© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Welcome, everyone to UMC's 2024 First 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 two hours after the conference is 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 first quarter of 2024. I am joined by Mr. Jason Wang, President of UMC; and Mr. Chi-Tung Liu, CFO of UMC. In a moment, we will hear our CFO present the first quarter financial results, followed by our President's key message to address UMC's focus and second quarter 2024 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 risk 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 filing 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 first quarter 2024 financial results.
Thank you, Michael. I would like to go through the first quarter '24 investor conference presentation material, which can be downloaded or viewed in real-time from our website.Starting on Page 4, the first quarter of 2024, consolidated revenue was TWD 54.63 billion with gross margin at 30.9%. The net income attributable to the stockholder of the parent was TWD 10.46 billion and the earnings per ordinary shares were TWD 0.84. Utilization rate in first quarter of '24 was 65% similar to 66% in Q4 of last year, however, the shipment has increased by about 4.5% sequentially.On Page 5, for the sequential financial comparison, revenue declined slightly to 54.6% -- sorry, TWD 54.6 billion. Gross margin was down 5.1% to 30.9 percentage points or TWD 16.899 billion. The operating expenses normally in first quarter is a seasonal low point. Therefore, we can see the operating expenses was down 13.4% to TWD 5.7 billion in Q1 '24. And our other operating income mainly is subsidies from government declined quite a bit to TWD 513 million in Q1. This is largely coming from our Xiamen operation. Their government subsidies recognition is in line with the depreciation curve, which has come down significantly in 2024. And overall, net income attributable to the shareholder of the parent was TWD 10.4 billion in Q1 versus TWD 13.1 billion in Q4 of last year. EPS was TWD 0.84 for first quarter.On Page 6, the year-over-year comparison, revenue also stayed similar range, almost a slight increase of 0.8%. And gross margin, however, declined from 35.5 percentage point to 30.9 percentage point in Q1, mainly due to increase in costs such as depreciation expenses. And for the non-operating income is also a big difference mainly due to our portfolio holdings, investment holdings. This is the mark-to-market gain, it's only about TWD 10 billion -- sorry, TWD 1 billion in Q1 versus TWD 4.6 billion in the same period of last year.On Page 7, our cash is now about TWD 119 billion, and our total equity is TWD 378 billion. Most of the increases in PP&E, property, plant and equipment, which right now stand at TWD 254 billion.On Page 8, there's a onetime annual adjustment in our ASP in Q1 of 2024, which was also the main reason offset the 4% to 5% increase in wafer shipment in Q1. So the magnitude is quite similar to that in the first quarter ASP.On Page 9, the original breakdown of our revenue stayed relatively similar quarter-over-quarter. Europe declined 3% from 11% in Q4 last year to 8% in the first quarter of this year.On next page, Page 10, there is also a big change in the IDM composition versus fabless revenue. So this quarter is 18% versus 82%, while last quarter was 22% versus 78%. On Page 11, the application breakdown remained relatively stable.On Page 12, we see a seasonal downward adjustment in some of our customers, which lead to a small decrease in our 22/28 revenue percentage point of 33%. And the rest of the technology geometries are relatively stable.On Page 13, our capacity breakdown in 12-inch equivalent capacity, most of the increase is coming from 12A in our Tainan fab, which is our P6 expansion, and there will be some spotty areas of efficiency improvement for some of our other fabs. Total 12-inch equivalent capacity is TWD 1.2 million in Q1 this year. Our CapEx after first quarter remain unchanged, still stay around TWD 3.3 billion cash-based CapEx for 2024. Majority of that will be attributable to 12-inch capacity expansion.So the above is a summary of UMC results for Q1 2024. 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 first quarter results.In the first quarter, our wafer shipments increased to 4.5% quarter-over-quarter as we saw a pickup in the computer segment. Despite a slight drop in the utilization rate to 65%, we were able to maintain relatively healthy margins due to a continuous cost control and operational efficiency efforts. Contribution from our specialty business increased to 57% of total revenue, driven by demand for power management ICs, RFSOI chips and silicon interposers for AI servers.During the quarter, our teams continue to make good progress on key pipeline projects, both customized solutions for customers as well as new technology platforms to serve high-growth segments within the 5G AIoT and automotive markets. This includes embedded high-voltage, embedded nonvolatile memories, RFSOI and 3D-IC solutions. In line with our policy to provide a stable and predictable dividend to our shareholders, UMC's Board of Directors recently approved the shareholder cash distribution of approximately TWD 3 per share, which will be a higher payout ratio than the previous year. This is subject to approval by shareholders at the Annual General Meeting in May.Looking ahead to the second quarter, we expect to see an increase in wafer shipments as the inventory situation in the computing, consumer and communications segment improves to a healthier level. As for the automotive industrial segment, demand remains mute as the pace of inventory digestion has been slower than anticipated. While we still expect some lingering impact on macro uncertainties and cost headwinds in the near term, UMC will continue to invest in technology, capacity and people to ensure UMC is ready to capture the next phase of growth driven by 5G and AI innovations.Now let's move on to the second quarter 2024 guidance. Our wafer shipments will increase by low single-digit percentage. ASP in U.S. dollars will remain firm. Gross margins will be approximately 30%. Capacity utilization rate will be in the mid-60% range. Our 2024 cash-based CapEx will be budgeted at USD 3.3 billion.That concludes my comments. Thank you all for your attention. Now we are ready for questions.
[Operator Instructions] First question, we'll have Randy Abrams, UBS.
The first question I wanted to ask on the 28-nanometer, where you mentioned it seasonally dipped in first quarter. Could you talk about the outlook, like if you see that recovering and how do you see the P6 as you bring up the rest of that capacity? How do you see application to build that capacity or keep it loaded?
For Q1, the 22 and 28, we saw a decline and because we're experiencing smartphone seasonality, which lead to a lighter 28 and 28 loading. We do expect to pick up the 28-nanometer wafer shipment in Q2 2024. The 28 loading will remain at a relatively healthy level based on current projections and which is supported by product in OLED drivers, largest technologies such as ISP, Wi-Fi and SOC process applications. After the Q1 2024, we will see higher wafer shipments driven by the demand from both communication and consumer segments.
Okay. And the blended pricing is firm overall, but you do have the mix shift coming back to 28. Could you talk about the like-for-like pricing environment if you see next few quarters negotiation? Would there be any like-for-like pressure or do you expect it would be like first quarter each year would be the time the bigger negotiation happens?
Well, I mean, that's the typical practice. And as we mentioned in January call, the ASP will remain firm after the one-off pricing refresh in Q1, and we are still expecting that. The company's pricing strategy has remained consistent, which is based on our value proposition. And at this time, we still expect the pricing will remain firm in the second half of 2024. However, we do believe the -- in terms of the like-to-like pricing adjustments and so on, we believe the focus should be elevating our customer's product competitiveness and to help them win more market share. So while our pricing strategy has remained consistent and align with our value proposition, but it also includes to stay competitive and resilient against the market dynamics, so then we will continue monitoring the market dynamics.
Okay. So it sounds like broadly firm, but some -- the flexibility were needed if it's -- like if it's a competitive situation.
That's correct, yes.
Okay. And to revisit the CapEx, I think you mentioned last quarter, majority of it was either residual for P6 or infrastructure. So lot of the equipment spend is still to come, but I think you were planning it to start up sometime around Q2 of next year. So is there a way to think about how much capacity in the first phase? And have you made that decision to bring up like a first phase, like if you have the amount or how much you might need to outlay? I would think it would be mostly in 2025.
Are you referring to P6 or P3?
Sorry, the Singapore. Actually, I was referring to -- because you're doing the construction of Singapore, so how much you bring up in Singapore in next year?
Yes so well, first of all, you're right. In last quarter, we did mention among the 2024 CapEx number, around 60% of the 2024 CapEx will be spent on the 12i P3 infrastructure and as well as some minimum tools and equivalents. So they are -- we are spending on the P3, but mainly for the infrastructure. And the -- given the current market dynamics and the customer alignment, we are projecting the 12i P3 production ramp will actually start in January 2026 now. And the P3 will ramp up with high volume starting from the second half of '26.
Okay. Does that imply like the big CapEx this year? Would you still -- it sounds like most of the equipment have been pushing to maybe some next year, some actually in 2026, so we should think -- I mean, is there a rough way to think about it because it would affect -- like depreciation has been rising, but that could maybe level it out with that push out?
Right. I mean, we are -- we certainly are managing that. We are projecting CapEx will peak out this year and will not impact the company cash dividend policy first. I believe that people care about that. As we have stated in the past, our CapEx strategy continue to remain disciplined and ROI driven and also along with our affordability. So that means we are managing that, but the CapEx will be peak out at this year, the 2024.
Yes. And I did see you have the TWD 3 dividend that you confirmed. Actually, the last question, it ties to the AI. Like a lot of the attention goes to the high-end, like the . But you did put in your remarks kind of a reminder on the -- you have the silicon interposer. But could you kind of outline where you still have opportunity, like that seems like the strongest -- still the strongest momentum driver, like where UMC can participate like [ silicon interposer ] or other components into AI or HPC?
I mean, we certainly would like to export more, but not yet. Well, at this point, while the recent focus of AI chip has been on the most advanced computational chips to run the AI module and the chips will require handling data transmission and power management are also equally important. And so UMC's focus is on those 2 areas, data transmission as well as the power management. Specifically, the AI server will near high-speed IO chips and memory controllers to handle the data transmission and the power management IC for every computing and memory units to optimize the power consumption. Similarly, technology offering from 55-nanometer, 12-nanometer, specialty technology, non-volatile memory, 3D-IC and they are all well suited for the Edge AI applications such as wearables, smartphone, et cetera. So in that, our solution will offer to enable those Edge AI devices to achieve the optimal balance among comp performance, [indiscernible], power efficiency, and we'll continue to work closely with the customers to bring those innovative Edge AI solution to the market.Well, I mean, like I said at the beginning, while we are still in the early stage of [ portfoliating ] the Edge AI, we are recognizing the growth potential of the AI market, and we are ready to capture those emerging opportunities. And based on the -- even we are early, but based on the current projection, we are expecting UMC's addressable market within the overall AI semiconductor market will be around 10% to 20%.
Okay. So you will capture 10% to 20% of the AI TAM with your portfolio, okay.
Randy, actually, that's more addressable.
Address, I should say not capture it yet. Solutions can target and then [indiscernible] that makes sense. And to clarify, too, on the silicon interposer, do you plan to expand from the 6,000? I think that was the last guidance you gave, 6,000, if you would add further capacity to that?
Yes, that's the current 2.5D interposers. Let me maybe elaborate a little bit more on our 3D-IC space. In addition to the interposers, our 3D-IC offering includes the wafer-to-wafer hybrid bonding, active intervals with TSV and the 2.5D interposer with TTC. And so the interposer is the one of the offerings and which that's where the market is today. But we are cautious in monitoring the expansion of the current interposer capacity and continue to focus on the expanding on the rest of the solutions. Those solutions aim to provide a cost-effective performance efficiency alternative for semiconductor devices through a vertical stack of silicon wafers or dies, which deliver small form factors, enhance the bandwidth and lower power consumption for various applications, including the Edge AI and the data center and communications. So we are cautious about the current solution of interposer expansion, but we aim for the future expansion for the interposer with TTC and other interval with TSV and as well as the wafer-to-wafer hybrid bonding.
Next one, Gokul Hariharan, JPMorgan.
My first question, Jason, I think last call, you had mentioned that your growth rate expectation for foundry is about high single digits and expect UMC to kind of grow at a similar pace or strive to grow at a similar pace. Is that still your expectation right now, or has anything changed given the slower automotive industrial recovery? Has anything changed on the numbers?
Gokul, yes, you remember well. And yes, let me give you a bit of an update. There are some changes. One, some area we did not change. For instance, we still project the semi industry will grow in the mid-single digits. That did not change. For the foundry, you will grow at the low-teens year-over-year and that is changed. However, majority of the 2024 growth in the foundry will be driven by the AI servers. We actually continue to monitor that. I think the biggest momentum is coming up on the AI servers. And therefore, the growth for the UMC addressable market now remain flattish for 2024. However, you're right, our intention is still to expect to outperform our addressable market, but I think the current addressable market projection for us is flattish for 2024.
Understood. That is quite clear. Second question is on some of the specialty projects that you have on RF-SOI that you called out and also the high-voltage driver ICs and OLED. Could you talk a little bit about your market presence, especially in RF-SOI, how big this is? Because it seems like you're starting to gain some market share from the current market leader on some of these platforms. So could you talk a little bit about the specialty platforms, especially RF-SOI and OLED driver IC?
Sure. I mean, first of all, we will continue to address all specialty technology, not just limited at RF-SOI. However, we see a good momentum from our RF-SOI market penetration. We are seeing a significant market share gain in that space. However, I don't have a specific number to share with you right now and the -- I may be able to update you next time. The -- well, for the non-volatile memory, and the -- we also gain some traction on that, while we have been maintaining our market position on the embedded high voltage.
Understood. One more question I had is on the -- as we are kind of moving towards spending primarily for the fab 12i P3, are there any CapEx offset that we will potentially get for the Singapore fab, or is it mostly the prepayment from customers and the government incentives are mostly on tax breaks and other kind of profit driven subsidies?
. Maybe Gokul, if you can repeat that question again?
Yes. So I'm asking for Singapore fab, do we get any capital offset, like CapEx offsets that offset our CapEx, or is it mostly going to be like tax breaks and other kind of subsidies, which kick in once you start production?
Yes, we cannot go into detail, but there are government incentives issued by the Singapore government, including both tax breaks as well as the subsidies on capital expenditures. And the overall package is equal or better than the -- our previous investment, the P1, P2 because of the more advanced technology investment we have in Singapore. So we are very grateful for the strong support from the Singapore government. And the enlarged hub in our Singapore operation actually will enable us to receive even more benefit as of this greater economical scale.
So the TWD 5 billion spend that we had projected earlier that is more like a gross CapEx number, we should assume?
Yes, roughly for P3, roughly.
Okay, understood. And one more question I had is on 8-inch, given 8-inch seems to be still pretty sluggish. Are there any plans on flexibility for the 8-inch capacity? Is there any plans to potentially convert 8-inch to other areas, the compound semiconductors, silicon carbide or any of those areas?
Well, first of all, we are continuing to anticipate pressures from the -- some of the 12-inch mature node fab that has impacted 8-inch supply chain. So we definitely need to address that. While certainly the main string application will remain an 8-inch node, we also try to expanding the technology offering to silicon carbide, gallium nitride, but they're still in a very early stage. Most of the silicon carbide today is at a 6-inch, and we are more focused on an 8-inch migration, and we can update more progress once that becomes mature.
Next one, Brad Lin, Bank of America.
I have 2 questions. One is on the silicon interposer or the advanced packaging. We have learned that the management is cautious or less active in expanding this part of the capacity. So we are on track to hit around 6k per month and no further expansion, should we assume that? And then a follow-up on that is that, we know that we are expanding the other wafer-to-wafer technology and other 3D solutions. So when should we expect this product business to well take off?
So first question is the interval. So yes, we will maintain that 6k capacity and we continue seeing that we had a stable run rate with -- for 2024. And for the others, some of the RF-SOI applications already starting to use the wafer-to-wafer hybrid bonding solution. So they will be ready for production in 2024 as well. And so the -- we gradually introduced some of the advanced die-to-die and wafer-to-wafer stacking solutions and gradually increase that capacity as well.
Got it. Glad to hear that. And then my second question will be on the -- well, our node migration. We have learned a good business on the 22 and 28 nanometer with the OLED driver IC and some products migrating to 22 and 28 from 40. And we would like to learn the outlook of 40 and the well utilization outlook and what products can we expect to help fill the 40-nanometer capacity gap?
So you're talking about 40 or 14?
40, four-zero.
For the mature 12-inch, the utilization rate for both 40 and 65 will remain flattish for the second half of 2024. And of course, this still highly depends on the pace of the overall market recovery, and we closely monitor that. At the same time, there are applications still coming into the 40-nanometers such as the RFSOI and as well the non-volatile memory. They are some of the new application will come in adopting the 40-nanometers.
Got it. Actually, I also wanted to ask about 14, one-four. So despite a limited capacity currently, there is still emerging business opportunities in the market. Does the firm plan to allocate more resources for the expansion on this 14?
Our current focus on the FinFET is on the 12-nanometer cooperation with the U.S. partners, which the program has kicked off at the beginning of this year. And the focus is try to -- making good progress on the co-development and which is on track. And the -- and we're also seeing a high level of interest, and we see numerous increase. At this moment, we are working with our customers to accelerate the ramp schedule for the 12-nanometers. So yes, I mean, the focus will be shifting to the 12-nanometer instead of 14.
Next one, Bruce Lu, Goldman Sachs.
I want to ask about the -- Jason, your view about the high [indiscernible]. If you look at your guidance, you're guiding for a flat to slight up for the total revenue, which is a sequential growth for your quarter revenues like, most like, it's flattish or slightly down every quarter for the next coming quarters. And your margin is somehow flat for 30% if there is no more ASP erosion. So which is -- there is no recovery at all for the industry. So it seems to me that this inventory [indiscernible] cycle is a lot longer than expected. So when do you see the inventory correction will -- the restocking will start to kick in and provide some minimal health for your business.
I mean, Bruce, the first off, let's talk about inventory. We have noticed the overall industry inventory level are continuing to improve. Specifically, we have seen inventory getting to a healthy level in the computing, consumer and communication market centers. However, we still observe our customers taking more of a conservative approach in their inventory restocking behavior. In other words, we've seen more of a rush order instead, the confident projection going forward. I think that mainly because the biggest uncertainty is still at overall macro outlook, which could impact the market dynamics.The other 2 market segments for the inventory digestion for the auto industrial segment is still slower, like I mentioned earlier. So I think the market is prudent. The customer are prudent about the market outlook. We do see the overall industry are recovering. And so we do think the foundry industry will grow in the low-teens percentage year-over-year. However, most of the resources is more allocated to the AI server. So I think the AI server will be the -- with more of a higher growth rate versus the rest of the market applications. So -- and since we have less exposure at the AI server, so I think our projection at this point is more conservative. However, we are optimistic about the inventory, the situation.
So if the AI continues to be strong for next year, you will continue to [ renew it ], is that the base assumption?
I mean -- well, the other things, we do foresee the auto and industrial segment, inventory health -- inventory situation will become more healthier by end of the year. So I think all market segments within UMC addressable will become much healthier in terms of their inventory situation starting from '25.
I see. Okay, the next question is for the [indiscernible] 3D-IC packaging, which Jason started to talk a bit more. I want to know the value proposition for UMC in this business because you don't have the [indiscernible], you don't have 3 or 5 nanometers. So which is somehow vertical integration is important, so where do you see your value when you don't have the advanced node dies for the packaging? And what's the profitability look like for this business, it's going to be the margin accretive for you?
Well, I mean, first of all, I mean, you're right. But if we look at this market specifically, the -- some device is looking from a form factor standpoint and some devices looking from the enhanced bandwidth standpoint and lower power consumption. So if you have a horizontal view, I think from the small form factor standpoint, some of the applications do not require most advanced technology nodes. And some of the applications will probably need a little better but not -- still not the most advanced, because they're seeking for the balance between the form factor and the higher bandwidth and the power consumption. But if you directly referring to the very super high bandwidth and which -- that's a space that we are not addressing. So there's still a sizable market within the stacking, the 3D-IC space for us.
So that is what you mentioned about like 15% to 20% of the addressable market.
That's among the --
The total.
Yes, the overall AI semiconductor market, yes. And that's part of --
Understood. And the profitability? Profitability for this business?
I mean we're very cautious about that, and we've been running the company, improved our structural probability for a few years already. And of course, the -- any solution we're offering and operational efficiency was always part of the consideration. So that will be a profitable operation for us, yes.
So we can consider as a margin accretive for this business?
I mean -- right now, I can't comment margin accretive, we'll try our best to maintain our structural profitability, yes.
Next on Charlie Chan, Morgan Stanley.
So first of all, Jason, great calls on the cycle and the market forecast. I think your [indiscernible] is converging to your kind of forecast on the [indiscernible] market. And also excellent job on the pricing discipline, so well done on the margin side. So a couple of questions from my side. So first of all, the CapEx, right, does that include some spending for the U.S. partner fab? Because you kind of mentioned that there could be some debottlenecking required for the U.S. operation.
I mean not for the 2024 number, that's not a whole lot. So that will not change the current CapEx projection. The -- when we say we peak out in 2024 in terms of CapEx and the 2025 will start declining, that's already including that assumption, yes.
Okay. So any CapEx for that debottlenecking should be more 2025?
Well, yes, most of them will happen there, yes.
Okay. Yes, and also staying on this 12-nanometer business, right, you said the -- you try to accelerate customer schedule. Do you think any kind of production can be ahead of 2027?
We certainly hope so. But cooperation like this scale naturally come with various kind of challenges so we have gone through a rigorous due diligence since day 1, and we have been proactively managing those potential challenges. So, so far, the collaboration has been a positive for us. And so we want to focus on to accelerate that, but I can't give you any specifics at this point, but I think we are -- I think I can tell you the project is on track, and we have a good confidence that we're making good progress right now.
Got you. So in terms of the customer's feedback or demand compared to maybe 3 months ago or 6 months ago, do you see more commitment from your customers or partners? So I remember the Intel fab capacity is 50k to 60K, right? Are you confident that you can fill that capacity?
I mean we never talked -- we never specifically talked about capacity size. In terms of the capacity arrangement, it's very typical we have to work with our customers and align with them based on their forecasts and then we can deploy that. So there's no specific capacity number at this point. The -- secondly, we are seeing more interest because we announced this early this year, obviously, that customers want to know more about it. And so we see a lot of interest and I think we're making some or good traction. But first thing first, we have to demonstrate that ourselves. So all hands on deck is we just focus on execution today. And given the project execution standpoint, we don't foresee any major risks now, yes.
Okay. Yes, so I heard that your progress also get on nerve of your industry peer. For example, at least a key customer, Novatek, right? I'm not sure whether there's some dynamic that you exceed Novatek both because of Novatek in the future, probably they won't depend on TSMC and TSMC also want to bundle with this customers tighter. So I'm not sure if there's already some counter move from your major competitor in 12-nanometer.
I mean first of all, we do not comment on any specific customer, we never do. And from the competition standpoint, we will look at this in a way of building a strong relationship with our customers. It can only forge upon a fundamental strength in technology leadership, manufacturing, capability, excellence, and on the capacity offering. And so you have to win back that. You don't win by relationship. Having a good relationship is always good, but that's not we focus on. And the -- we will continue to strengthen our competitive advantage in supporting our customers and winning their business and building that relationship.
Got you. So wish you success. So may I switch gears to some near-term or financial questions?
Sure, absolutely, yes.
So Chi-Tung, on the gross margin side, I'm wondering how you model the kind of electricity cost impact to your gross margin? And also, is it -- depreciation still growing like 20% year-on-year for 2024?
Yes, depreciation should grow around 20% or a little bit less, maybe given the current new schedule for Singapore P3. And the utility impact is actually after the -- all the costs together blended together, is actually rather minimal. And our goal is always try to offset that through our operating efficiencies and also the benefit from the larger economy of scale. So the third forecast is a small impact to our operating, gross -- to our gross margin.
Okay. So it sounds like you don't intend to pass through this kind of minimal additional cost to your customers. Do I interpret this comment, right?
I mean, the cost and the pricing is -- to us is actually a 2 different thing. The -- from an ASP standpoint, I mean, we -- our pricing strategy is, like I said, is consistent based on our value proposition. And then we -- and we have to align with that, that includes the competitive resilience, right? But the cost, we always want to try the aggressive cost reduction efforts to compensate or offset some of the headwinds and to maintain structural probability. So we -- it's not a cost-plus business model for us.
I see. So in the Q&A with Bruce, I seem to hear you kind of admit that the second half margin can be also at a similar level. Did I hear you right?
I mean we're typically giving guidance quarter-over-quarter, but at this point, we more look at the market outlook and the ASP projection will stay firm. But we do withstand the headwinds, like you said, the 2D cost increased, inflationary cost increase, depreciation cost increase, and so there are some headwinds that we have to deal with. So we will continue aggressively spend effort to improve our cost structure. But -- and so the goal will be, at least, that's continue improve our structural profitability from that standpoint.I mean, I have to say, currently, we have been navigating the industry dynamics, even with the utilization rate below 70%, right? I mean -- so I think we have demonstrated that for the past period. And now we're dealing with the headwinds in depreciation of all those inflationary cost pressure, and I think we'll continue to do so. And so despite those challenges, we will have a relentless effort to over those challenges, yes.
I see. The last one from me, so on the end market trends, one observation. So I fully respect and you have been right about the cycle recovery, the market forecast, right? But recently, we seem to hear that smartphone supply chain continue to see all the cuts. One of your major customers will report this Friday, I guess. So we are seeing inventory destocking, right? Some tough situation for China smartphone still. You seem to say that inventories are healthy. So can I know how you're going to reconcile those data points?
In the -- I mean, I kind of touched that. So first, we are cautiously optimistic about the rest of the year. If I put it this way, at this moment, we foresee the Q1 2024 could be the button. And the biggest uncertainty is the overall macro outlook that could impact end market dynamics. So because the visibility -- insufficient visibility at this point, given the customer's practice -- rush order practice approach. So we feel optimistic about it, but we need to be cautious about it. If the end market does not recover as we expected, we could have a challenging second half. But however, at this point, we have not seen that. So we really hope that if the market continues digesting the inventory, gradually improving -- the end market demand gradually improving, we actually foresee the Q1 '24 will be a [ baton ] for the year.
Awesome execution on the pricing discipline and also the 12-nanometer strategy.
Next one, Jason Tsang, CLSA.
Wonder if you can give us more details in terms of demand in second half. I mean, can you give us outlook or more details for 28-nanometers or [indiscernible] nanometers or 8-inch.
Well, I mean, let's talk about the Q2 since we are in the current quarter. And the Q2 outlook and we look at the applications, we expect the wafer demand in consumer and computing segment will grow while the automotive industrial segment will remain soft as they still digesting the inventory. So they're still in the mixed bag for the Q2, but overall shipment, I think, will be -- we'll grow quarter-over-quarter sequentially. The -- if you break it down to a technology node, the 28 and 22 will gradually improve. And for the mature nodes, the 12-inch, 40-inch and 65-inch will stay flattish, while the 8-inch will stay flattish as well. And so that will be the current projection for us.
Okay, thank you. So can we expect that there will be a normal seasonality demand in Q3? And how can we expect -- whether your utilization rate can reach maybe around 70% in this transitional half season? Can we expect that kind of seasonality or demand in Q3?
Well, I mean the -- it's -- at this time, the market is still lack of a sufficient visibility like I said, for the second half. I try to give you an example that we have observed rush order from customers as they continue to managing their business prudently. And for auto and industrial, they have a slower inventory digestion. So we think that by end of the year, I think auto and the industrial will be at a much healthier position. So given the mix of that, it's hard to tell you the -- if the Q3 will recover to 70% utilization rate. We certainly see the -- some of the market segment has a healthy inventory. So they have -- the demand will more directly linked to the end market needs and -- versus the other -- the auto and the industrial still have to digest their existing inventory on hand.So if the macro situation is healthier, if the macro is a recover, which we expect the computer, consumer and the communication segment, the demand will redirect to us to improve the loading situation. So there's so many different variables right now. And given the insufficient visibility, I can't give you anything specific, but we definitely feel optimistic about that because the inventory situation is improving in many of the market segments already. We just have to continue monitoring the progress, and we will keep you updated quarter-by-quarter.
Great. So my last question is in terms of 28-nanometers or 22 because on the demand side, actually, we found some of the clients are migrating from 22/28 to maybe FinFET process nodes, including SSD controller or Wi-Fi 7. And on the supply side, we found that your competitor in China is playing into [ intentional mass auction ] for 28 high-voltage process [ node ]. So the demand is moving into the FinFET process now. And on the supply side, we know that UMC has kept good positions in 28-nanometers. But on the supply side, we found that the new players are planning to enter the high-voltage process market. So how do you look at -- or how do you plan to keep your market shares or keep your technology leadership in high voltage or in those kind of niche markets or your products or to maintain your market shares?
Fundamentally, the answer to that is we need to stay competitive. In the past and even currently, right now, at this moment, we work closely with customer -- strategic customer to build specialty technologies, differentiation and lasting value for them to enhance their -- our product portfolio as well as our customers' product portfolio and with a diversified capacity. And so that -- we want to do that to offer our customer with a competitive solution. And so I think that's the answer to that. The competition is everywhere. And the key answer to that is we have to stay competitive, and we'll definitely strive to do that.
Next one, Laura Chen, Citi.
I think my question is also kind of a follow-up on the CapEx outlook and also the depreciation. Since Jason, you mentioned that the CapEx will pick up in this year, I'm just wondering that looking forward, how should we look at the depreciation cost trend into next year? How would that impact our gross margin?
The depreciation impact will be a kind of a delay factor. So if the CapEx is big in 2024, I think the depreciation expense probably will be peak 2 or 3 years later. So the trend will still be on a minor upward trend all the way to 2026.
Yes, so the magnitude of the depreciation cost increase for the following few years, you expect that would be more like a steady increase or that will be gradually like a scale up because still, in the past 2 years, we see the CapEx increase.
Keeping our ROI-driven base CapEx, and we are managing our overall depreciation expenses as a percentage of overall revenue. So I do expect the increased magnitude should be under control and also acceptable by the investor community.
Okay, thank you. And also for the AI opportunities, I understand that, Jason, you mentioned that will be 10%, 20% addressable market? And where are we now? And do we expect there will be a secular growth for the following 2, 3 years?
Well, I mean, Laura, like I said, we're still in the early stage of the proliferation of the AI -- Edge AI. Most of the focus today is running the computation chips for the AI modules today. So we have a very small exposure there. And -- but we do expect the Edge AI will start coming in. And the -- there are some of the projects that we have in discussion with customers. They all have a very high expectation. The market was taking out soon, but I don't have any specifics today. We do project those applications or those features will be used in the Edge AI in the next few years, and that will probably be representing at 10% to 20%, but it could be even higher. But at this point, we project about 10% to 20% of the AI silicon, yes.
Ladies and gentlemen, we are taking the last one. The last question, Timm Schulze-Melander, Redburn Atlantic.
I just had 2, if I could sneak them in, please. The first one was on the advanced packaging and chip integration. You talked about hybrid bonding moving into greater volume in 2024. Just wondering if you could share a few more details about the end application and if that's the partnership that you had press-released with ASE, Faraday and Cadence, is that the application? And then I had a quick follow-up, please.
Well, first, for the wafer-to-wafer hybrid bonding solution today, the applications for the RF front-end module, and the -- for the -- it's not completely correlated with the ecosystem press-release. But to address the -- your part of the question about the ecosystem compensation, for the ecosystem, we are not talking about one specific solution. We're talking about the UMC overall 3D-IC solution is leveraging the entire ecosystem to support that. We do not do turnkey solution, means the end-to-end solution. We're only providing wafer-to-wafer hybrid bonding or the interposer and solution while we have all the ecosystem partners to support the customer to provide the total solution. So that's not only limited with the hybrid bonding.
Okay. But for that hybrid bonding, are you actually doing that wafer-to-wafer bond yourself in a UMC facility, or is that being outsourced or shared with some partners?
It's in-house, yes.
Okay. And maybe just if I can, one sort of much, much bigger picture, long-term question. The last year or so have seen significant capacity expansion in more mature nodes among China-based chip makers across a whole range of products. Just as you look out 3, 5 years, how does that sort of influence the sort of strategic land shape -- landscape and the amount of competition that you see or that you expect UMC to face?
Well, I mean, we kind of touched that earlier. The -- it's a couple of things that we see that driving the landscape changes. One is we do see more capacity build up expanding for geopolitical reasons and/or for the reason the semiconductor industry has become so important. So people want to build up more capacity regionally. And so there are multiple reasons that drives the overall capacity buildup. And the -- fundamentally, I mean, we believe we need to stay competitive. So we're working closely with our customers to build differentiation and to provide value to enhance our product portfolio. And the end goal is, we will strive to minimize the -- some of the commodities exposure in some of the selected markets, if those capacity build up and it could have affected the market -- the end market with some of the commoditized products. And so we need to continue to differentiate ourselves to minimize the exposure with that.If we look at the buildup situation, if they are driven by the geopolitical tension, that will, in our view, impose not only the constraint, but it also presents some of the opportunity as a customer seeking alternative to support their sourcing objective. And for UMC, we are well-positioned because of our geographically diversified manufacturing sites, which allows us to work with the customers to navigate this geopolitical concurrence. We are only the few foundry that have fabs across Singapore, Japan, Taiwan and China and back with a comprehensive technology portfolio, not single node, not limited technology offering, but very comprehensive technology portfolio.And that could be bridged between fab, which actually is very beneficial to some of the customers if they want to have a fab flexibility, the cross fab flexibility. And each of our sites is equaled with sizable capacity that can serve numerous market segments. And each of our regional manufacturing site also being well established with a still and complete ecosystem, and we can run very highly efficient local operations.And if you look at ourselves, if you look at longer term, besides our strong manufacturing footprint in Asia, now we also extend our capacity offering into Arizona. You saw the cooperation with the U.S. partner in 12-nanometer. That's another testament that we're expanding our technology offering as well. And so this will continue to further our competitiveness and to strengthen our customer supply chain resilience. So I think we can bring value to customers, and that's how you compete. I think we are positioning ourselves to continue improve our market relevance, our market position. And hopefully, the customer will see that, and they recognize the value, then they will stay competitive and to differentiate ourselves. I hope that gives you a bit of context in terms of bigger picture.
Ladies and gentlemen, we thank you for all your questions. That concludes today's Q&A session. I now turn things over to UMC Head of IR for closing remarks. Thank you.
Thank you for attending UMC conference call today. We appreciate your questions. As always, if you have any additional follow-up questions, please feel free to contact UMC at ir@umc.com. Have a nice day.
Thank you. Ladies and gentlemen, that concludes our conference for first quarter 2024. 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, and goodbye.