Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR (TSM.US) FY26Q2 conference call: Q2 gross margin of 67.7% beats guidance; full-year capital expenditures raised to $60-$64 billion.

date
21:30 16/07/2026
avatar
GMT Eight
The company has revised its full-year 2026 U.S. dollar revenue growth rate to "slightly above 40%" (previously over 30%).
Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR (TSM.US) held its FY26Q2 financial report conference call. The company announced a cash dividend of TWD 467 billion for 2025, representing a 28.6% year-on-year increase, with a full-year cash dividend per share of TWD 18; the cash dividend per share for 2026 will increase to TWD 24, a further 33% year-on-year increase, with management committing to continue the increase in 2027. In terms of performance guidance, the company expects revenue for Q3 2026 to be between $44.6 billion and $45.8 billion, corresponding to a 12% quarter-on-quarter increase and a 37% year-on-year increase; gross margin is expected to be 65%-67%, and operating profit margin is projected to be 56%-58%. The company has increased its full-year revenue growth rate for 2026 to "slightly above 40%" (previously over 30%). Regarding gross margin, the Q2 gross margin increased by 150bps to 67.7% quarter-on-quarter, slightly exceeding guidance; the Q3 gross margin is expected to decrease by 1.7% quarter-on-quarter to 66%, mainly due to the steep ramp-up of the 2nm process diluting gross margin by approximately 3-4%. The ramp-up of the 2nm process in the second half of the year is expected to dilute gross margin by 3-4%; the ramp-up of overseas fab is expected to dilute gross margin by 2-3% in the early stages, increasing to 3-4% later. In terms of capital expenditure, the company has increased its full-year 2026 capital budget to $60-64 billion (previously around $52-56 billion in January and around $56 billion in April). Approximately 70%-80% of the budget will be used for advanced processes, with around 10%-20% allocated for advanced packaging, testing, etc. The capital expenditure for Q2 was approximately $15.7 billion; cash and securities at the end of the quarter were TWD 3.5 trillion (approximately $110 billion); the ramp-up of the N2 process led to an increase in inventory days from 80 to 87. Q&A: Q: Will this current upcycle provide capital expenditure guidance for the next three years (26-28) as it did in 2021? A: Currently, there are no specific numbers to share. We are investing in capital expenditures this year for future business opportunities, and as long as there are opportunities, we will not hesitate to invest. As mentioned in our prepared remarks, we have a strong confidence in the multi-year AI trend, which is why we are increasing capital spending and adjusting this year's capital budget. Previously, we mentioned that the capital spending for the next three years will significantly exceed the past three years, and now the situation is that the capital spending for the next three years will significantly increase compared to the past three years. Q: With an additional $100 billion investment in Arizona, totaling approximately $265 billion in cumulative investment, how will the capacity expansion plan in Arizona be implemented over the next few years? A: The timetable depends on market conditions. With the current strong trend, we announced additional investment in Arizona. How many fabs will we build? Many. Specifically, we may build about 4 more fabs. (Confirming that this includes previous funding arrangements.) Q: How does Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR respond to competition from semiconductor foundry competitors such as Samsung (earning high profits through the storage business) and Intel Corporation (receiving US policy support)? Some US customers are in contact with these competitors, and ASML has announced an expansion of EUV capacity for 2028. Are you concerned about competitors taking more capacity and competing with you? A: Yes, I envy a competitor in South Korea who has made a lot of money, and another competitor in the US who has received strong support from the US government - incidentally, we have also received government support, just not publicly announced. However, there are no shortcuts, meaning we must return to the fundamentals in the semiconductor industry. We welcome and appreciate government support, and money is of course important. But what is most important is always technology, manufacturing, and customer trust, which have not changed in my 37-40 years of career. This has always been the key to Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's success. From a competitive perspective, choosing a technology and ramping up production is not like going to 7-Eleven to buy milk - it's not that you see a better bottle of milk today, so you go to the next store, and if you don't like it, you switch to another store. You have to actually use it, test chips, work together, prepare capacity, and ramp up, which takes about 5 years. Q: Since customers and their customers have shown strong signals and raised annual guidance, will the 5-year (semiconductor/AI) CAGR be revised? It was previously around the high 50% range; also, as the cost of storage is rising, which accounts for a considerable portion of AI capital expenditures, how do you view Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's contribution to AI-related growth? A: If you have read our information - we continue to increase investment, raising capital expenditures, all for good reasons. If you ask about the CAGR for AI, I won't give you a number, but the answer is stronger, stronger, stronger. I won't give a number today because it is still increasing, stronger than what we previously said. Q: With the increased competition in advanced packaging, especially EMT (Embedded/Local Interconnect Packaging) gaining market attention, how will Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR respond? A: Our packaging capacity is very tight, and it has constrained customer growth. Having more flexibility in the market can actually help Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's wafer business (which is the main business of Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR). This technology looks promising according to reports, and we also hope it will be successful to help offload some of Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's orders. We are currently working hard to narrow the gap between demand and capacity, so we welcome more alternative solutions to provide flexibility for customers. Q: As a new technology, how will Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR handle special requirements if problems arise after customers adopt it, and are there alternative solutions? A: Our first priority is to support the success of our customers. Whatever helps our customers' businesses, we are willing to do. Q: When planning capacity expansion, in addition to considering customer and customer demand, will you also consider the competitive pressure brought by competitors building capacity (as a market leader, long-term supply exceeding demand is not ideal)? How long will it take to meet current demand? In addition to chip shortages, there are also issues such as data center delays and power supply - how do you incorporate these into your planning framework? A: Every time we consider business, competition is the primary factor we consider; then we assess our position, evaluate demand with a combination of a bottom-up and top-down approach, which is the most challenging part of our daily work. We make many judgments, communicate more cautiously with customers and their customers (i.e., CSP) to get all demand inputs before making decisions. It's important to note that, I believe every customer tells me the truth, but the sum of all customers' "truths" may not necessarily be the truth - because all CEOs tend to be aggressive, that's their job. So we have to make very prudent judgments, which may not be entirely correct, but because this involves large amounts of capital (this year's capital expenditures have increased from around $52-56 billion to $60-64 billion, and will continue to increase), we are being very cautious. We are also monitoring the progress of AI data centers - construction, site selection, demand, rents, etc., to ensure that Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's chips are not stockpiled. Q: Despite such a large-scale capacity expansion plan, do you still expect a supply shortage by the end of next year? A: You're asking me to guarantee that. I believe that demand will be very strong from now until around 2029-2030. This trend is so robust that I believe we are witnessing a whole new industry - what I call the AI industry, which has already deeply penetrated daily life and will affect industries such as automotive, humanoid Siasun Robot & Automation and all industries. Given the amount of money being poured in by everyone (including all CSPs), it is very important for the world. This demand will continue, and at its core are semiconductor chips, with the majority being from Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR. Q: Regarding profitability: historically, the profitability of semiconductor foundry (especially advanced processes) should be higher than that of storage; however, Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR is no longer the most profitable semiconductor manufacturer. Does this mean that the pressure on conveying and capturing value from customers is less than a year ago? A: Your question is actually very simple - what is Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's pricing strategy and what should be the profit margin? Of course, the higher, the better. But we are partners, and as I have said many times, the success of our customers is crucial, and I do not want to extract too much value from the market; and we are a company that customers can trust, we won't suddenly increase prices by 4 or 5 times, because that way customers cannot survive. We earn the value we deserve and ensure that the profit and gross margin are sufficient to support long-term sustainable expansion, which is beneficial for both customers and Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR, and that is our belief. So, I do indeed envy the storage companies with an 86% gross margin - I would be happy with 68%. In short, we are a very trustworthy company. Q: With AI demand significantly surpassing other end markets and the top five customers reaching a historical high proportion, how do you view the risk of customer concentration? A: This is not a concern for us. The fact that customers are getting bigger and bigger makes us happy, with some customers growing very rapidly. And it's not as simple as "big customers getting bigger", in fact, there are many new players emerging in the AI industry. Q: Your direct customers are providing financing/investment for their end customers to support AI demand, will Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR consider providing such investments or financing arrangements for customers' customers (end customers)? A: To answer directly: every company has different considerations and strategies. So far, Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR does not make these financial arrangements because we have been working smoothly and successfully with existing customers in the current model. Q: With the additional $100 billion investment in the US, how will the distribution of this investment over the next 3-5 years proceed, and what will be the pace? A: We do have a plan, but the progress and timing mostly depend on market conditions and customer demand. If you want a definite timetable, we don't have that today, but we have a plan and will push the progress forward as much as possible. (Addendum: We will also push forward with the new facilities and factories in Taiwan and Japan as soon as possible, as the current supply-demand gap is quite large, and we are working to narrow this gap.) Q: In the HPC computing sector - (CPO/co-packaged optics, etc.) platform, when will it contribute significantly to revenue? A: It has already begun production and will ramp up gradually over time. I believe that AI data centers need to reduce power consumption and increase communication channel bandwidth, so this technology demand will continue to grow and become quite important in the coming years. Q: Agentic AI vs CPU growth potential - how do you view the growth potential and visibility of different AI chips (GPU accelerators vs CPU/XPU)? A: I cannot provide specific numbers, but I can share that they all use Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR and the same advanced processes. We are working with customers to allocate wafer supply to balance the proportions of CPUs, GPUs, and XPUs. Q: Regarding advanced packaging, you previously announced a roadmap with a 14x reticle size to achieve larger AI packaging, and recently showcased the development of CoWoS glass substrates in Japan; how are glass cores, glass substrates, and glass carriers progressing? A: CoWoS is still the mainstream at the moment, and we are developing alternative solutions to reduce costs, and we are collaborating with substrate suppliers to help customers bring products to market. The pooling (related production lines) we announced several quarters ago are under construction and will take about a year to mature, at which point we will be able to start volume production with customers. Q: Can you quantify the outlook for sales growth in the coming years? Additionally, could you break down the key drivers of the current capital expenditure increase and demand this year - is it still mainly from cloud computing, or is it spreading to edge computing? Are there also factors of price increases in the equipment supply chain? A: Revenue is related to investment - we predict demand, assess, and then decide on capital expenditures. The next few years will be very good for Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR; that's all I can say. (Key drivers: Everything related to AI.) Q: The competition in backend packaging is intensifying (especially Intel Corporation EMT); are you concerned about the erosion of the overall foundry value from manufacturing to packaging at Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR? A: Front-end wafer business and backend business are two different things and are not the same. And our backend capacity is in short supply, and the gap is larger, so I welcome competition to provide flexibility for my customers in terms of packaging my front-end wafers, which ultimately helps the wafer sales of Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR. That's our attitude. Q: You previously mentioned that High-NA equipment is too expensive; how do customers view the challenge of stitching brought by a smaller exposure field? Even if the technology improves, will this slow down the adoption of High-NA? A: You have a deep understanding of High-NA. Currently, its exposure field is only half the size, and we have factored this into manufacturing costs. Regardless of whether we use it or not, High-NA is a very good tool with excellent performance. Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR has expressed its intentions to cooperate with ASML and is working to make it more suitable for manufacturing in terms of cost and maturity. We will always consider technological maturity, cost, and then decide whether to adopt it or not. Q: The market generally assumes that demand for 3nm and below exceeds your capacity by 30%-50%, is the actual gap even greater? A: We do not have specific numbers to share. The gap is significant. Q: It was previously mentioned at a technical forum that the 2nm family capacity for '26-'28 is expected to grow at a CAGR of 17%, and N3 and N5 at a CAGR of 25%, are these assumptions still valid today? Have there been any changes in the past quarter? A: The numbers we presented at the technical forum - they are now larger, that's all I'll say. Q: Advanced packaging capital expenditures have always been bundled with testing, photomask manufacturing, etc., accounting for about 10%-20% of total capital expenditures, but what is the actual proportion of investment in advanced packaging? How do you view the gap between packaging revenue and capital expenditure proportions? Should it be considered as a separate capital expenditure item? A: We work very hard to ensure accurate capital expenditure figures, but there needs to be flexibility between the front end and back end. Sometimes there are bottlenecks, and we invest more to buy equipment for those bottlenecks, sometimes at the front end, sometimes at the back end. In the long run, this ratio is roughly the level I have shared, around 10%-20% for the backend, and this range itself is quite wide. To be honest, as time goes on, some customer products require more testing, and as testing demand increases, we will invest more capital expenditure in testing equipment, packaging, or other areas, so it is not possible to specifically delineate how much investment goes into each area. Q: From the beginning of the year to now, Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR has raised capital expenditure guidance by nearly $10 billion, where does the additional space come from? Compared to six months ago, is it due to changes in CPU/accelerators, storage vendors, or the expansion of back-end capacity? A: The most important reason is the continuous growth in demand, and we feel the pressure from customers driving (more accurately, in collaboration with Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR) expansion, which is one of the main reasons. The second reason is inflation - now we are buying equipment at inflated prices.