Thank you so much for joining the earnings call of Kioxia Holdings Corporation today. For the participants who are joining online, if you are using a nickname to log in, please change it to your name and the name of your affiliation. To change your login name, please log out once, go back to the registration page on this Zoom webinar where you can change your name and the name of your affiliation, and log in again. Thank you for your kind understanding. The session will begin soon. Thank you. Thank you for waiting. We will start the earnings call of Kioxia Holdings Corporation for the first quarter of fiscal year 2026. I am today's facilitator, Nishida, in charge of IR from Corporate Communication Division. With me here are Executive Vice President and Executive Officer, Chief Financial Officer, Yoshihiko Kawamura, and Head of Corporate Communication Division, Toshiaki Fujikawa.
Let me inform you on the process of today's call. After outlining the disclaimer, Mr. Fujikawa will explain about the overview of the NAND market outlook and enhancement of our production capability. After that, Mr. Kawamura will go through the financial results of the first quarter of fiscal year 2026 and the guidance of the second quarter, that will be followed by the question- and- answers. The entire session will end at 5:00 P.M. Please be advised that today's session will be recorded for on-demand viewing later. We would like to start with a disclaimer. Forward-looking statements are prepared based upon our expectations and projections in light of the information currently available to us, which involve various risks and uncertainties. Such risks and uncertainties may cause our actual results to be materially different from any future results expressed or implied by these forward-looking statements.
We undertake no obligation to update any forward-looking statement included in this material. For more of the disclaimer, please refer to this slide on the screen. Now, let us hand over to Mr. Fujikawa.
This is Fujikawa. First, let me talk about the current state of the NAND market. Currently, growth in AI demand is driving the NAND market, with data centers and enterprise applications leading the growth. Notably, agentic AI applications are a key growth driver for NAND demand, impacting even demand for traditional servers. The smartphone market is seeing a decline in shipments of low-priced models, while high-end models are performing steadily. As a result, we anticipate smartphone NAND demand to remain flat compared to calendar year 2025. The increased storage capacity of high-end models is supporting demand.
The PC market is expected to see a decrease in shipments due to rising BOM costs, leading to a slight decline in NAND demand compared to calendar year 2025. In light of these market trends, we see AI-driven data center and enterprise demand as sustaining market growth, and we believe that the NAND market will continue to grow. For calendar year 2026, in terms of bit growth, we currently anticipate the NAND market to grow in the high teens percentage range, in line with how we see our own bit growth for the year. Looking ahead to calendar year 2027, we forecast that demand will exceed supply. We will continue to closely monitor market trends and respond appropriately to demand. Next, let us explain about our production capability to support the growing demands in the AI era. We are continuously improving productivity by leveraging big data and AI technologies.
Through continuous capital investments, we are transitioning to next-generation products to strengthen our competitiveness. Additionally, our long-term joint venture partnership with Sandisk allows us to leverage economies of scale through ongoing joint development and co-investment. We have a multi-site production system comprising two locations, our Yokkaichi plant and the Kitakami plant, which addresses the increasing demand for NAND and allows us to achieve a stable supply. Both plants excel in terms of business continuity planning. The Yokkaichi plant is one of the world's largest flash memory manufacturing bases and serves as a key site equipped with leading-edge R&D. It enables smooth transitions from product development to mass production, and close collaboration between development and manufacturing department facilities facilitates efficient launches while leveraging advanced technologies and know-how. It also houses advanced R&D space for developing next-generation memory, playing a crucial role in future technological innovations.
The K2 fab at our Kitakami plant began operation in September 2025 and began production of the 10th-generation bics flash this year. With the introduction of the latest equipment, the plant achieves high productivity and contributes to further improvements to yield and productivity for products developed and mass-produced at Yokkaichi. It also boasts different site expandability, positioning it as a key manufacturing base supporting future demand growth. Through the AI-powered productivity improvements, technology node migration facilitated by continuous capital expenditures, JV partnership, and a multi-site production system, we are steadily expanding our production base to support long-term demand growth. To continue addressing demand growth and investing in next-generation technologies, we plan to spend an average of JPY 470 billion annually on capital expenditures from FY 2026 to FY 2028. That concludes my part. Thank you.
Next, we will ask Mr. Kawamura to take over with the presentation.
I am Kawamura, CFO in charge of finance department. Thank you for squeezing out of your busy time to join our earnings call. Please refer to page eight. That lays out our financial highlights. April to June are first quarter's financial results. Across the board, we achieved record high numbers in both revenue and profit. Let us take you through from the left to right. Left-hand side, revenue, record high. The revenue increased 1.8 x quarter-on-quarter, and JPY 1,767.1 billion. Next, non-GAAP gross profit 2.1 x quarter-on-quarter, JPY 1,405.5 billion. 80% of the gross margin and the net gross margin, excluding joint venture related items, was 82%. The non-GAAP operating profit hitting the record high as well. Quarter-on-quarter, 2.2 x, JPY 1,326.2 billion, exceeding JPY 1 trillion.
This level last year, the entire year's non-GAAP operating profit was JPY 876.2 billion. Now, for this quarter already exceeded the entire year of the last year. The margin was 75%. We added the new indicator, which is the core free cash flow. I will explain this definition later, but excluding large size amount A. This represents the pure cash generated from operation. Basically, the operating cash flow deducting the net CapEx Q-on-Q, 3.4 x, hitting a record high JPY 827.2 billion. As a result, on the right-hand side, we achieved a net cash position. We repaid the senior loan entirely in the first quarter, achieving the net cash position. Next slide, please. This slide lays out the quarter-on-quarter and the comparison against the guidance.
From the top to the bottom, revenue, non-GAAP operating profit, net non-GAAP net income, non-GAAP EBITDA, which is the biggest chunk of the cash flow, and earning per share non-GAAP and FX. From the left to the right, fourth quarter's results and guidance we announced on May 15th, and the results of the first quarter. The revenue guidance was JPY 1,750 billion, and actual over the first quarter, JPY 1,767.1 billion. Q-on-Q 76.2% up. Year-on-year 415.5% increase. The next, non-GAAP operating profit. The guidance, JPY 1,300 billion. The results is JPY 1,326.2 billion, margin 75%. Q-on-Q 121.4% up. Year-on-year 2,833.2% up. Astonishing numbers. Non-GAAP net income guidance, JPY 870 billion. The actual was JPY 887 billion, margin 50.2%. Q-on-Q 116.4% up. Year-on-year 4,692.5%. Again, the astonishing numbers we see here.
As I said earlier, senior loan has been repaid, and that's been factored in here. The non-GAAP EBITDA, please look at the results, JPY 1,402.1 billion, margin 79.3%. Quarter-on-quarter 107.6%. Year-on-year 1,022.7%. Quite high numbers we recognized. On the right, selling price and the gigabyte volume. The selling price went up by 70%, like for like blended ASP both increased by 70%. The quarter-on-quarter shipment volume also increased the low single digit. Therefore, they both contributed to our results. Our earning per share, second from the bottom, guidance JPY 1,593.15, but actual JPY 1,621.81. FX, against the guidance of JPY 159, the actual was the JPY 160. That's at actual.
Yesterday, according to the news media, there was some intervention, but this FX has been quite flat for the entire term. There are some items which are not included items. That is there's some phasing timing of the volume, which has been deferred to the second quarter from the first quarter. Also, another point is that there are employees reward, that we are going to pay the special compensation to be paid to the employees to reward them. The allowance for it has been recognized, JPY 50 billion a year. That will be starting from this fiscal year. Also, another point is that there are some difference in the accounting system. From the non-GAAP to IFRS, there are two items to be added. Please go to the page 26 in appendix.
That is the equity-based compensation, JPY 19.4 billion, and some litigation related, JPY 36.6 billion, to be recognized in IFRS from non-GAAP. Please go to next slide. Page 10 and page 11 shows by application. This is the sales and sales actual by application. The inside blue indicates smart devices. Red indicates SSD and storage, followed by others. In the middle, you see the revenue per application. Smart device, JPY 525.7 billion. Quarter-on-quarter 55.8% up. Year-on-year increased by 565.1%. Main reasons on the right-hand side, it's driven by the selling price, significant increase. Also, the revenue hit the record high in this segment. Red indicates SSD and storage. Revenue was a JPY 1,174.7 billion. Quarter-on-quarter 95.7%. Year-on-year 440.3%. Significant increase. The reasons are laid out on the right-hand side.
The PC segment, which accounts for less than 40% right now of this segment. Mainly, quarter-on-quarter sales went up driven by the selling price increase. The data center enterprise, relatively speaking, is increasing in the SSD and storage, this time accounting for over 60%, driven by the growing AI server demands. The volume hit the record high, and driven by the selling price, the revenue also hit the record high. The BiCS8 now is exceeding 50% over the entire production right now. The others, JPY 66.7 billion. Quarter-over-quarter, 2.3% up. Year-over-year, 44% up. Page 11, please. Still by application. The color legend is the same as the previous slide. Blue, smart devices, and red indicates SSD and storage. Gray shows others.
From the left, fiscal year 2025 quarterly basis, first quarter, second, third quarter, and fourth quarter of fiscal year 2025. The revenue total, JPY 1,002.9 billion. JPY 65.2 billion coming from the others, and the blue, 3,373. The red numbers show 6,003, 6,003. If you go to the first quarter of this fiscal year, red increased to JPY 1,174.7 billion, almost double quarter-over-quarter. If you compare that to the right-hand side, that's the yearly results of fiscal year 2025. SSD storage was JPY 1,362.6 billion. The first quarter alone is close enough to the entire fiscal year 2025's number over SSD storage, which indicates SSD and storage is a driving force of our revenue and the profit, sustaining our profitability. FX, at the very bottom you see. From the left, last year, first quarter, JPY 145.
Since then, the weak yen is progressing.
Moving on to page 12, please. This will be cash flow. We are discussing cash flow here. I did mention this earlier, but from this time, we are using the word core cash flow. This is going to be a new disclosure item. This is going to be a new KPI that we would like to add on, and we will be leverage using this going forward as well. What is this core free cash flow? Investment cash flow includes M&A, and therefore, there's a Taiwanese Nanya investment that we made this time around, and this was quite significant. If we include this, the operational stamina cannot be discussed. We wanted to have a separate core cash flow that is excluding this investment. Please look at the left bar chart.
For March end of 2026, we have JPY 470.7 billion remaining as cash, this will be record high. Looking at operating cash flow, it is JPY 866.3 billion. At the very right, when you look at operating cash flow, this is redundant, but there with ASP increase and data center enterprise SSD, a sales increase is also affected as well. Looking at BiCS8 expansion as well. As a result of that, quarter-over-quarter is about three times more, this is record high as well. This time as well, for quarter one, the operating cash flow is JPY 866.3 billion, as you can see on the very left-hand side. Moving on to the right-hand side, which would be the -39.2. You can see the investment cash flow. How are we calculating this?
This is CapEx investment, which is JPY 52.4 billion, which is gross CapEx. From here, there is subsidy, which is JPY 12 billion that will be deducted. On the very right, there would be the JPY 1.2 billion, which is adjusted. That would be, it is -52, and then there is + JPY 12 billion. Therefore, that would be the -39.1. With this deduction, the core free cash flow is going to be JPY 827.2 billion. At the very right. When it comes to other financial assets, this is going to be the acquisition of equity at Nanya. That is a -JPY 78.2 billion. For investment cash flow on the very right, gross CapEx, as we mentioned before, JPY 52.4 billion. The investment in Taiwanese Nanya is JPY 78.2 billion.
For the 8th generation as well as 10th-generation bics flash equipment investment is starting. That is also inclusive in this number. On the right, which will be financial cash flow. As we mentioned before, the senior loan full repayment of JPY 407.5 billion is priced in. As a result, on the very right, for FY 2026, it is going to be JPY 791.0 billion. Please move to page 13. This will be surrounding the balance sheet. When it comes to our equity ratio, this has improved quite significantly. Please look at the very left, and this will be the asset side. The left hand will be for FY 2025, and then the right will be the end of the first quarter of this fiscal year.
When you look at the assets, what is increased most significant will be the cash flow that we discussed before as well at the very top, cash and cash equivalent. That is JPY 791.0. For trade and other receivables. Since business grows, this is going to grow as well. That is also on the right at JPY 1,151.6 billion. For inventories. This is increased by JPY 25.5 billion. This is with DRAM as a core. We are strategically trying to ramp up our inventory. Outside of that, there is no significant difference. At the very right, there will be the inventory, days of inventory. It is about 100 days or maybe a little less than 100 days. It is not that much of a change.
At the very bottom, when you look at others, that is the Nanya equity acquisition. That is inclusive in this number. With that, asset is JPY 4,730.5. Excuse me, JPY 4,730.5. On the very right, for the end of this first quarter, the biggest change, it will be the senior loan full repayment. That will be second from the top. This will be bonds and borrowings. This has reduced quite significantly by JPY 400 billion. Furthermore, when it comes to other liabilities, this is on the rise. It is JPY 442.9- JPY 781.4. There is quite a bit of profitability that we booked, meaning we need to pay quite a bit of tax. Corporate tax, which has not been paid yet, is also captured here under other liabilities.
There is JPY 2,404.5 billion in terms of capital, it is going to be 51% in terms of our equity ratio. For the second quarter and third quarter, we do believe that there will be more ramp-up in the capital. Therefore, from a capital perspective, it will be quite robust. Moving on to the next page, please. This will be surrounding enhancing our financial structure, and I would like to discuss more substance. This will be the net debt to equity ratio and exactly how that has been evolving. When you look at the chart on the very left to right, on the very left will be last fiscal year, quarter one, quarter two, quarter three. This fiscal year's quarter one at the very right.
The blue area will be the liability and also capital on the red, and equity on the red. Quarter one had more liabilities, and the same situation lingers into quarter two. Coming into quarter three of last fiscal year, the D/E ratio goes to 80%, there seems to be more of an equity side than the liability side. In quarter four, this reversal improves even further and D/E ratio goes to 39%. In the first quarter of this year, our capital has been ramping up even further, and we have more cash ramp-up. With that, as you can see here, the D/E ratio is now at - 8%, this is how much improvement we have been able to achieve. We are in a total cash positive position. Moving on to page 15, please.
This will be the second quarter guidance. How we disclose the guidance. We used to have a range. However, from the quarter one, we decided to go with just one number. On the very left, we have the quarter one results, and based on this quarter one results, we have been developing the quarter two guidance. On the right will be the quarter-on-quarter difference from quarter one results as well as quarter two guidance. Revenue results, as I said, was JPY 1,767.1 billion, we're saying that guidance-wise it's going to be JPY 2,390.0 billion. Quarter-on-quarter, it's going to be a 35%+ increase. For non-GAAP operating profit for the quarter one results is JPY 1,326.2 billion, this is going to be guided as JPY 1,900 billion.
The margin is going to be 79.9%, and quarter-over-quarter it will be a 43.3% increase. For non-GAAP profit before tax. This is going to be a new disclosure item. In addition to core cash flow, we decided to disclose non-GAAP profit before tax as well. Traditionally, it was basically operating profit minus our profit before tax, and we did not really disclose this number. You needed to have some assumptions, therefore we decided to disclose this number so that it would be a very solid number that you would be able to understand. This will be non-GAAP profit before tax that we decided to add on as a part of the disclosure. Results was JPY 1,291.5 billion. Quarter two guidance, we're saying JPY 1,880.0 billion. Margin-wise, net 78.7%. Increase will be +45.6%. Non-GAAP net income.
Quarter one results is JPY 887 billion and guidance is saying JPY 1,280 billion, margin at 53.6%, and also increase of 44.3% quarter-over-quarter. It's a very high number. As a result of that, moving on to the next item, which is non-GAAP earnings per share. Quarter one results, JPY 1,621.81 billion, which is going to be JPY 2,335.70 billion, and it will be JPY +713.89. The exchange rate is going to be JPY 162 to the dollar. FOREX sensitivity across the three months on a quarterly basis, revenue is going to be JPY 14 billion with JPY 1 change, and then operating profit will be moving by JPY 13 billion. This is exactly how much I think increase we will see with a weaker yen.
Even though it's not described here, when it comes to our net income, there's also R&D investment that we are going to be enhancing, and that will be inclusive as well. It is about JPY 200 billion for the full-year that we are expecting, and therefore it will be BiCS 11 as well as AI inference SSDs. This is going to be a concentrated investment that we will be making, and therefore our R&D investments will go into these 10th and 11th generation BiCS, as well as SSD for inference. That will be the overview for our guidance. Moving on to page 16, please. This is going to be exactly how we are continuing to introduce initiatives in order to boost our corporate value, and one will be stock split. The investors.
We want to make an environment that will be easier for the investors to invest. We want more investor base. We want to expand our investor base, and that will be the purpose of the stock split. It will be a ratio of 3: 1 in terms of our company's common stock split. This will be on September 30th, and effective date will be October 1st. Therefore, after JASDAQ, it is going to be approval. It will be a ratio of 3:1, and then it will be a share buyback, and this will be a repurchasing program. The purpose is to improve our capital efficiency and also enhance our shareholder return. The total number of shares to be acquired will be 30 million shares, and that will be the maximum that we will be considering.
This is going to be 5.5% of the total shares outstanding that we will be considering, it is going to be JPY 800 billion, which will be the maximum for the total amount of acquisition. Acquisition period will be from the beginning of next week, which will be from August 3rd to October 30th of 2026. This is intended that we are hoping to actually purchase from the market from the Tokyo Stock Exchange. These are two things that we wanted to introduce. Lastly, this will be the summary for today. Please go to page 17. Three things. The first will be, Fujikawa explained prior to my presentation surrounding accelerated growth driven by AI demand. In the quarter one, when it comes to our revenue as well as operating profit, we booked something that was on record high.
This was exponential. That growth trend, as we mentioned in the guidance, it will continue into quarter two as well. For agentic AI, with the widespread adoption of agentic AI, the NAND demand growth will be even more robust. It is just still in the incipient stages, and we do believe that there is more to come. When it comes to our major customers, we will have a long-term agreement that we will be signing in calendar year 2028. Most, we would say 50% of our shipment will be covered by LTA if things go well. This is what we are trying to aim for. Therefore, it is a long-term revenue visibility that we are hoping to establish so that we can ensure a deep engagement with the customers. It is not just going to be selling through marketing.
When it comes to NAND, it is a very specific product. Customer specification will be very important for us to actually design and also engineer. Therefore, we hope to create a very friendly relationship with our customers. In light of the AI demand growth, we are hoping to expand our manufacturing infrastructure and capture expanding AI demand. This will be investing into Yokkaichi as well as Kitakami, which will be continuing. The second will be financial strength enabling our future growth investments that I mentioned before as well. For the second and third quarter as well, we do hope that our equity capital will be increasing, and therefore we will have an even more robust financial soundness. Therefore, we will have more flexibility as a company in order to enjoy a very strong position.
Therefore, even if there is a big downturn, we will be able to cope without any issues. That will be a very robust financial standing that we will enjoy. Thirdly, this is something that I did touch upon as well, which will be continued value creation and corporate value boost driven by our strategic priorities. Basically, we want to be friendly to the capital markets as well. We will make growth investments. We did mention CapEx, and that we discussed mainly today. There are three types of investments, next will be R&D investments. That is something that I did mention previously, but it will be in the order of JPY 200 billion. Also human capital investment as well. That is going to be enhanced as well.
Capital efficiency improvement is something that we will constantly work on. We want to boost our EPS. That's something that we are constantly engaging in. We want to make sure that we discuss shareholder return as well, so that we are going to be a very broad capital market-friendly initiative that we do want to execute. Stock split that we mentioned before, that is going to be one to three, and the share buyback will also be conducted with JPY 800 billion as a max. We'll do this in a very agile manner. That will be my presentation. Thank you.
We will move on to the Q&A session. The Head of Finance & Accounting Division, Akira Menju, will be coming online now. Mr. Kawamura, Mr. Fujikawa, and Mr. Menju will be taking your questions. If you have a question, please tap the raise-a-hand icon on the Zoom screen. We will appoint you in order. When you are appointed, please unmute yourself to speak up. If you would like to ask questions in writing, please use the Q&A feature in Zoom. Feel free to submit them in either Japanese or English. Please allow us to limit the number of questions to two per person. If we have time left before closing, we will take additional questions. Please kindly note that because of the time constraint, we might not be able to take all the questions since we will wrap up the Q&A session at 5:00 P.M.
Please raise your hand if you have any questions. From CLSA, Ms. Yoshida, please. Please unmute yourself to speak up. One question. If you have two questions, please start with the first question, followed by the second question.
I am Yoshida speaking. Thank you. Congratulations on the good results, and also that we see the strong numbers that are in the second quarter's guidance. The big growth effects and the ASP, what is the composition of them? Which is the driving force, and what about by application? I'm a bit mindful of the seasonality of the smartphones. Also, what's your view of your business results for the second half of the year?
Thank you for your question. Numbers by application, I will ask Mr. Menju to take that question.
Thank you for your question. Revenue for the second quarter, JPY 2,390 billion, and a 35% Q-on-Q is expected. First quarter, Q-on-Q, 76% up. ASP accounts for 70% as a driver. Significant increase in ASP. This time, 35%. The breakdown of it, the biggest driver should be ASP up continuously from the first quarter. It is expected that ASP will continuously grow. In terms of the volume, last quarter, a low single-digit percentage increase. The second quarter, it is expected to grow as well. FX, JPY 160- JPY 162. Around 1%. These are major drivers behind it. When it comes to the by application, the first quarter, when you look at the volume, low single-digit percent growth. Data center enterprise has driven the entire business.
Data center enterprise, so that we allocated our business volume to that sub-segment. In the second quarter, continuously, the demand will be strong in data center and enterprise, which will drive the volume, and the smarter device as well. Of course, there must be some seasonality, but the volume is expected to grow the second quarter. Thank you.
Toward the end of the fiscal year, we have to carefully see how it will play out. We already presented the second quarter as guidance. When we look ahead in the third and fourth quarters, our production plan remains intact. The sales plan, which might be adjusted a little bit, but overall, it will remain unchanged. The third and fourth quarters, we will follow continuously the strong growing demands, our financial results will follow, we believe.
I would like to confirm one thing regarding the volume. Are you expecting the two-digit growth?
We are not disclosing specifically, so it's really hard to take the question, but we can safely say that it is expected to grow.
Thank you. My second question is about the shareholders' return. Unlike what you said in the past, you seem to accelerate the timing of the shareholders' return. Instead of with the dividend, it seems that you will start with the share buyback, which might be related to the stock split. What's the background of your decision? Also, if there is any update over your future shareholders' returns policies or initiatives.
We kept talking about dividend. Of course, still we are discussing it because we have ample cash flow.
Second half of the year, we assume that it might be possible for us to consider some kind of a dividend pay. Still, that's existing annual discussion on the table. The share buyback. Right now, as you are familiar with, our share price is fluctuating, and right now it's downward. In order to improve the capital efficiency, it's a good timing for us to adjust the number of shares in the market. The share buyback will contribute to the better EPS. Inevitably, it will be the shareholders' return, as well as dividend. By coincidence, because of the share price trend, we decided to take the share buyback first, but our policy remains unchanged. Thank you.
From UBS Securities, Kenji , please.
Thank you. This is Kenji from UBS. This is surrounding hyperscaler customers included. We would like to better understand your status with LTA. I think it may be difficult to mention specific customers, but we would like to understand your progress to the extent possible. Thank you. That's number one.
I would like to invite Fujikawa to respond. Thank you.
In terms of LTA, as Kawamura has introduced before in the presentation, as Kioxia, there are three reasons why we would like to go for LTA. One, we want to visualize a long-term demand, and we want to have a backdrop to exactly how we actually make growth investments. The third will be for enterprise and hyperscaler key players. We want to really have a future-looking business and also a technological business engagement.
These are the three things that we would like to focus on when we actually engage in our LTA activities. When it comes to customers as well as conditions, it is very bespoke and very customized, and therefore, it's not really standardized across the different customers. There's NDA as well, so we can't really discuss any details. At the Investor Day in June, I do believe that Ota, our president, did make some introductions surrounding this area. We would like to aim for 50% of our total shipment volume in terms of the percentage of LTA. There is going to be more LTA engagement that we would like to deepen. 50% is something that we are trying to establish as a goal, and we are pretty much on track in trying to attain that goal. Thank you.
When it comes to the term, exactly how long would it be when it's five to six years? What are you aiming for in terms of your LTA length?
Allowing myself to respond once again, please. Generally speaking, say by 2027 to 2028 is what we are discussing with the customers at the moment. However, most recently, there are further longer-term framework that customers are wanting to sign. These are some voices that we do hear, and we are proactively engaging on that front as well.
Thank you. The second question is surrounding AI inference, how we capture that. AI inference demand appears to be rising. We do believe that there could be a tightening that could happen going into next year. When it comes to AI inference area, how do you see the development?
Most recently, the open source model from China is now starting to become more visible. The U.S. frontier model and also the two big manufacturers inference AI Chinese makers, I personally think that it's not that different. Between the U.S. and Chinese model, is there maybe a change that we will be seeing in the NAND market?
We would like to invite Fujikawa to respond to that question as well.
First, when it comes to AI inference flash memory demand going into the future, as you have rightly pointed out, inference AI is in the incipient stages, as Kawamura had discussed as well. As Kioxia inference AI or agentic AI flash memory demand, we are still in the entrance point of its demand increase. Specifically, when it comes to agentic AI and also storage dependency is going to be increasing significantly with that.
The inference server SSD as well as input output per second, the very high, super high IOPS SSD that we enjoy, that demand is going to be created going forward. Therefore, at this moment, we do believe that the industry analysts or maybe the research companies are saying that there is a demand data that they are introducing. The inference AI growth SSD demand and flash memory demand, exactly how much that is priced in, as you have rightly pointed out, it is something that we do not know. There is, of course, upside potential, we do believe. You mentioned China, the open source, I think more affordable AI model, exactly what that impact could look like.
Jevons paradox is something that is discussed in the industry quite frequently. As Kioxia, when it comes to AI cost reduction, the system cost reduction, and the specification cost reduction, that means that many more users will be using AI. Therefore AI's barrier to entry is going to be lower, and therefore there will be more users that will be able to start using AI. Meaning that there will be more data that will be used and that will be generated. That is a positive cycle that we do foresee. Thank you.
Next from Nomura. Sorry, Ms. Virginia.
Thank you. I'm Virginia from Nomura. Can you hear me? Thank you. I have a question about the production, which might be overlapping from the previous person. I wonder how frequently you have interaction with the critical customers. As we talked about earlier, that open source model for your customers. I think that this is something they have never anticipated, such as AI and others. I wonder if such the new demands are emerging, and if there are such new demands coming from the customers, do you have the customer access so that you'll be able to get that information of such new demands?
Mr. Fujikawa?
Yes. In the U.S., Silicon Valley, Texas, or the Shanghai, Shenzhen in China. In each location, we have the talented sales engineers in each location who have the interaction with customers daily basis or a few times a day. They have interaction with customers in each location. Not just the customers procurement departments, but also they have the deep engagement with the technology department of the customers so that we can detect any change of the project status or the demand change. As you pointed out, any new technology transformation coming outside of that circle. That might have impact, of course. However, that whenever customers detect such change, if they change any plan because of that, they will be able to get information from the customers.
Thank you. My second question about CapEx. It seems that you are in the dilemma situation. Even if you are in the disciplined manner trying to make the CapEx in a prudent way, if there are other of your competitors are aggressive in making investment, they will steal the pie from you gaining shares. If they are aggressive in increasing the capacity, are you trying to follow the competitors or are you trying to get away from the commoditization? Just like a super high ops you are currently developing to be unique.
Well, this is Kawamura. I will take your question. We are not pursuing a market share for making investment.
As Mr. Fujikawa mentioned earlier, we are leveraging the technology advantage, specifically higher sequence speed, or the better energy efficiency, or the higher product quality, we will make investment in those areas. In terms of the customer engagement, we are managing it quite well. We are not selling commodity products, it requires the high level customization on the engineering. The customer engagement is much stronger in Kioxia than any other players. Instead of pursuing the quantity, that we will pursue the quality and profitability. That's our focus. In terms of CapEx, JPY 450 billion this year JPY 470 billion in three years. We will be flexible to be aligned with the market change. Fortunately, we have ample cash flow to cover the CapEx.
Again, that instead of pursuing the quantity, but that we will make sufficient investment in R&D or some specific facilities to cope with market change. We are not necessarily following the competitors and trying to increase the market share by making significant investment.
Thank you for a very clear response.
Thank you.
From Goldman Sachs, Shuhei, please.
Thank you. The first question is surrounding your quarterly earnings. The first quarter as well as the second quarter guidance. For the first quarter, when it comes to your shipping volume, it seems like it has been delayed and running into the second quarter. Exactly how large a scale of that pushback, and can you explain the backdrop to that as well? When it comes to your outlook for sales in the second quarter, it seems like the biggest revenue increase will be supported by the ASP increase. When it comes to what percentage of the shipment volume is already fixed in terms of pricing, and how much percentage is still, I think, fluctuating as a price?
I would like to invite Menju to respond.
For the first quarter volume, it was slipped into the second quarter. Originally, at the quarter end, we were supposed to deliver. However, there was a delivery slip. Gigabyte volume-wise, it is not that significant. Surrounding the second quarter, the volume will be increasing. Exactly how much of the pricing is already set in terms of that question, is that we are currently still negotiating price, so about Thank you. When it comes to the shipment slip, maybe it's a mid-single digit, I think, increase. It's low single digit, maybe that gap portion is what slipped. I'm not saying all of that gap. However, I think that would be the larger portion. Thank you.
The second question is for the CFO, Kawamura-san. Can you give us the backdrop to why you decided to go for a buyback? You did explain.
This JPY 800 billion, how did you come to that? The window is until end of October. Thinking about your liquidity, it seems like a very short period, and it is a pretty large-scale buyback that you are going for in a very limited timeframe. From next year and beyond, how are you observing the business environment? I think that's probably relevant with your decision as well. When it comes to some research companies in the latter half of next year, the NAND demand supply dynamics may be, I think, deteriorating. Can you maybe share with us your view as well?
This is Kawamura. I would like to respond. Thank you. As for the backdrop to why we decided on a buyback this time around, as we discussed.
In the past couple of weeks, especially in the month, there has been a decline in our stock price. If stock prices decline this much, then this is a great opportunity for us to be improving our EPS, and that will be the backdrop. The second, it will be that looking when we discuss our earnings number, we do have quite a bit of a cash flow generation that is ramping up. How do we use this cash flow? First it will go to CapEx as well as R&D. We want to return to employees as well, and we want to allocate well. We still have some cash remaining, therefore, we have about JPY 800 billion worth of share buyback that we would be able to go for. These are the two predominant reasons.
In terms of the short window, we wanted to be very tactical, we wanted to address this very smoothly, that's why we decided to go for this period. In the second quarter and beyond, as we discussed previously, cash is ramping up in a very steady manner at the moment, that's what we continue to expect. Therefore, within this fiscal year, we do believe that there will be further ramp-up in our cash. Then including dividend, the next round of shareholder return is something that we will be deliberating.
For fiscal year 2027 and beyond, in terms of how the demand supply dynamics will look like, when it comes to our observation for fiscal year 2027 as well, we do believe that a very strong situation will continue, though, for cash flow generation in fiscal year 2027, we will pretty much keep the current plan. It will remain pretty much unchanged. Shareholder return buyback as well as dividend will be discussed. For fiscal year 2027, we have no intention of changing our outlook significantly. Thank you.
From BofA, Mr. Hirakawa, what's your question?
Thank you. Can you hear me?
Yes.
Thank you. I am Hirakawa speaking. My first question is a follow-up question from the previous. FY 2027 demand continuously will exceed supply. Looking at the demands, consumer segment, enterprise SSD, the supply, also the Chinese players that are coming into the space. Considering such a situation, how do you see supply/demand situation?
I will ask Mr. Fujikawa to take that question.
As I touched upon while taking the previous questions, according to research companies' demand data, the generative AI and the swimlane, the open flash coming from the generative AI and the agentic AI, that demands are exploding. The question is that how much we are catching up with it in the entire sector.
As Mr. Kawamura said earlier, that we are trying to meet customers' demands in terms of the production and technology development. We are ready to accommodate such customer demands. Not just the Chinese players, but we have to be mindful of the global competition. Kioxia, in that sense, we have multiple locations with customers, including China, that we are offering services to the multiple locations globally. It means that our products are well received or they're appreciated by the key players globally. As you see, the backlog, which shows that how strong AI demands are. Also, that we are leveraging the LTA, leveraging our competitiveness. In that sense, we believe that we'll be able to maintain our technology leadership position then focusing on reliability of our products.
We will continuously engage with the customers under the umbrella of LTAs, we will focus on the investment roadmap so that we'll be able to maintain good position.
Thank you. The follow-up question is regarding data center demands that will increase. Makes sense. There must be some risks, such as the smartphones and consumer segments, which might deteriorate the market. How do you see that potential scenario? How are you going to mitigate such risks coming from those segments?
Again, Mr. Fujikawa will take the question.
Not just NAND flash, DRAM and SoC also, or maybe substrates as well. BOM cost increases. Therefore, the consumer segments, PC, smartphones segments are affected, especially low-end and mid-end models are affected when you see the shipment volumes, which are affected obviously. There is the replacement cycle, which might get longer.
The QLC NAND SSD for data center to replace nearline hard disk. That's an opportunity. Entire market is skewed toward the high-end price, that transition may be slower than expected. The question is that when the supply/demand will be balanced. Nobody see it when. When the ASP is normalized, it means that we'll be able to access such demands, which we cannot access yet this moment. Recently, how people see this situation is that the edge segments is consuming the AI to process the data through the smartphones or PCs. Many people are discussing that. I think that the things that will move toward the edge computing. Even in the consumer segment, the NAND demands is expected to grow. Not to peak out immediately.
Some people say that, for your reference, some people say that it's not going to peak out so quickly.
Let me add some comments. This is Fujikawa again. Yesterday and today, we have made a press release about for edge AI or their on-device AI. For such use cases, UFS 5.0 new standard for such use cases, which ensure high speed, we are continuously developing that interface, showing our technology engagement with the customers is quite strong so that we'll be able to detect early sign of demands. Thank you.
Thank you.
Since it is time, we would like to entertain one last question from Citigroup Securities. Mr. Fujiwara, please.
Thank you. This is Fujiwara from Citigroup Securities. I also have a question, please. You are going for buyback, I think it was addressed as question a couple of times as well. Once again, when it comes to the capital for shareholder return, exactly what is your thinking on that front? You have more excess free cash flow that you want to generate, I do believe that that outlook is looking, I think, very promising because the payback period is quick. Maybe that capital could be seeing a shift or a change from here and on for every quarter. Maybe you will be considering a certain level of shareholder return. Can you just learn exactly what your thinking is on this front as well?
I myself, Kawamura, would like to respond. Specifically, when it comes to cash flow generation, it's not as if it is front-forwarded compared to our plan. It is as per our plan. We are going for an JPY 800 billion max of buyback that we are going to be announcing exactly how we address the defining stock price. We do believe improving our EPS, it's a great timing to do so. These are things that we deliberated. In terms of the JPY 800 billion, this will be shareholder return as well as the next round of shareholder return is not going to be impacted with this JPY 800 billion max amount that we may be buying. The second, third, and fourth quarter cash flow generation is looking very strong.
Therefore, it's not as if there is a very big change that we foresee in terms of the capital that we may be able to use. It's not as if we're rushing into this. There's cash flow generation. Basically, as we mentioned before, there are the three types of investments that we will make, CapEx, R&D, HR investments. We have working capital that we are going to secure, and there will be excessive cash flow that we will be able to generate, and about 50% of that will be allocated to shareholder return. That basic thinking remains pretty much unchanged. Therefore, there is solid cash flow generation, and we would be able to invest within that cash flow.
If there is excess cash flow, even after that 50% as a benchmark, we would like to continue to consider our shareholder return. That will continue to be our policy and thinking. Thank you.
With that, we would like to conclude the Q&A session. For those that are on live streaming, if you try to exit, there will be a questionnaire sheet. This will be very helpful for our future IR activity, we would very much appreciate your populating the questionnaire. We would like to conclude the March end 2027, I think, earnings call for Kioxia Holdings. Thank you very much for attending despite your busy schedules