just wondering what impacts your first quarter outlook here, if your customers are seeing stronger inventory pulling in the traditional low season. Why is your shipment for first quarter is still relatively flat? And then, what's your puts and takes for the first quarter overall business outlook? Just wondering whether -- which part of the business is actually relatively stronger and weak?

For Q1, by segment, we are actually in line with our addressable market seasonality. We didn't see a significant changes due to the inventory restocking. But if you're looking into by applications, we expect the revenue contribution from consumer segment to increase driven by the WiFi and DTV and set-up box, while the revenue from the communication and automotive will decline due to a softer demand of ISP and DDI products.

Okay. That's pretty clear. And then since you just mentioned about seasonality, are you expecting this year's seasonality to look pretty similar to the previous few years that first quarter could be relatively light and second quarter and third quarter, you will be able to see a relative strength into the year?

I can have -- probably provide you with this. If we look at the whole year, with the new project of a multiple specialty technology across the embedded high-voltage, non-volatile memory, power management, IC, RF SOI, it supports the end markets in communication, consumer, automotive and AI servers which will ramp in second half 2026. So, we're more looking at this year that our second half will outperform the first year -- first half, I'm sorry, the second half will be better than the first half. So that may be the deviate from the traditional seasonality. But as far as for us, we think the overall shipment for the year will be a growth year and as well as second half will be better than the first half.

Okay. Yes. And last question before I jump back in the queue is that, just based on the comment you had just now, what is the underlying market unit demand assumption you have right now? Is it smartphone -- is it the overall smartphone market will actually grow or decline based on your current base case scenario that second half will be better? Or it is actually already factoring a relatively more conservative expectation that smartphone TV, PC, this kind of consumer markets will actually see a unit decline?

Always with the current forecast from our customers. I mean, we do see gains on product segments, all applications. We do see some share gains on those applications. So right now, the forecast does show us that's more of a share gain in the market -- end market demand associated.

Next one, Felix Pan, KGI.

I just have a couple of questions about the future growth driver, particularly in the remarks, you mentioned about the advanced packaging and silicon photonics. So my first question will be besides the Interposer, what else we might have, some engagement for advanced packaging? And for Interposer, what's the capacity expansion plan for 2026? And my second question will be the silicon photonics, particularly in the Singapore fab, a lot of rumor about your potential customer. Is there any color, any client engagement or any contribution can generate from this segment? Any color will be grateful. Thanks.

Okay. A big question. So, let me see if I can cover -- cover that. And well, if I look back, I mean, I understand you asked for 2026. But let me look back this. We have delivered a very solid 2025 performance with a 12.3% shipment growth and 5.3% revenue growth, which outperformed our addressable market. This result is supported by our differentiated 22-nanometer technology and other specialty offering across both 12-inch and 8-inch amid a world-class market, a broad-based market demand recovery. And building on the 2025, we do view 2026 as a year of both continuity and evolution. We believe the UMC will once again taking shares and outperform its addressable market, and we will also see several positive inflation. First of all, as our guidance suggests, we are seeing a more favorable pricing environment. This will result of tighter supply globally as well as our differentiated technology and geographical footprint, which will drive our growth for the next few years. We are on track with our 12-nanometer cooperation with Intel, which should start see tape-out in 2027. Now that's the existing one. And your question about silicon photonics and advanced packaging. Secondly, we see 2026 as a pivotal year for those high performance, high potential opportunities such like the silicon photonics and advanced packaging. And we are making those deliberate choice, working with INEX to invest and scale them into a significant driver for our future. If you ask specifically about the advanced packaging. And there are two distinct opportunities for advanced packaging. One, we call enablers, the other we call 10 extenders. Major to explain this. I know it's long, but bear with me. So the fourth enabler, we are seeing the 2.5D and 3D packaging as well as the chiplet move well beyond just the data center and ultra high-end chip and start to spread across the broader market. Over time, we expect that advanced packaging to be adopted even on mature nodes. A good example is RF SOI. We have mentioned many times where we're already in production. In addition to the RF SOI, we are also exploring other applications with leading partners and believe we are at least 2 to 3 years ahead of our competition. What this really means for customers is better power efficiency, small form factor, and differentiated products. And for UMC, it is a strategic win-win. We believe our leadership in advanced packaging will enable us to capture more shares, sustain our higher ASC and drive better margin in many of our already established business in the long run. On the 10 extenders, we also believe that advanced packaging will help UMC address new opportunities. For example, customers are coming to us for AI-related applications. This is not necessarily just the XPU related, but we are adding value by stacking memory with the logic, adding DTC to the stack or selling the discrete DTC. We are also working with our partners to enable a total solution. Meanwhile, we are working with more than 10 customers in advanced packaging currently and expand more than 20 new tape-outs in 2026. We foresee revenue in 2027 will be a significant year for us. And the capacity question you have that capacity plan will be aligned with the customer ramp plan and market outlook. You also asked about silicon photonics. For silicon photonics, we are developing solutions, which includes ASIC, OIO, OCS, and CPO. Our collaboration with INEX allow us to deliver industry standard PDK to our customers in 2027. In addition to platform preparation, we also work with the customer on captive technology of 12-inch PIC aiming for possible product, which is expected to ramp this year. We will also combine our advanced packaging know-how with the silicon photonics as many of the applications require the integration and different substrates, process, technology and materials. Looking ahead to achieve 1.6T bandwidth and beyond, we're working with both customers and vendors for the test finding on heterogeneous material such as the TFLM. Those technologies could also be used in additional applications such as quantum computing. Again, we hope to integrate the new material the advanced packaging technology as well. So those are all integrated altogether. That's why I gave you a bit of a longer answer. I hope that explains it.

Yes. Okay. But just -- let me just a quick follow up and rephrase my question. So for silicon photonics, what's the earliest timetable we can see the revenue contribution, like most likely?

For the 12-inch PIC, for the pluggable product, we'll be expecting to ramp this year.

Okay. And about the -- because as I know about the Interposer, currently is the -- Interposer is also the bottleneck for our partner to expand their capacity. So, is there any color we can give -- how much capacity growth for the Interposer, like how much year-on-year growth or something like that?

Well, right now, the capacity planning will be aligned with the customer for the 2027 ramp. So, we will probably provide you some clarity when that comes. Right now, in 2026, we will focus on the tape-out.

Next one, Gokul Hariharan, JPMorgan.

Could you go a little bit deeper into that advanced packaging comment that you made? What is the involvement level of UMC in some of these advanced packaging solutions? Are you doing full stack? Or is it basically like previously where you were largely focused on the Interposer side of the equation? And in terms of the tape-outs that you have, what are the nature of these tape-outs? Are these mostly data center ASIC-related products? Or is this a much wider array of products other than just data center ASIC?

Sure. Well, first of all, we have reported in our advanced packaging space. We have building up some of the capability from wafer-to-wafer stacking and TSC as well as Interposer, the 2.5D and the many different capabilities. And then the way we see it, like I explained, for the enabler is we can apply those to many of the current products that we currently serve. And then -- and one example I mentioned is the RF SOI. So we have wafer-to-wafer hybrid bonding with the RF SOI solution for the mobile space already. And then, some of the capability can be built for the DTC for the stacking as well as some of the customers looking at discrete DTC already. And we are combining some of the capability into segments, the logic and the memory. Of course, we do not provide memory ourselves. So the customer will have to provide memory wafer to us. And so then we can provide wafer-to-wafer hybrid bonding on those. So on one hand, the way we see it is advanced packing is a capability per se, and it implies to the product and then we call it enabler and also expander. And the -- meanwhile, the product coverage is all the way from the mobile space, power management discussion, the AI-related -- AI-related product and also for the BCD application as well. So they were -- it's our belief is for a better reason or for the higher performance reason, and many different applications will start adapting the advanced packaging. So we think this is going to be a broad success on the advanced packaging space.

Got it. And any plans to further expand your Interposer capacity? I think we had expanded, I think up to 6,000 and then kind of stopped it there. Now some of that demand seems to be kind of coming back for some of -- one of your customers in China. So, is there any plans to expand the capacity further?

There are discussions around that. Right now, if you look at the technology itself, we have some common tools in place already, which that we can leverage of our 40-nanometer capacity, of our 65-nanometer capacity. From those common tool space, we're already allocating to this area. Now for the unique tools, then we will put in the plan for the future expansion and for the customer ramp profile. And we believe that will probably happen in 2027.

Got it. Understood. That's clear. Another question I had is on the -- just your expectations for the communication, consumer segment which is north of 70% of revenue, given all these concerns about smartphone, PC. How are you budgeting for this? Are your customers telling you that they are really concerned about this memory cost inflation? Or right now, you still don't really hear that from the customers that that's going to be a big issue from a unit perspective going through the year?

Well, we're also cautious about that topic. As of today, we have not observed any demand impact on our customers' forecast for the year, despite the recent surge in that price. And our technology predominantly supported customers addressing the higher end of the market segment, where the demand tends to be more resilient in the past and in the period of memory tightness. So, because the supply usually typically prioritize in such high-end higher-value device. While we remain attentive to the potential impact on the memory market and our current assessment is that any potential headwinds are probably manageable, and we will continue monitoring the situation properly with our customers together.

Got it. My last question is on the geographic split of revenues. I think, could you talk a little bit about the Intel 12-nanometer progress? And any color on how you will be booking revenues or profits from this partnership given the fabless, Intel fab, while you are essentially the provider of customers and some degree of IP as well into it? And secondly, on the Xiamen's fab, what's the strategy for the Xiamen's fab medium to long term, given many of your semiconductor peers in Taiwan have kind of progressively exited capacity in Mainland China?

Well, for the 12-nanometer project with Intel, overall, the 12-nanometer cooperation project with Intel continues to advance smoothly. We remain on schedule to deliver the PDK and associated IP to customer in 2026. Furthermore, we anticipate the product tape-out will commence in 2027, making the significant step towards to commercialized deployment and future revenue growth. Right now, UMC and Intel are working closely to ensure successful tape-outs and an efficient ramp up for the mass production. As the project advance, it is expected to further strengthen USD position in the U.S., right, for customer as well for us. Because the geo diversification manufacturing. Right now, the application on the 12-nanometer cooperation, including products on digital TV, WiFi connectivity and high-speed interface products. In terms of the business model, and it's probably not available for us to comment. But it is a win-win strategy that we see and will be very synergetic for both parties as well as for our customers. And we have very high confidence this will be a win-win model.

Okay. And any thoughts for the Xiamen capacity?

Yes. For the Xiamen, I kind of touched that earlier as well. I look at Xiamen, not just Xiamen itself, our core part of our competitive advantage is our geographically diverse manufacturing footprint. And the Xiamen play one of the important space for us and particularly for the local customer. So -- and at this point, the fab is actually at a full capacity. We are running at a full utilization as well. And we see -- we continue seeing many different engagements coming to this and we will across regionally optimize it from the customer engagement and product loading standpoint.

Okay. Just one more on blended ASP. I think, Jason, you mentioned that the ASP environment is more favorable this year. But overall utilization is still in the mid-70s as of Q1, right? So do you expect that this year, we could see a scenario that we could see blended ASPs moving up meaningfully like 5% to 10% or something like that, like we have had in the past or that requires a much higher level of utilization that is probably not happening this year, given your low single-digit foundry growth expectation?

Sure. I mean, the high utilization is one of the important factors, but that's not the only factor. We want to make sure the pricing strategy is enabled not only ourselves and our customers to be competitive as well. So -- but we do see the pricing environment is getting more favorable to foundry because of the loading reason. And so -- but the magnitude of that, we probably have to continue to manage it. And if we have a clarity, we will share that with you.

Next one Alex Chang, BNP.

I just have a very quick one. I just saw the company announced that they started the mass production of SuperFlash Generation 4. So just wonder how much revenue contribution from the non-volatile memory business in the past quarter or maybe past year? And also how much revenue is contributed by the power management ICs for the server-related applications?

I mean, we don't have a breakdown to provide. And the way that we break it down is based on specialty technology that includes the high-voltage and non-volatile memory and the PCB space. Right now, the specialty revenue representing about 50% of our overall revenue. And I can let you know the high voltage is about 30% of that. And the rest of that, I would say, is a combination of the non-volatile memory as well as the DCB.

Next one is Laura Chen from Citi.

I just want to follow up on the deterioration rate and also the gross margin outlook. Jason, you mentioned that the pricing environment seems to be improving more favorable. And together with firm shipment and better product mix as well as the utilization rate, so how should we think about the gross margin trend? You guided that will be high 20% for Q1. But with these favorable factors, how should we think about the margins throughout the year? That's my first question.

Yes. Gross margin can be highly dependent upon utilization rate, ASP, product mix, depreciation and foreign exchange rate. So there's a lot of variables. So beyond this quarter, it's difficult for us to give a firm outlook. For the first quarter guidance, which is high 20s, is mainly due to the higher cost, especially the higher depreciation expenses. As I mentioned, it will grow by low teens in the full year of 2026. As for 2026, we will continue to cope with higher depreciation expenses as well as the other inflationary pressure for our production, raw material and other costs. To mitigate and cope with the headwinds, we will continue with our cost reduction efforts and also all the activities to improve our productivity and drive operation efficiency. And these measures hopefully will help UMC to deliver a stable EBITDA margin and ensure our long-term financial resilience to remain intact. As a matter of fact, our 2025 EBITDA margin is actually a good improvement compared to that of 2024.

Yes, sure. And also, I think for the advanced packaging and as well as the silicon photonics is one of the key things that UMC may have a great opportunity. We know that UMC has already working on advanced packaging, previously on Interposer, probably now we'll see more various different design. So could you share with us what's about the revenue contribution of your advanced packaging right now? And how would that look like in 2, 3 years?

Currently, the Interposer was exposed to very limited customer base and also narrow application. While we have engaged with more than 10 customers and expecting more than 20 new tape-outs in 2026, we do foresee that revenue of packaging -- advanced packaging growth in 2027 will be significant.

So, significantly means that, could that be like 5%, 10% or higher?

I am expecting more than that. But I mean, if you're referring to the overall revenue contribution, we'll probably give you more guidance later. But if you're looking at the packaging itself, it's going to be significantly larger than what we're shipping today.

Next, we'll have Bruce Lu, Goldman Sachs for questions.

I want to go a little bit deeper for the silicon photonics. I mean, as you might know that your peers like GlobalFoundries, Taiwan Semi, pretty vocal about that. Can you tell us how big do you think the addressable market for silicon photonics for you guys in 2 years? And how do you win market? What is the competitive advantage for you in this business? I mean, other than working with INEX?

Well, I mean, the Singapore facility is not going to only serving the silicon photonics. Singapore facility is one of our important manufacturing site, they serve our worldwide customers, all different applications. And so it's part of our geographical diverse manufacturing strategy. So...

No, no, no, my question is for silicon photonics, our business strategy?

The silicon photonics strategy in Singapore. Okay.

No, no, no, no. I'm sorry, let me rephrase my question. So the growth driver for UMC, one of it is the CPO, I'm assuming having more business in the silicon photonics. In -- for your peers like GlobalFoundries or Taiwan Semi, they are pretty vocal about the silicon photonics and have meaningful revenue contribution already. For UMC perspective, what is your competitive advantage for UMC to win this business? And how much business you can win or how big is the addressable market for you in 2 years?

Got it. So, for the silicon photonics, our strategy is simple. Our cooperation with INEX allowed us to deliver the industry standard PDK to our customer in 2027, particularly in 12-inch. So many of our competitors is today at 8-inch and we are focused on this in 12-inch. And as we believe the 12-inch will have that advantage. And right now, we already have certain products that have proven that performance is a better and a pluggable product, and which we will expect to ramp this year. And meanwhile, we're also combining the silicon photonics with our advanced packaging know-how, so for many different type of applications then we can integrate that. So by doing that, we think we will be even providing even more value from advanced packaging combining with silicon photonics at 12-inch. I think that's where we believe we are competitive.

But that's mostly for plug-in, right? Because if you don't have the EIC, the pure CPO product might not be your key growth driver?

You're correct. We're not looking at a completely CPO package. We're looking at particularly in the PIC and OIO and OCS.

I see. I understand. That's very clear. Next one is -- and we see that the progress for the Intel project for 12 nanometers is pretty smooth. I just want to know what is the next step? I mean, when we can see a further collaboration in 10, 7 nanometers and beyond? I mean, obviously, whatever you said, the advantage at 12-inch, you can also use the same argument for 7-nanometer. What's stopping you to do that?

Well, you're also right on that. And our focus right now is on delivering the 12-nanometer platform to customers. In the future, should it make sense for both UMC and Intel as well as our customers, we will surely consider expanding our collaboration to other derivatives as well as the technologies. Yes.

But what is stopping now? What is the show stopper now?

It's not -- I won't call it stopping. I think the focus is a focus on 12-nanometer. We have to deliver a 12-nanometer today, and make sure that we deliver that program. We execute it well. And I think anything that makes sense on that, unlike you said, I think there will be a discussion, yes.

I see. Because we already assumed that you can deliver something in '27. So given that working for 7 nanometers, maybe you need 2, 3 years, we want to see the project kickoff as soon as possible.

[Operator Instructions] Now we'll have our last question, [ Sappho ] Neuberger Berman.

It's been a while. And congrats on the progress you've made throughout this couple of years. I just have a few questions. The first one is, on the market dynamics, I think previously, Sunny has asked about the TSMC is shrinking or defocusing on this mature foundry process. And it looks like not just TSMC, but also the other foundries are -- seems to be doing some leading-edge logic foundry seems to be doing the same thing. And also Powerchip recently just reached agreement with Micron as well as Intel fab, which means they're trying to streamline and re-org some of the foundry process, too. So it seems like there's a lot of supply is kind of being taken away because of the rolling out effects from the AI and crowding out some of these older nodes. On the supply side, it seems to be that actually decreasing. And on the demand side, if you look at, I think, TI just to report overnight. I think it seems like that there's been more obvious recovery on the analog MCU space. So on demand side, that's also improving. But the supply side, that's actually decreasing. So it looks like supply-demand dynamics is moving to a more favorable situation. I think that's the point why you were mentioning the pricing dynamics favorable this year. So I'm just curious about your view, if we try to compare the current like the mature foundries dynamic situation right now versus, I mean, back in 2021 when there is a severe shortage back then. How would you compare this time around versus last cycle?

I mean, that's a really good question. I mean, we saw on the market movement, the changes. And we also deep dive on this demand and supply outlook. And we think whether this is short term or long term. If you look at the driver behind us, we see -- you mentioned this is truly more of the AI phenomenon ripple effect. And so we see that AI remains to be very strong, at least in the foreseeable future. And I think this momentum will continue driving the overall demand. And meanwhile, in many of this -- the capability -- AI capability we portfoliating to even the other end market devices in the Edge AI as well. So as in this will continue. And from an economic standpoint, building any of the mature facility is not justifiable. So we do think that this could last longer compared to the over time. And I think the situation could be more of a structure going forward. And -- but again, this is at a very early stage of this market movement. So we'll pay attention to it, and we'll continue monitoring the progress. Meanwhile, like I said earlier, I think it is more a favorable pricing environment. But more importantly is we need to prepare ourselves to cope with this market dynamic. So we are welcoming all the opportunity that for us to engage in supporting the customer. And -- but the important focus today is we have to get ourselves ready to capture those opportunities.

Got it. Another question I have is your earlier comments on the pricing. I think the -- you offer some of the annual -- maybe some discount to some of our strategic clients for their share again, but also net-net wise, also seems to be pricing is going up for a majority of the clients. So net-net, it's going to still be the -- ASP still be positive. But I'm just curious about, for those clients that you're offering some discount at the beginning of the year, when it down the road is, if the next few months or quarter situation has become tighter -- and would you be able to reprice with these customers?

Those discussions will be ongoing. We're always working with our customers to reflect the market dynamics as well as the cost increases. So I'm sure, and I believe this conversation will surely happen. It happened in the past, it will happen now and will happen in the future. So the pricing discussion will continue. And I think customers understand that. And we just have to continue monitoring the market dynamic and maintain our competitiveness on both the customer and ourselves.

Yes. Well, a thought on that because of some of the pricing that started to affect it on the January 1 this year, this was actually negotiated already in fourth quarter last year, right?

That's -- some alignment on that on both volume and the pricing. So if volume has changed, of course, that's a different topic. So, there are some volume dynamic in that as well.

Yes. My question is actually is that because a lot of the pricing that's effective on January 1, beginning of the year, it was actually communicated 1 or 2 months ago before that, toward the end of last year when the time that the supply demand dynamics haven't been really that tight as compared to some of the changes that happened in the just past couple of weeks. Am I getting that right?

Yes, you're right. Yes. But those also is on certain conditions. So given the condition has changed, the some of the pricing are dynamic.

Yes. Yes. Exactly, that is what I'm trying to discuss with you. Because we also saw a lot of the other different components, different subsectors within the tech or semi supply chain that such as memory, I think the pricing were still down in July, August, but all of a sudden, September prices going up. So I'm just curious about that because when you negotiate some of this discount months ago, the supply-demand dynamics was not the same as today. So things remain fluid, dynamic and it still continue to be flexible and it's going to be dynamic and open for changes down the road, if things are moving more favorably.

I think the core of the pricing strategy is that it has to be consistent, and it has to anchor with the value that we deliver and also the customers' competitiveness. That is the core. Then usually, that is how we're centering about the pricing discussion. So that core is not compromised. Now if the condition has changed, yes, they always have some flexibility to it. So, one is called pricing strategy and position, another is core pricing negotiation. So there will be some flexibility, yes.

Yes. And the condition has started to change now.

Yes. So we do think the pricing discussion will be more favorable now, yes.

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

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 [email protected]. Have a good day.

Thank you. And ladies and gentlemen, that concludes our conference for fourth quarter 2025. 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. Thank you, again. Goodbye.

Welcome, everyone, to UMC's 2025 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 2025. I'm joined by Mr. Jason Wang, 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 2025 guidance. Once our President and 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, Financial 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 description 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 presentations 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 2025 financial results.

Thank you, Michael. I'd like to go through the third quarter 2025 investor conference presentation material, which can be downloaded or viewed in real time from our website. Starting on Page 4, in third quarter of 2025, consolidated revenue was TWD 59.13 billion, with gross margin at 29.8%. Net income attributable to the stockholder of the parent was TWD 14.98 billion and the earnings per ordinary share were TWD 1.2. Capacity utilization rate climbed to 78% in that quarter with wafer shipment just marked 1 million 12-inch equivalent wafers. On Page 5, on the sequential comparison, third quarter revenue of TWD 59.12 billion increased slightly compared to the previous quarter, mainly due to higher wafer shipment, although the NT dollar exchange rate was an unfavorable factor of around 3%. Gross margin also climbed on back of the better capacity utilization rate to 29.8%. And net income reached nearly TWD 15 billion or an EPS of TWD 1.2 per share in NT dollar terms. On year-over-year comparison, on Page 6, for the first 3 quarters, revenue grew 2.2% year-over-year to TWD 175.7 billion. Gross margin was around 28.4% or nearly TWD 50 billion for the first 3 quarters of 2025. Overall, net income for the first 3 quarters is down to TWD 2.54 per share compared to TWD 3.12 in the previous 3 quarters of 2024. On Page 7, cash still above TWD 100 billion, and total equity of the company is now TWD 361 billion at the end of third quarter of 2025. ASP on Page 8 shows we remain firm for the past 2 quarters. On Page 9, for revenue breakdown, we see -- we can see that the North America represents about 25% of the total revenue in the third quarter, which is 5% higher compared to 20% in the previous quarter. On the contrary, Asia declined by nearly 4 percentage points to 63% in the third quarter of 2025. IDM versus fabless remain unchanged on Page 10 for the third quarter of 2025. On Page 11, we noticed the communication and computers edge up in terms of sales mix when consumers declined by nearly 4 percentage points to 29% in the third quarter. On Page 12, the segment sales breakdown by technology, 22 and 28 still remain our main technology node, when 22 continued to climb in terms of percentage. Total 22 and 28 revenue reached about 35%. For 40-nanometer and 65-nanometer revenue, somewhat unchanged, in about 17% and 18%, respectively. For our quarterly capacity for the third quarter, we see a minor increase coming out of our 12x Xiamen fab with now the monthly capacity is nearly 32,000 wafers per month, and total available capacity will remain flat for the coming quarters. On the last page of my presentation, our annual CapEx is heading to our budget number of $1.8 billion with 90% in 12-inch and 10% in 8-inch. The above is a summary of UMC results for third quarter of 2025. More details are available in the report, which has been posted on our website. I will now turn the call over to President of UMC, Mr. Jason Wang.

Thank you, Chi-Tung. Good evening, everyone. Here, I would like to share UMC's third quarter results. In the third quarter, we observed demand growth across most market segments, which drove a 3.4% increase in wafer shipments and improved utilization rate to 78%. In particular, we benefited from a pickup in sales of smartphones and notebooks, driving replenishment order from customers. Our 22-nanometer technology platform continues to provide us with the differentiation in the market, with 22-nanometer revenue now accounting for more than 10% of the total sales in 2025 alone, we are projecting over 50 product tape-outs, and we expect 22-nanometer contribution will continue to increase in 2026. Aligned with our strategy of providing customers with highly differentiated specialty technologies, we recently announced the readiness of our 55-nanometer BCD platform. In addition to mobile and consumer applications, the new platform is also complemented with the most rigorous automotive standards for automotive and industrial use. Looking ahead to the fourth quarter, we are anticipating wafer shipment to be comparable with third quarter's volumes, wrapping up 2025 with shipment growth in the low teens. UMC continues to deliver competitive process technologies that enable diverse applications, which position the company to benefit from a broad-based market recovery. With the 22-nanometer logic and specialty platform, in particular, we expect to drive growth. Now, let's move on to fourth quarter 2025 guidance. Our wafer shipment will remain flat. ASP in U.S. dollar will remain firm. Gross margin will be approximately in the high 20% range. Capacity utilization rate will be in the mid-70% range. Our 2025 cash-based CapEx budget will remain unchanged at USD 1.8 billion. That concludes my comments. Thank you all for your attention. Now, we are ready for questions.

[Operator Instructions] First, we'll have [indiscernible], Bank of America for questions.

My first question is regarding the near-term outlook. Could you discuss more in detail on how you see the business by end market is trending into the current quarter in fourth quarter? It seems the guidance is above seasonal, so just wondering if there's anything driving that. And also just your initial view into first half next year, did you get any feedback from your customers on the potential restocking, or in general, they are still pretty conservative at this stage?

Sure. I mean, while we're going into Q4, we can -- as I said, we wrap up the 2025 shipments to a low teens. It's now that we project the 2025 shipment growth was supported by our differentiated 22-nanometer technology and other specialty offerings across both 12- and 8-inch amid a broad-based market demand recovery. On the 12-inch size, shipment growth was driven by a strong momentum from 22-nanometer logic for ISP, Wi-Fi connectivity as well as the high-end smartphone display driver IC. In addition to 22 and 28, overall 12-inch wafer shipment will outpace our addressable market due to our comprehensive value-added specialty portfolio of nonvolatile memory, RFSOI and BCD. On the 8-inch size, we expect a high single-digit growth in 2025, mainly led by the PMIC and LDDI. So in summary, the strength of 22-nanometer and the specialty process across both 12-inch and 8-inch platform underpin our confidence in achieving a low teens percentage shipment growth in 2025. So for 2025 Q4, we remain -- the shipment outlook remain flat. If we're looking into the early look of 2026, I think we're still going to experience some seasonality, but if I look at the entire year, despite the ongoing global economic geopolitical uncertainty, we believe our 2025 business growth momentum will continue into 2026, where we expect the wafer shipment will increase year-over-year. In addition to some TAM expansion, our 22 eHV -- 22-nanometer eHV platform, which is serving the high-end smartphone OLED display driver application, will be one of the key growth engines. We expect the overall 22- and 28-nanometer revenue to achieve double-digit year-over-year growth in 2026. However, there's still going to be some seasonality that we may have to go through. So Q1 may be one of the challenging quarter for the year. Supported by the strong customer adoption of our 22-nanometer technology, in addition, our technology readiness in RFSOI for smartphone RF front-end device will also fuel our growth in 2026. And besides the growth of the communications segment, we also foresee our enhanced version of PMIC solution, which will also continue to drive recovery in our 8-inch segment. In 2025, we foresee PMIC business will grow in the high single-digit range, and this growth momentum will extend into 2026. Our effort on the enhancing our technology competitiveness, particularly for the PMIC application, have started to yield some tangible results, and that will actually help us with the -- to strengthen our position in this market segment and for 2026 growth. If we look beyond 2026, we'll continue to develop new derivative technology to enhance our differentiated and our competitive -- enhance our competitive position. Furthermore, we are expanding our addressable market into 12-nanometer FinFET, as you know, and as well as some of the advanced packaging space. The UMC portfolio -- technology portfolio is well positioned to serve a growing demand of the power efficiency optimization, high-bandwidth data transfer as well as the improved connectivity. So I think in general, we are relatively confident in 2026, but it's still kind of early to go into the quarterly guidance.

Yes. That's pretty intensive. And I think just a quick follow-up to my first question is just when you mentioned the growth momentum could continue into 2026 compared with 2025, are you saying that the wafer shipment could actually still be growing by low teens next year at least? Because you mentioned a lot of growth drivers by applications just now, especially on 22-nanometer, 28-nanometer and also 8-inch. So just wondering whether you are implying that the wafer shipment could grow by another low teens at least for 2026?

I mean, we're not giving the blended -- the wafer shipment at this time. We're probably ready to provide you more clarity into Q1, but 22 and 28 particularly, yes, I think that when we go into 2026, we're still expecting a double-digit year-over-year growth.

And then my second question would be on your gross margin trend. I think for the fourth quarter, you guided flat shipment and also pricing. The FX seems to be -- foreign exchange seems to be more favorable at this stage. So why does the gross margin does not go higher than the third quarter? I'm just curious why that is the case. Or should we think about high 70% utilization is going to translate into like high 20 percentage gross margins going forward?

Gross margin in third quarter is actually, in fact, slightly higher than that of the previous quarter. The gross margin also primarily depends on utilization rate, ASP, product mix, depreciation and foreign exchange. As you know, even though the foreign exchange rate may be on a forecast basis, better than forecast, but still appreciate against U.S. dollars, our key receivable currency, so still in an unfavorable situation, as I mentioned earlier, that almost eat up about 3% of our total revenue. And we do expect the Q4 '25 gross margin still will remain in the bandwidth of high 20 percentage range. Despite the variables such as our depreciation, we will still see quarterly increase. And this year, we are facing 20% plus increase in annual depreciation expenses. So I hope that answers your questions.

Yes. And then just a relevant follow-up is in your cost structure. You have been able to manage the other manufacturing cost item quite nicely down in third quarter despite the fact that the labor cost is higher, electricity cost is higher and also the material or even the wafer shipment is slightly higher compared with second quarter. So could you just elaborate in more detail on how should we think about the other manufacturing costs, which I believe should be mostly variable cost? How should we think about that going to trend?

So part of our employee compensation is bonus, which is based upon profit sharing. So when we have a better quarter-over-quarter profit in the third quarter, we do have to factor in higher bonus, which increased the compensation expenses in the third quarter.

But it was still down compared with second quarter. So I was just wondering if there's any reason driving that decline and would that trend continue.

No, the trend will not continue. It will fluctuate along with our rolling profit recognition.

Next one, Charlie Chan, Morgan Stanley.

Congratulations for very strong results, especially on the gross margin side. So maybe starting with the so-called geopolitical uncertainty. So, Jason, can you elaborate a little bit what kind of macro uncertainty you see will continue in 2026? And I was asked by one of your customers about -- there seems to be some speculation about semi-tariff may come next January. So any kind of impact -- potential impact to your business or operation? And also another uncertainty, it was a couple of weeks ago, right, the rare earth kind of supply. Does your team run through some analysis about the potential impact if rare earth will be restricted again?

Sure. A couple of things, right? I mean, you mentioned about geopolitical dynamics on the tariffs. So maybe I'll start up on the tariff first. We do understand there are uncertainties and risks from the potential impact of tariffs, and we will remain cautious of those potential business impacts, and we'll be mindful in our business planning going into 2026. At this current point, we haven't seen anything yet, but we are cautious. The -- amidst the uncertainties, we'll also continue to focus on the fundamental of our business. That is the technology differentiation, manufacturing excellence and then customer trust to further strengthen our competitiveness -- competitive position. So I think we still have to go back to the fundamentals. For UMC, to address the geopolitical concerns, I do believe that UMC has a geo-diversified manufacturing site across the globe. And the global semiconductor landscape is evolving. Customer and governments are increasingly emphasizing the geographic diversification and supply chain resilience along with the tariff. But to address the structural changes and align with the customer needs, our strategic initiative, including the capacity buildup in Singapore and the U.S. and are designed to complement our Taiwan facility, will enable us to better support our customers across multiple regions. Over the long term, we are targeting a balanced capacity split between Taiwan and overseas locations, but we welcome any opportunity from our customer. Whether this is an impact or opportunity to us, we will probably have to position ourselves and ready for that dynamic changes. Yes.

So specific on semi tariff, right, I think we also went through this discussion last quarter or 2 quarters ago. So do you also hear that next January could be a final implementation of this semi tariff? And secondly, would UMC can get exemption from the semi tariff?

Well, I mean, your guess will be as good as my guess. So I'm not going to guess here.

I watch TV only.

Yes. So we're going to be cautious about this, and we're closely monitoring the progress and developments. And at the same time, given that we are investing into the U.S., so we're definitely going to present our case. But there's nothing else to update here. But if there's anything, we will definitely recall back.

Okay. Got you. And second question is about the -- your gross margin sustainability. I know this quarter, next quarter, some puts and takes, right? But just overall, right, next year, it seems like some of your industry peer, may just call it TSMC, kind of hike their wafer price. And recently, we are seeing that the back-end foundry, though it's not like your industry peer, but it's kind of your downstream supply chain, right, also attempt to hike the back-end foundry service price. So what was the UMC's kind of sort of potential wafer price hike into next year?

Well, like Chi-Tung mentioned earlier, margin reflects the result of ASP loading certain variable factors. So let's take the ASP specifically. For the ASP outlook, our 2025 ASP performance has remained firm amid a dynamic business environment, and it has remained stable at a healthy level throughout the year. And so -- and we expect the ASP will remain firm in Q4 2025. And for the 2026 outlook on ASP, we will provide more detail in the upcoming January 2026 conference call, as we are going through some discussion with our customers aligning that. So we probably have more detail to report in the next conference call.

Okay. And on the cost side, expense side, Jason, you said at some interview that your team want to drive some costs down. But I feel like most of the components whatsoever. Most of what I'm hearing this commodity cost may go up, right? So on the cost side, do you have any preliminary outlook for 2026?

Without getting into specific cost projection or outlook, I think we can probably update you of the view in cost, our view of cost -- about cost. Cost competitiveness is always a mutual goal for us and our suppliers together, so in order to be competitive. So we're closely working with our suppliers. We'll continue to drive towards cost savings in 2026, and that has been going on for many years, but we are continuing to doing that into 2026. But that includes the combination of both internal and external efforts. It's not only working with the supplier, it's also internal efforts. For example, we have already started leveraging some smart manufacturing and AI technologies internally to enhance our fab efficiency and enabling our long-term operational competitiveness. So that's also a major piece of driving our cost goal. So I think there's many of the initiatives that we're deploying, and we working with the supplier -- supply chain is just one of them.

Okay. Okay. And last one, I will be back to the queue. So I know your company and your team have been running through a lot of strategic or marketing research, right? So recently, we picked up one data point I would like to share with you and also consult your view. Because of the T-glass shortage, right, we're starting to see tightness of BT substrate supply. From your perspective or UMC's perspective, would that kind of constrain your -- some of your customers' demand, for example, the consumer or smartphone SoC demand into 2026?

Well, we really haven't seen that, but we are closely monitoring the entire supply chain resilience. The current market is driven by this AI momentum. So there are various areas demonstrating potential supply concern. But so far, we have not seen any impact to us. But like you said, we all look out there and see if there's going to be any. But meanwhile, we are managing -- from our internal perspective, we are managing our supply resilience point of view. We want to ensure the supply assurance and as well as the -- both from supply and demand -- supply and demand as well as the quality standard and cost. So I think that's always been our initiative internally. So I would just have to say we haven't seen any impact on the recent market dynamic, but it's something always on our radar screen, and we continue monitoring it.

Yes. How about smartphone or PC demand recovery, if you have a crystal ball? Do you think that 2 major segments of the end demand will significantly recovery next year?

Well, I mean, at least for the Q4 '25, we expect the wafer shipment will remain flat, and the markets reflect pretty healthy inventory level as well. We see slightly communication segment decline in our segment, but the computing, consumer, automotive are slightly increased. So I'm not sure that's affected by that particular supply issue, but it reflects probably more end demand associated.

Next one, Laura Chen, Citi.

My first question is also about the margin outlook. Chi-Tung, you mentioned that the depreciation cost for this year were up about 20% plus year-on-year. But we know that actually in the first half, the depreciation cost increased almost like 30%. So does that mean that depreciation cost year-on-year increase trend to slowing down into Q4? With overall your utilization rate and also ASP seems to be resilient and also higher exposure on 28-nanometer, should we be looking for some of the potential upside of the gross margin?

Well, other than depreciation, there are other factors. Like Jason mentioned, we will have a clear view on the ASP, which is an important component for the margin equation. But just on depreciation alone, yes, the increased magnitude, we're down to about low teens in the year of 2026 versus 20-something in the 2025. And in the previous quarter, we also mentioned either '26 or '27 should be the peak of the recent depreciation curve. So on that regard, it does provide a good floor for helping our EBITDA margin.

Okay. Great. And also the second question is, I recall that we mentioned about the Interposer business before. We know that the AI demand is surging. So I just want to understand UMC, do you have any updated view on the Interposer strategy? And also, we know that UMC also have wafer-to-wafer technology. So just wondering what's the plan here. And also, do you want to further expand the capacities on Interposer?

Well, the latest development on the advanced packaging space, we will continue preparing our advanced packaging solution for this growing market associated with the energy consumption of cloud AI and the edge AI market. For UMC, we are developing the 2.5D Interposer with DTC, the deep trench capacitor, and discrete DTC to address the power efficiency requirement in all AI, HPC, PC, notebook and smartphone space. And second, UMC is leveraging the scalable 3D wafer-to-wafer packaging stacking and the TSV to enhance the -- enhance our specialty technology offering. We are in the mass production of extremely small form factor for the 5G and 6G RFIC right now by leveraging the wafer-to-wafer stacking technology. Based on the success of the 5G and 6G RFIC that works through the wafer-to-wafer stacking, we are also developing memory-to-memory stacking and memory-to-logic stacking service for the high-bandwidth computing requirements. So our technology really is associated with the center with the DTC capability and the wafer-to-wafer stacking capability. Right now, still within our current capacity size, there's no expansion planned, but there are a lot of customer interests and engagement being developed right now.

Okay. Great. Can you also give us some like idea how is that kind of business opportunity growing into the next few years?

I mean, as we anticipated, the cloud AI and the edge AI market will probably taking out in the next 2 years or so. And so we think preparing those technology capability today will position us well to serve that market when the market comes. I think many customers are engaging in that discussion and exploring the product roadmap at this stage. But in terms of the actual volume and the ramp-up schedule, I would expect it's going to probably be in late 2026 or sometime in 2027.

Next one, Sunny Lin, UBS.

Congrats on the very good outlook. Very glad to see business stabilizing and improving. So my first question is on the pricing. I understand more specific guidance should be provided in January or in early 2026, but I want to get a bit more color on the latest progress on your engagement with the clients. So in 2024 and 2025, basically, you provided roughly mid-single-digit type of price reset for across the board. And so how should we expect like going to early 2026? Would it be fair to assume that now given the improving supply/demand, even if any price decline should be lower than the magnitude in early 2024 and early 2025?

Well, I mean, this is definitely -- I mean, that's our goal, right? I mean -- but while we are still in discussion and aligning with our customers, I can't really quote that. I have to really see the data before I can comment about it. But throughout the annual discussions and the patterns in January, we'll probably continue engaging in similar discussion. But in terms of the magnitude of it, I think it's kind of too early to guide at this point.

Got it. Maybe a follow-up on blended ASP. There are still some concerns that there may be some overhang from LTAs expiring in the coming few quarters that could weigh on your blended ASP. And so Jason, could you maybe provide a bit more color on if any impact or that impact is already gone mostly?

I mean, LTA is one of the mechanisms that help us and our customers working other partners, not only based on the ASP, it's also we based on that, providing a mutual commitment for us to put in capacity to support the customer. At the same time, the customer demonstrate some commitment for the business engagement. So LTA will continue serving that purpose. Well, given the market dynamics, we're always working closely with our customers and to support them and gaining market shares without losing the market share and gaining the market shares and also with the market dynamics in terms of commercial needs, so -- but at the same time, we have to balance in terms of CapEx returns. So it is a complicated process and discussion, and we've been doing that for the past 2 years, and we'll continue supporting our customers to march into that direction, finding a win-win solution based on the LTA arrangement. But the future commitment of LTA remains intact, yes.

Got it. So maybe one question on 2026, just to make sure that I got the right number. So for 2026, Jason earlier, did you mention the target would be to grow business by double digit?

I mentioned about the 22- and 28-nanometer that we expect the momentum will go into 2026, and we expect a double-digit growth year-over-year, yes. For the...

Got it. And maybe a question on Singapore expansion. So if any like latest update that you could share with us in terms of how quickly the capacities will be ramped in 2026?

We -- I think the milestone has not changed. We project that the 12 IP3 production ramp will start in January 2026, and it will ramp up with a higher volume starting in second half of 2026. And that milestone schedule remains.

Got it. Maybe last question. So in terms of dividend policy, given the improving cash flow outlook in the coming few years, would the company consider maybe revisiting the dividend policy to change to like absolute cash dividend? Would that be possible?

It's not impossible, but we always try to strike a good balance between the high percentage payout ratio and absolute dividends. So I think that strategy or that position will continue.

Next one, Gokul Hariharan, JPMorgan.

So just wanted to understand a little bit more on the pricing. I know that you're in pricing negotiations with customers. Could we talk a little bit about 22 and 28? How is the pricing trend there? Do you expect that there is any concession that you may need to make on 22 and 28 pricing or that is going to be reasonably firm? And maybe also the same question on the 8-inch portion of the capacity as well, given some of your competitors are also kind of putting down or kind of exiting some of the 8-inch capacity?

Well, our pricing strategy has been very consistent, and we will work closely with our customers and -- for protecting and gaining market shares. So that remains. That will not change. So in the particular number, the ASP guidance, I think it's better that we have all the picture together and to share with you. But in terms of pricing strategy and positioning, that has not changed. We do believe that the pricing is a combination of our value proposition from technology differentiation, our manufacturing capability, reliable capacity and the diversified manufacturing locations, so on. So we think there's a lot to offer. And along with the mutual commitment with many of the customers, we believe that we will strive to a right balance for the pricing discussion. However, again, from the specific guidance on ASP outlook, I will probably prefer to wait until we finish up. I don't want to mislead you at this point. So -- but -- and that goal is whether it's 22- or 28-nanometer and as well as the 8-inch because each technology node has a different market dynamics, and we will work within that dynamics. Meanwhile, you're talking about if we see anything on the 8-inch opportunity or due to any other, our peers. We don't typically comment about our competitors. We believe our market share increase in 2025 in 8-inch, but not just 8-inch, overall 8-inch and 12-inch legacy nodes. And we believe those nodes remain a sweet spot for a wide range of analog reach products. So we'll continue to strengthen our product portfolio, focus on those spaces. And hopefully, we can increase our market shares. We continue to optimize our existing platform and developing a new solution to better address that market need. This is the area that UMC has built some long-standing relationship and trusted relationship with our customers. So we believe this structural trend will reinforce our position as the preferred foundry partner for customers in this needs. And that will actually help us to sustain our maybe growth in both 8- and 12-inch legacy nodes over the long run.

Got it. Yes, clear on the pricing that we can wait for January. But I think I just wanted to also ask on the semiconductor Section 232 tariffs. How are the discussions with your customers going? And let's say, there is a 15% to 20% tariffs on exports, which needs to be offset with any kind of U.S. investment or U.S. capacity that you have. How does UMC manage that situation? And which are the investments, or if any, that can qualify for that kind of an offset? I mean, for some of your peers, I think that is pretty clear. But I just wanted to understand how UMC is considering the situation.

Well, I kind of touched that earlier. Our -- we have been a very diversified manufacturing -- look, we have a very diversified manufacturing location in the past. And so we have very -- I think we're pretty much very complement to the current market dynamic. The current geographically discussion on diversification, supply chain resilience, I think our past initiatives serve that, and so we'll just continue. We may alter that, making some adjustment about that strategy, but not significantly. For instance, we're including building capacity in Singapore and U.S., and it's very much aligned to that direction. Of course, the tariff situation, whether it is X percentage, we don't know yet for Taiwan, but we know some areas already came out at 15% and which -- that's where we have our manufacturing sites. So customers are in discussion in interest of making sure that they have access to those facilities and to those locations. So we are definitely entertaining that conversation in a manner of growing our business engagement. So we hope that becomes more of an opportunity to us, not just a negative impact. Now, for some area that is not clear yet, and we have to navigate through that, it's our belief that we have very smart people in this industry. And despite how -- which direction it goes, we will navigate through this process and finding a win-win solution of mutual benefits.

Yes, just following up on that, Jason, I think geographical diversification is one aspect, but also the second aspect is U.S. capacity, right? So is your understanding that your 12-nanometer collaboration with Intel kind of counts as U.S. investment and U.S. capacity, given I think the total investment is actually quite small, even though you are actually shouldering a lot of the technology-related task.

Well, I mean, I can't comment about the big or small, but the investment is investment, and we are putting capacity in the U.S. And the starting point of the 12-nanometer only lays a solid foundation for us to explore maybe even other collaboration opportunity as well. So that also -- if there's anything to update, we will update you, but that could also represent even more investment, right? So -- but it's just -- we're not ready to update you anything yet. But even I look at the 12-nanometer today, that is quite significant in terms of investment.

Got it. Maybe one last question on the advanced packaging bit. I think you last time updated, I think, around 6K or so of wafer capacity for 2.5D IC packaging. Is that still where we are in terms of the capacity? And for your 2.5D packaging with deep trench capacitor, what is the application? Is it slightly different application that you're targeting compared to the mainstream market and that's why you're kind of waiting on the capacity expansion while the industry is still like really asking for a lot of capacity?

No, the 2.5D Interposer 6K today stays there. There is no expansion plan beyond that given the technology road map migrating to the DTC and we're developing the DTC capability. And for that, we're serving the AI, HPC, PC, notebook and smartphone space. And so our advanced packaging roadmap will center on the DTC going into, yes, 2026.

And would you say that the 6K is now fully utilized or you still have a lot of slack in that 6K capacity right now?

I mean, as the product is migrating to DTC, that's why we're not expanding the capacity on the 2.5D right now.

Okay. Okay. Fair enough. And this DTC capacity, how significant do you think it is going to become in terms of revenues? Let's say, I think you were expecting end of '26 ramp-up, so let's say, in 2027, is that a fairly significant part of your total portfolio? Or is it still going to be quite small, similar to the Interposer-related revenues that has been more like a single digit -- low single-digit kind of percentage of revenue?

I think it's kind of too early to predict that. A part of the market is associated with the edge AI market and which we have to wait until that has more clarity. And so I think at this point, it's too early to project that. But in terms of technology-wise, I think that's definitely the core of the next generation. So we need to make sure that we have prepared for it.

Next one, Janco Venter, Arete.

I just wanted to follow up on the investment into the U.S. and just get an update on the state of the PDK. And then also, we just want to understand the business model around this engagement on 12-nanometer. Is it revenue share? Is it profit share? And then just secondly, on that, will it be cannibalistic to the 22, 28-nanometer customers as you start migrating to 12-nanometer? Any color that you can add to that to help us just understand this opportunity would be quite helpful.

Sure. From a project standpoint, the -- currently, the 12-nanometer cooperation with Intel is progressing well and remain on track according to the project milestone. And we expect the early PDK will be ready for the first wave of customers in January 2026. And both UMC and Intel are aligning with the customer device spec to facilitate the ramp-up. Overall, the collaboration is proceeding as scheduled, and customer product tape-out is expected at beginning of 2027. So that is the update on the 12-nanometers. The business model itself, we are working collaboratively together and engaging with the customer and the actual business model that we're probably not elaborate to share right now. But once -- I think that will be a -- the business revenue recognition, once it's ready, we'll update that. And the cooperation model is actually very structured and -- but just we'll probably have to report that after we're into production. I think that's the 2 questions you have, right? Did I miss any?

Yes. That makes sense. Yes, that's right. Maybe just one follow-up. And I think you touched on this earlier where you talked about potentially looking at further investments. Now, if we look at -- actually, we were trying to understand if there's scope perhaps to extend this agreement to single-digit nodes because if you look at Intel's business, they fully depreciated 7-nanometer. And it seems like an obvious area to extend the agreement. Is this something that you would potentially be looking at? And does that make strategic sense for UMC?

Well, I mean, the -- yes, the simple answer is yes, right? And -- but we have to starting from -- we have to start it from the 12-nanometer. So we had to make sure that executed well, so we can lay a solid foundation on that. For technology beyond 12-nanometer, we are open to explore the future opportunity through the partnership arrangement that are mutually beneficial. I would say the cooperation with Intel is strengthening UMC's strategic position in U.S. significantly and for the U.S. market and also broaden our addressable market while adhering our disciplined CapEx approach. So we are very committed to this partnership. And so far, the project is actually progressing well.

Next one, Bruce Lu, Goldman Sachs.

Can you hear me?

Yes.

Yes. I just wanted to follow up the -- for the U.S. collaboration beyond 12-nanometer. What are the showstopper for us to move beyond 12-nanometer at the current stage? Or do we consider to go backwards to do like relative mature node capacity in U.S.?

I mean, that's an interesting question, right? I mean, the -- I think when we talk about this cooperation with Intel strengthened our positioning in the U.S. market, hopefully, we're not only limited at 12 nanometers and that if we can have a full potential of this position. And we -- so that's why we're actually very open to explore the future opportunities through this. So I don't think there's a -- I won't call any showstopper, but I think as long as it's mutually beneficial, I mean, we will definitely open to explore that. Now, is the exploration limited to the more advanced node or backward? I think we are also open to that. We're not limiting ourselves with that collaboration.

No, Jason, the question is that it's clearly mutually beneficial, right? So who has the ball? I mean, who doesn't want to move on?

I think, in any of the engagement, not just this, you require the market validation, you need to make sure you're doing your due diligence. So I think I will probably comment that all conversations are open and the due diligence need to be in place before we move forward. So it's not truly a showstopper. It's not going which sport, it is we have to make sure we conduct the appropriate process.

So in other ways, the prerequisite condition would be that you probably need to deliver 12 nanometers with like decent size of revenue, decent size of customer, then both sides might consider to move it on. Is that the right consideration?

Not -- I mean, I won't say that it is a prerequisite, but that is one of the important considerations. But more importantly is if this collaboration is economically beneficial to both sides. And so I think that the -- once we are more mature and ready, and we definitely will update you, but again, our position on this topic is we are open to that exploration.

Okay. So when can we expect to see the meaningful revenue contribution from 12-nanometer?

Well, I mean, right now, for the early product tape-out, it is going to be in 2027. And so we're probably going to start seeing some contribution in 2027, but then ramping after that though.

And in the interest of time, we're taking the last question. Last one, Charlie Chan, Morgan Stanley.

So it's actually wafer-on-wafer related. So, Jason, can you share with us who could be kind of memory partners? I mean, it seems like it requires a lot of so-called customized design interface, et cetera. So are those more Taiwanese partners or you have some global top memory partners for wafer-on-wafer? And secondly, if you can, can you share some potential kind of end applications and the timing for wafer-on-wafer?

On the wafer-to-wafer stacking capabilities, we are in mass production for some of the extremely small form factor devices in the RFIC space. We're talking about that because we believe if you look at the market is going, and we believe this technology will serve more than just a small form factor. It provides the option for the memory to memory, the logic to logic, logic to memory stacking options. So by providing the option to the customer, they say they can explore many different product applications. So at this point, the advanced packaging technology is developing into 2 cornerstones. One is the DTC capability. Another is on the wafer-to-wafer stacking capability. And then we -- once the technology is ready, then we can explore to many different applications.

On this wafer-on-wafer, do you see kind of advantage or differentiation to industry, for example, TSMC or China's -- I'm not sure, maybe XMC, yes, any sort of differentiation you may have?

Well, I mean, the developing differentiated technology is definitely on mandate. So we continue driving that technology differentiation. But at the same time, you have to make sure that you're part of the ecosystem, where the market is going. So we see this from a market standpoint. From a technology/product migration standpoint, we believe these are 2 very important capability and technology. So we're preparing ourselves to get that ready, and then, we can start exploring different business opportunities.

And ladies and gentlemen, thank you all for your questions. That concludes today's Q&A session. I'll turn it 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 [email protected]. Have a good day.

And ladies and gentlemen, that concludes our conference for third quarter 2025. We 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. Thank you again. Goodbye.