管理層發言
Welcome, everyone, to UMC's 2025 Second Quarter Earnings Conference Call. This conference call is now being broadcast live over the Internet. A webcast replay will be available within 2 hours after the conference ends. 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 second 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 second quarter financial results, followed by our President's key message to address UMC's focus and third 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 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 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 second quarter 2025 financial results.
Thank you, Michael. I'd like to go through the 2Q '25 investor conference presentation material, which can be downloaded or viewed in real time from our website. Starting on Page 4. Second quarter of 2025, consolidated revenue was TWD 58.8 billion, with a gross margin at around 28.7%. The net income attributable to the stockholder of the parent was TWD 8.9 billion and earnings per ordinary shares were TWD 0.71. Wafer shipment in quarter 2 increased to 967,000, up about 6.3% quarter-over-quarter. However, the effective NT dollar exchange rate also appreciated a similar magnitude from TWD 30.81 in Q1 to TWD 30.81 in Q2. And utilization rate increased from 69% in Q1 to 76% in quarter 2. Revenue as a result increased about 1.6% sequentially to TWD 58.75 billion. Gross margin, as we mentioned earlier, reached 28.7% or TWD 16.8 billion. This is already factored in around 3% of the Forex impact, 3 percentage points in quarter 2. Net income reached TWD 8.8 billion or 15.1% net income percentage rate. EPS is TWD 0.71 in the second quarter compared to TWD 0.62 in the previous quarter. On Page 6, for first half comparison, revenue increased by 4.7% to TWD 116 billion. Gross margin reached 27.7% compared to 33.1% in the same period of 2024. Net income attributable to the shareholders of the parent for the first half of 2025 was TWD 16.67 billion or TWD 1.34 in EPS terms. Cash remained over TWD 100 billion, reaching about TWD 111 billion at the end of the first half of 2025. And total equity for the company is now around TWD 337.4 billion. ASP edged up a little bit in the second quarter, mainly due to the better product mix. On Page 9 for revenue breakdown, there's literally no change on a sequential comparison basis. Europe increased to 8% and Asia reached about 67% IDMs up slightly to 19% compared to 18% in the previous quarter. In terms of application breakdown, the change is also very minor. Consumer went down to 33% by 1% and communication increased by 1% to 41%. Advanced technology revenue continued to increase and now revenue below 40-nanometer represents more than half of the total revenue, reaching 55% in quarter 2 when 22 and 28-nanometer represent 40% of the company's total revenue. On Page 13, the capacity breakdown, we will continue to see some minor capacity increase for the third quarter, capacity increase will come from mainly 12X Xiamen. And after the first 6 months, our CapEx budget for year 2025 remains unchanged at an estimate of USD 1.8 billion. And the above is the summary of UMC results for the second 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 the President of UMC, Mr. Jason Wang.
Thank you, Chi-Tung. Good evening, everyone. Here, I would like to share UMC's second quarter results. In the second quarter, the utilization rate increased to 76% as wafer shipment grew 6.2% quarter-over-quarter, primarily driven by communications in imaging signal processors, NAND controllers, WiFi, and LCD controllers. While we experienced an increase in the overall utilization and the growth of our 22- and 28-nanometer portfolio, the unfavorable foreign exchange movement of the NT dollar kept our gross margin at 28.7% by nearly 3 percentage points. Revenue from our 22- and 28-nanometer portfolio continued to grow sequentially, now accounting for 40% of total sales, a record high in both percentage and absolute dollar terms. Our industry-leading 22- and 28-nanometer solutions continue to win adoption by customers, and we expect to see further market share gains in wireless communications over the coming quarters. We have always believed that with the right differentiation, 22/28-nanometer is a strong and long-lasting node with a robust product pipeline. In addition, the new Phase 3 facility at our Singapore Fab 12i set to start production in 2026 will enable UMC to better serve customers seeking diversified manufacturing for enhanced supply chain resilience. Looking ahead to the third quarter, we expect a mild increase in wafer shipments. However, adverse foreign exchange movement will lead to a decline in NT dollar revenue. We are closely monitoring the near-term uncertainties and risks as the market anticipates U.S. tariff policies. To navigate macro and geopolitical headwinds, including foreign exchange risks, UMC will continue to actively manage our foreign exchange exposure and maintain financial flexibility to enhance our financial structure and business resilience. Now let's move on to the third quarter 2025 guidance. Our wafer shipment will increase by low single-digit percentage. However, NT dollar-denominated revenue is fully exposed to fluctuation in the foreign exchange rate. For instance, a 5% appreciation in the NT dollar will result in a corresponding 5% reduction in reported NT dollar revenue. ASP in the U.S. dollar will remain firm. Q3 gross margin will be approximately Q2 gross margin subject to the foreign exchange effect. Therefore, our Q3 gross margin will be approximately equal to that of Q2 under the assumption the foreign exchange rate is at the current level. 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.
分析師問答
Now first, we'll have Brad Lin from Bank of America for questions.
I have 2 questions. The first one will be on the ASP trend. So what's the initial outlook and view on the ASP trend into 2026, given the higher expense and cost? Obviously, we are happy to learn the stable ASP in near term. But yes, any initial view for 2026?
Typically, we don't provide guidance beyond 2025. While we can discuss the near-term outlook for average selling prices, it's important to consider a longer-term projection as well. Our strategy is to continue differentiating our technology and product offerings while maintaining and improving our average selling price resilience. We aim to create a greater distinction in our technology offerings and increase revenue from those specific nodes. After rolling out our 22- and 28-nanometer technologies, we will continue to provide specialty technology in the 40- and 55-nanometer range, although the revenue contribution in those areas will likely decrease as we compete with pricing from foundries. In the near term, our CFO noted a low single-digit increase in our Q2 average selling price driven by the higher mix of 22 and 28 nanometer products. We expect the product mix to remain stable in Q3, which suggests that average selling prices will remain steady for this year.
Got it. And my second question would be, well, we have seen, well, in the presentation, 14-nanometer and below mix are listed in the slide at 0 for a while but still listed in there. So should we expect the number to increase? And will that be from 12-nanometer or potentially also 6-nanometer?
The 12-nanometer technology is still a bit beyond our current scope. However, our collaboration with Intel on this project is progressing well and stays aligned with our project milestones. Both teams are currently focused on verifying the silicon performance for the pilot line, and we anticipate that the early Process Development Kit will be available for the initial group of customers by June 2026. Consequently, customer product tape-outs are slated to start in 2027. That's the status of the 12-nanometer project, and we will continue down this path. As for anything beyond the 12-nanometer, we do not have any specific plans right now. Our development efforts will concentrate on this technology, which will expand our specialty technology portfolio in both directions. This remains part of our roadmap, and we will provide more concrete updates when available. At this stage, the priority is to deliver a highly competitive solution for mass production at 12-nanometers in collaboration with our partners. For developments beyond that, we will look into future opportunities through partnerships that we believe will be beneficial for all involved.
Next one, Charlie Chan, Morgan Stanley.
My first question is about the tariff impacts on your customers' behavior? Do you see kind of pull in? And what does it impact your second half sustainability or outlook?
Sure. We do observe some demand upside in Q2 and as well as Q3, partly driven by the inventory buildup in anticipation of potential U.S. tariffs. And so for UMC's first half '25 results, which is in line with our guidance as the Q2 wafer shipment increased 6.2% quarter-over-quarter. While the Q3 demand increased on a higher base, we expect the shipment will still grow mildly sequentially. And so there are some observations about that. But given the 2025 market dynamics, such as the adjustment to U.S. policies and ongoing geopolitical and macro uncertainty, the usual seasonal pattern may be different. Along with our customers, we're closely monitoring those end market signals.
I see. And I think lots of discussion about the future advanced packaging technology. So Jason, can you share with us about your business development here? And also, I think you have some interposer capacity, how are we going to utilize those capacities going forward? And maybe some color about the potential applications.
We are focused on capitalizing on the advanced packaging opportunity. We are preparing our advanced packaging solution to meet the increasing energy demands of cloud AI and the anticipated growth in the edge AI market. To address the power efficiency needs of high-performance processors, UMC is developing a 2.5D interposer with DTC and discrete DTC, which will form our upcoming roadmap. The current interposer is progressing to the next generation, and we are eagerly awaiting its introduction, expecting to ramp up production thereafter. Additionally, UMC is utilizing scalable 3D wafer-to-wafer stacking and TSV to boost the competitiveness of our specialty technology. We are currently in mass production for the ultra-compact form factor used in 5G and 6G RFIC devices. Following the success of the 5G and 6G RFIC with wafer-to-wafer stacking, we are also working on memory-to-memory and memory-to-logic stacking solutions to address high-bandwidth computing needs.
Okay. My last question is about your comments on the semiconductor cycle. This is the third consecutive year without a recovery in the second half. What is your perspective on what's happening in the semiconductor industry? Why aren't we seeing the usual seasonality or cyclicality? In the past, we experienced cycles of upswings, shortages, overcapacity, and then corrections, but that seems absent now.
The visibility is currently lower than before. At the beginning of 2025, we anticipated that our growth outlook for that year would be slightly better than our addressable market, which we believe will grow gradually at a low single-digit rate. We still expect our growth outlook for 2025 to remain the same. However, for 2026 and beyond, we need to work closely with our customers to understand their visibility and monitor the DOI situation. As of now, the DOI seems to be improving, reaching a healthier level about one or two quarters ago. The computer, consumer, and communication segments remain strong, while the automotive and industrial sectors are still high. By keeping an eye on macroeconomic factors and the DOI, we can hope for an eventual upcycle, though visibility is quite low at this time.
So maybe tied again Brad's question about wafer pricing. So yes, because obviously, FX impact over Taiwan, Taiwan foundry a lot in terms of gross margin. Would there be a factor you can put on the table to negotiate with your customers for next year's pricing?
I mean we continue working with our customers in terms of pricing compensation closely, but those are more of a tactical conversation. I think fundamentally, like I reported earlier, our key focus is to try to differentiate our technology offering and ensure that we can continue to enhance our product mix to improve the ASP resilience. And that's what we're marching towards. And we have a very clear roadmap today on many fronts of our technology development. So our goal is to further widen the gap in technology offerings and increase the revenue contribution from those respective nodes and technology offerings, which we think that we can ensure that our ASP can remain resilient.
Next one, Gokul Hariharan, JPMorgan.
First of all, for the Singapore fab 28-nanometer and 22-nanometer expansion, could you talk a little bit about what is the current pace of the ramp-up and the kind of customers that you're ramping up there? Obviously, some of the pricing negotiations that you had back in '22 and '23 had some price escalators. Could you talk a little bit about whether those price escalators still exist given the environment has definitely changed somewhat? So that's on the 28-nanometer part, yes.
Sure. Well, for the 12i Singapore facility, given the current market dynamics and customer alignment, we project the 12i Phase 3 production ramp will start in January 2026. And it will ramp up with a higher volume starting in the second half of 2026, and that's the current ramp plan. Many of this ramp schedule and alignment is based on the customers' close communications. And right now, given the application ramp-up, it's going to be mainly in the communication with our 22-nanometer high-voltage devices, and we still believe our 22 and 28-nanometer high solution are differentiated from the market. And so the ASPs still remain very healthy at this point.
Got it. Secondly, regarding gross margins, we are currently utilizing around mid-70s capacity, while our gross margins are in the mid- to high 20s range. Depreciation has started to increase again and is expected to continue growing over the next couple of years with the introduction of 12i. Could you discuss the realistic pathway for us to return to the mid-30s or low to mid-30s gross margins we've previously mentioned? While currency fluctuations are beyond our control, could you elaborate on other factors? Is aiming for that kind of goal realistic? Considering previous questions, can pricing effectively help us achieve that, or is it more difficult to leverage price as a strategy?
Absolutely. I mean it's definitely our mission to continue to improve the gross margin back to a reasonable level. Given the current loading is fluctuating around the 70%, that's definitely putting some pressure in terms of the gross margin, while the depreciation increases. And so the focus is very clear. I kind of answered Charlie earlier that we are focused on technology development, technology offerings, even new technology offerings, and partnership engagement. With the product mix improved, we think that we have a path going back to a reasonable level. For the past, we have maintained our foundry shares in our addressable market segment. Based on our current design pipeline, we anticipate more share gains in 2026 as well as going into 2027, particularly in the 22- and 28-nanometer bucket today. Now while we rolled out the other technology offerings, we think this will continue to improve and we will definitely march in the direction to go back to the right level of the gross margin level.
I want to add that our annual depreciation growth is expected to peak. In 2023, our depreciation expense rose by over 20% compared to the previous year, and a similar increase is anticipated for 2024 and 2025. However, for 2026 and 2027, the growth rate is likely to be much lower, possibly falling to single digits. We are optimistic that this will lead to a better cost structure as we move into 2026 and 2027.
Maybe one more question on the high-voltage side for 28 and 22. Jason, do we have a pathway below 22-nanometer for high-voltage given there's been some discussion about some of the driver IC-related products moving below that, be it to some kind of a FinFET node, but enabling high voltage?
It's definitely on our roadmap today. We still believe the 22 High D will be the most compelling and competitive solution now and in the coming years. The FinFET solution for high voltage is also part of our plans.
And any timeline in terms of when you think customers will start demanding this?
We are still aligning with our customers and considering the value proposition of the 22-nanometer technology compared to the next generation. We are working closely on this, but I don’t have a specific time frame and I prefer not to estimate at this point. Based on the data we have, we anticipate that the 22-nanometer high voltage technology will experience a delay, likely needing another year to up to two years.
Understood. Maybe one last question. Several of the consumer fabless companies are guiding down Q3 quite meaningfully. Your own wafer orders are slightly moving up in Q3. Should we expect that there could be a pickup in Q4? Like every year seems to be a different seasonality, but just wanted to understand how you think about that inventory cycle for many of the Asian consumer fabless companies, which are your key customers as well.
Sure. I mean the inventory situation actually is quite healthy with still a major segment already. And auto and industrial, I think they're still kind of high, but the rest is actually quite healthy. At this point, given the visibility, we do not guide Q4 at this time. But our view for the full year 2025 will remain unchanged. And again, I kind of touched that earlier that we expect our addressable market will grow by low single digits, and we will still outgrow the addressable market in 2025. And the biggest challenge nowadays is really the visibility given the macro uncertainties and the geopolitical concerns. I think the customer is being cautious. It doesn't mean that they don't have demand. The question is they kind of want to play this thing in a different manner. So we're working closely with them. And meanwhile, Q2 is growing, Q3 slightly, sequentially. And Q4, we just have to play and see, and we will definitely report that in next quarter. But meanwhile, we've seen the overall 2025 projection, it's still unchanged.
Next one, Sunny Lin, UBS.
So my first question is on 28-nanometer. So if we look at Q2, Jason, what's driving the revenue upfront? Is it driven by the 22-nanometer migration? Or is it through a product mix upgrade? And then looking ahead, could you share a bit more on your share gain in wireless communications and maybe some of the other products going to 2026?
For the near term, the revenue increase from the 22 and 28 segments is primarily attributed to the communications segment in Q2, with a strong emphasis on communications. We are very confident in the ongoing growth of our 22- and 28-nanometer business in 2025 and beyond, extending into 2026. The positive demand outlook is bolstered by the continuous tape-out momentum across various applications, thanks to our customers, and is further enhanced by UMC’s unique technology and regional manufacturing capabilities. This encompasses our 12i fab in Singapore, where the P3 fab expansion is progressing as planned, and we expect to start ramping up in 2026. This will begin contributing to revenue in the latter half of 2026 and will further bolster our capacity for 22 and 28 nanometers to meet rising demand. The combination of our technology offerings, manufacturing quality, and strategically positioned capacity will ensure that our 22 and 28 nanometer segments continue to be the main growth drivers through 2026.
So on 12i, would you be able to price the wafers a bit higher, given a higher cost structure? And when you talk about high volume production starting from the second half of 2026, any type of capacity that we should expect?
We don't want to specify the exact capacity size, but we are rapidly increasing our P3. Our combined capacity for 22 and 28 at our facilities is strong, and previously, the utilization rates for 22 and 28 exceeded the corporate average. Even now, they remain above the corporate average. I'm sorry, but I didn't catch your earlier question.
Pricing for Singapore. Would you be able to price a bit higher given cost is higher as well?
I missed that point, and it's a sensitive topic. Currently, our pricing strategy is anchored in our technology and the value we provide. This serves as the baseline for our average selling price. We need to collaborate with our customers to understand their requirements, ensuring they remain competitive while acknowledging the unique advantages of our technology and its geolocation benefits. We'll discuss this with our customers, focusing on our technology's differentiation and their competitive edge.
Got it. That's helpful. I have a question on the Intel partnership. Seems like Intel is becoming less proactive in pursuing the foundry ambitions with the new management. So I wonder how does that affect the business development with UMC? And let's say, if Intel wants to scale down and then look to maybe sell the capacities, in that case, would UMC be interested in acquiring the capacity, assuming the price is reasonable?
It's difficult to address any speculation at this moment. I prefer not to comment on the partner's priority within the company. However, I can share insights about our program. The collaboration with Intel is progressing very well, and we are on track to meet our established milestones. Importantly, both parties are fully committed to this 12-nanometer cooperation. I foresee no changes right now, and we maintain very high expectations for this program.
Next, we'll have Laura Chen from Citi. I'm sorry, Laura just dropped the line, so we'll take the next one, Jason Zhang from CLSA.
Just wanted to follow up on the impact from the FX ratio. Can you provide your FX ratios for Q3?
So first of all, every 1% move appreciation of NT dollars against U.S. dollars, it will erode our gross margin of about 0.4 to 0.5 percentage points. That's where the 3 percentage point erosion comes from on the back of the 6% plus NT dollar appreciation against U.S. dollars. And for Q3, we don't do forecasts, but we are using the current Forex rate, which is nearly 29.8 when we give out our guidance. And a reminder, for quarter 2, the weighted average was 30.81.
My second question is in terms of the competition. It seems like your Chinese competitors now have better or higher utilization currently. So do we see better market or lower competition in the mature nodes? And how can UMC benefit from this lower competition?
Well, at this point, more than half of our revenue comes from specialty technology solutions, which serve our customer demand in differentiated technologies. For instance, our 22/28, I kind of touched on earlier, is probably the most competitive solution in the high-end smartphone OLED display market. In addition, our 22 ultra-low leakage and low-power technology will deliver another 30% to 50% better power saving compared to standard 28. So we are positioning ourselves as a specialty foundry partner focused on low leakage, low-power logic, embedded high voltage, BCD, embedded non-volatile memory, RFSOI solutions. We want to continue to provide specialty technology where the percentage of revenue contribution in this space will increase and the percentage of the revenue contribution competing with the Chinese foundries will continue to decline. I think that's our focus. And I think that we've made quite a bit of progress already, and we think there's more room for us to improve on that.
Next one, Laura Chen, Citi.
Just a quick follow-up. I want to understand your view on the long-term gross margin outlook. We understand that there's a lot of moving parts, rising depreciation, and also currencies, etc. But we do see that recently the utilization rate is kind of improving back to high 70%. And as we're moving into Q3 with the wafer shipment also going up, so what's our view on our so-called long-term gross margin target? If you can give us some more color on that.
Currently, a mid-70% utilization rate is not ideal. Improving loading efficiency will be a key priority. To enhance loading, we need to offer competitive solutions to our customers. Our focus will be on technological differentiation, developing new technologies, and engaging with key customer partners. By doing so, we believe loading will improve, which will also contribute to healthier gross margins. We have seen a significant increase in depreciation costs over the past couple of years, and after 2025, we expect the rate of increase to stabilize. As we enhance loading and manage costs effectively, we will also focus on managing average selling prices. With better solutions and a diversified manufacturing base, we anticipate that our branded average selling prices will remain strong. Additionally, we are committed to progressing with our 12-nanometer development and improving our product mix. All of these strategies will help us strengthen our market presence and financial performance. We have made strides in enhancing our structural profitability and breakeven point, and we are already seeing positive effects. However, there is still work ahead. Combining all these efforts, we believe we have a clear plan to achieve better results.
Chi-Tung, can you also remind us what will be the depreciation cost increase for this year or maybe next year?
This year is low 20% year-over-year. Next year is still a very rough estimate. But as I mentioned, the magnitude of increase will decline significantly maybe to below 10%.
Okay. My next question is also about our operation in China. As we know, we still have 2 fabs in China. Even though there's always very fierce competition, do we see any possibility that our IDM customers, if they want to, like entering the Chinese market, they can also leverage our capacity there, thus to be kind of differentiation as well? So can you give us more updates on your current strategy in China?
Well, I mean, first of all, with our diversified manufacturing sites, we'll definitely be able to serve different customer needs. And if there is a customer who needs their product to be produced in our China facility, that's something that we very much welcome. The same thing that we have a customer moving from China to other locations, and we very much welcome that. And we believe with a diversified manufacturing offering will give us the benefit of supporting customers with their supply chain resilience needs. Right now, for the IDM customer moving into the China facility, certainly some signals, but I think the signal goes bilateral multiple different ways. And so we are working closely with different customers. And hopefully, we can fulfill their desired needs.
Next one, Tim Schultz indiscernible.
I had 2, please. The first one is on pricing behavior and particularly just how rivals are behaving in terms of pricing in the communications segment. Is that disciplined pricing, particularly given the steady improvements in days of inventory? Or is pricing more challenging? And then I had a follow-up.
When there's ample capacity available, pricing becomes a CapEx topic. So not until the capacity becomes tightened, I think the pricing will always be a topic. And so I think from a behavior standpoint, it's really subject to the capacity situation. So given that the current capacity situation in different regions is different, I think that conversation is still quite often.
Okay. That's very helpful. The second one was in terms of the collaboration with Intel. Good to know that the PDK 2026 production, 2027 is still on track. I had a 2-parter there. It's just in terms of the work you're doing with your partner, do you see any impact from the headcount reductions? Does that influence that cooperation in any way? And then the second part, talking about gross margins and the outlook in '27, '28 this journey to get back into the 30s, obviously, loadings are the most critical factor. But does this cooperation with Intel play a material part in your medium-term gross margin outlook?
Yes, in terms of absolute dollar amounts, it will. Given our business model and the commitments from our partners, the outlook is quite positive. The program is expanding from R&D development into preparation for high-volume production, which has led to increased involvement from various organizations. While I cannot comment on the second part of your question, I can say that we are witnessing much more activity from different departments as we transition from R&D to high-volume production. This means that, as we broaden the scope of our activities, we are seeing greater engagement with the program today.
And now we are taking the last question, Alex Chang, BNP.
I only have one follow-up question regarding to your China business. So can you comment like in terms of utilization, how is your China fab utilization versus the overall utilization? And in terms of the price pressure, have you seen it eased in recent months? Or what is the outlook for the price pressure in China?
The 12X facility is currently operating at full capacity, which is higher than our corporate average. As our various sites primarily function as manufacturing facilities with centralized business management, there is currently no pricing differentiation among our locations.
Ladies and gentlemen, we thank you for all your questions. That concludes today's Q&A session. I'll turn it over to UMC Head of IR for closing comments.
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 ir@umc.com. Have a good day.
Thank you. And ladies and gentlemen, that concludes our conference for the second quarter of 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.