Sumco Corp

TSE:3436

Earnings Call Transcript 2023-Q2

[Takayuki Komori]

Thank you for your participation today. This is the results briefing for Q2 of the Fiscal Year Ending December 2023.

Before starting the presentation, allow me to confirm today's materials, which consists of three items: The brief statement on consolidated financial results for Q2 fiscal 2023; the announcement regarding interim dividend and the presentation deck entitled Results for Second Quarter of Fiscal Year 2023. This will be a roughly one-hour briefing, which will end at 5:00 p.m.

Next, a disclaimer. The estimates, expectations, forecasts and other future information discussed here and shown in today's materials were prepared based on the information available to the company as of today, and on certain assumptions and qualifications, including our subjective judgment. Actual financial performance or results may differ substantially from the future information contained in this material due to risk factors, including domestic and global economic conditions, trends in the semiconductor market and foreign exchange rates.

We will have presentations today from Representative Director, Chairman and CEO, Mayuki Hashimoto; and Vice President, CFO, Michiharu Takii. Hiroshi Itoh, Executive Officer and General Manager of Accounting is also on hand.

Chairman and CEO, Hashimoto, will discuss our forecast and operating environment to be followed by an explanation of the financial results by CFO, Takii. We have set aside time for a Q&A session as well.

I will now hand over to Chairman, Hashimoto.

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

I am Chairman Hashimoto. I will start with the overview on slide 5 of the presentation. This is a summary of the Q2 results. Q2 earnings exceeded our forecasts, profits in particular overshot our forecast largely as a result of lower costs, particularly electricity costs. Turning to our Q3 forecast, we are guiding for sales of JPY 101 billion and operating an ordinary profit both at JPY 11 billion. Net profit is projected to be JPY 7 billion. Profits are down roughly JPY 10 billion Q-on-Q. The decline breaks down into an increase in depreciation which accounts for close to half of the decline. Lower sales which account for around half with higher materials cost accounting for around 10% CFO Takii will discuss this in more detail later.

On shareholder returns, please turn to page 6. We have reflected the overshoot in the interim dividend raising it to JPY 42 per share. Next page please. Slide 7 shows the market environment for silicon wafers. First looking back at the results for Q2. As stated here, sales were largely unchanged from Q1, for 300 millimeter in both logic and memory, customers continue to adjust production so conditions are soft. Conditions for 200 millimeter and smaller size wafers are similarly weak. However, LTA prices are being respected. On the outlook for Q3, we expect to see a further deterioration in the market. Earlier I mentioned that half of the sequential profit decline of JPY 10 billion was due to the drop in sales, reflecting expected declines in sales of both 300 millimeter and 200 millimeters. The drop in sales reflects a sequential decline in volumes of roughly 10%. On our outlook going forward, as I have been saying to date, this recession is different from previous down cycles. Memory is particularly challenging. Logic is also weak but not as weak as memory. We are seeing an exacerbation of the polarization between winners and losers in logic. There are customers that are aggressively cutting production but others where the cuts have been more modest.

For memory, the downturn is across the board with almost all players implementing significant production cuts. In terms of applications, automotive applications, such as MOSFET and IGBT have been firm overall, with some customers indicating their desire to increase LTA volumes. Up to now, in down cycles typically, the weakness is across the board with all players struggling, but this time is uncharacteristically different. Any recovery is unlikely to be across the board. Instead, we believe the timing of recovery is likely to be different for individual players.

Page 8, please. This slide shows the wafer trend for 200 millimeters by quarter. As shown here, there was a decline of around 10%. Page 9, please. If you look at the trend for 300 millimeters optically, it appears as if there was an increase from Q1 into Q2. This is probably a reflection of some customers choosing not to buy in March but instead inspecting and confirming acceptance in one fell swoop in April. In reality, I believe conditions in terms of the correction were similar for Q1 and Q2.

Page 10, please, as shown here, inventory levels continue to rise getting close to a peak in our view. Page 11, please, if we look at the breakout between logic and memory, we have seen a relative flattening in logic which we hope is an early sign of recovery. However, from memory, inventory is continuing to rise sharply. Customers have indicated they will undertake substantial production cuts, but inventory is unlikely to decline until these cuts are implemented. That said in Q3 and Q4 with almost no exception, customers will make major production cuts. So we hope to see some improvement in conditions.

Paage12, please, this chart plots trends for inventory demand and price for 300 millimeter logic use epi wafers, the blue line is demand and the orange line is inventory. In the case of logic, usually when the blue line which represents demand rises, inventory represented by the orange line falls a statement of the obvious perhaps, but this is the typical pattern for epi wafers. Furthermore, normally after four to five recessionary quarters, the market recovers. Page 13, please. However, from memory if we look at the period between 2010 and 2015, you can see that despite an increase in demand inventory only starts to decline in 2016. As you can see, inventory also fell in 2021. But is now rising rapidly. As I touched upon earlier, memory fabrication is typically targeted production. So wafer demand tends to lag, there is a time lag before a recovery in the memory market is reflected in wafer demand. Rather than being made to order memory is made to stock in anticipation of demand.

Memory makers also tend to carry inventory of polished wafers. All of these factors contribute to a time lag before a market recovery flows through to wafers. Back to page 12. In contrast to this as shown on page 12 logic is made to order so wafer trends typically reflect market trends in real time, with wafer demand recovering when the market recovers. This is the backdrop to our view, that logic should recover in the first half of 2024 and memory in second half 2024. However, we believe the recovery in wafers for memory, given the time lag is likely to be at the end of 2024. For logic, our current assumption is that the recovery in wafers should be mid-2024. That said it is very difficult to read the geopolitical situation as a result of the US continuing to tighten its sanctions against China. Consumer appetite for items like smartphones in China, which is already waning could be dampened further. What happens here could have implications for the market going forward. Developments in Ukraine could also impact future trends. Currently, it is very difficult to read future trends.

Page 14, please. This is the outlook for supply demand, including historical data which we always share with you. Our forecasts are based on the blue dotted line, which is PPP GDP growth. If you look at the historical trends, you can see that the trend diverges from PPP GDP growth from time to time only to revert. Currently, there is a significant divergence to the downside from PPP GDP. We don't know when the demand forecasts will revert to the blue dotted line. But for now, we think it may take some time. Maybe conditions will be suboptimal until around 2025 or 2026. Our assumption is that from the second half of 2023 into 2024, in particular, is likely to be very challenging. This completes my section of the presentation. I will hand over to CFO, Takii to talk about details of our earnings.

[Michiharu Takii, Chief Financial Officer and Vice Chairman]

I, Takii, will present the earnings in more detail, please turn to page 16. This slide shows the results for the first half. First half sales rose JPY 13.2 billion, up slightly year-on-year, but OP was largely flat year-on-year down JPY 3 billion. Ordinary profit was also similar. However, we generated extraordinary profits in Q1 on the negative goodwill associated with the acquisition of the polysilicon business. This resulted in a significant year-on-year increase in first half net profit attributable to owners of the parent. As was the case in Q1, CapEx is up substantially year-on-year on continued Greenfield investments. On the back of new CapEx depreciation also rose. There was a positive tailwind from the weaker yen at slightly below 135 yen to the dollar. OPM and other metrics are as shown at the bottom of the table.

Please turn to page 17. This is the analysis of changes in operating income. First, the sequential analysis, Q2 operating profit dropped JPY 5.1 billion Q-on-Q. We show the analysis below. Costs such as materials cost improved JPY 1.3 billion Q-on-Q, but depreciation increased Q-on-Q. Sales related variance fell JPY 5 billion Q-on-Q mainly as a result of lower sales and production volumes. Sales as shown above rose JPY 0.8 billion Q-on-Q but this is the result of a JPY 5 billion contribution from the polysilicon business from Q2. Sales of wafers fell as a result of production cuts. This is why sales related variance was minus JPY 5 billion. The Q2 JPY 20.8 billion OP was JPY 3 billion better than expected. The main factors were a slightly lower than expected depreciation level, which came in JPY 0.6 billion below plan, the weak yen and lower than expected costs.

We had expected an increase in materials and maintenance and repair costs, but we're able to make progress with cost reductions. Hence the JPY 3 billion sequential overshoot of our expectations for OP, on the right we show the year-on-year change for first half. If you look at the waterfall chart costs rose a significant JPY 15 billion year-on-year. Materials costs increased JPY 5 billion , electric power unit costs rose JPY 6 million, labor expenses were up JPY 2 billion and R&D costs up JPY 1 billion. The increase in labor expenses includes a rise in base pay, depreciation also increased year-on-year. On sales related variance, the contribution from the rise in LTA prices from the start of a new fiscal year was JPY 13 billion , but volume decreased by JPY 9 billion for a net positive of JPY 4 billion. Forex had a positive impact of JPY 11 billion.

Page 18, please. Slide 18 shows the balance sheet and cash flow statement. Cash and time deposits on the balance sheet fell more than JPY 50 billion. I will explain this in more detail in talking about the cash flow statement. Raw materials and supplies rose JPY 20 billion, the impact of the acquisition of the polysilicon business pushed up polysilicon inventory by JPY 3 billion. Tangible assets increased significantly on the back of the increase in CapEx. Interest bearing debt increased JPY 24.8 billion to JPY 166.1 billion. If we look at the capital account, retained earnings increased reflecting the impact of higher net profits. At the bottom of the table, we show the equity ratio and a number of other metrics for your reference. On the cash flow statement to the right, operating cash flow was a positive JPY 53 billion. Capital expenditures were slightly less than JPY 150 billion on an acceptance basis as mentioned earlier, proceeds from the acquisition of shares of subsidiaries, which refers to cash and cash equivalents acquired through the acquisition of Mitsubishi Materials, polysilicon business was a positive JPY 9.9 billion , pushing up the overall cash on hand. Free cash flow was a negative JPY 65.4 billion. After dividends paid, the repayment of debt and new borrowings of JPY 40 billion, we get to the net increase in cash and time deposits highlighted earlier of JPY 51.1 billion. The net increase in debt was JPY 24.8 billion.

We chose to increase our borrowings to have some breathing room when taking into consideration our future needs. Page 20, please. On page 20, we show our forecast for Q3. As shown here we project Q3 sales of JPY 101 billion down JPY 10 billion Q-on-Q. Nine months sales are largely unchanged year-on-year. The forecast for Q3 operating income is JPY 11 billion halved Q-on-Q. As a result, we project Q3 net profit of Q-on-Q 7 billion. Nine month net profit is projected to be up year-on-year reflecting the extraordinary gains from Q1. The Forex assumption is 138 yen to the dollar.

Page 21, please. This is the analysis of change in operating profit. First, the sequential change. The key factor for Q3 is a decline in wafer volumes, which is why we expect sales to fall JPY 9.7 billion. We do expect a slight tailwind from a weaker yen. In looking at change in operating profit in the bar chart below, we expect materials unit costs to rise slightly in second half. Depreciation is also expected to increase. Under sales related variance, we expect volumes to fall reflecting the marginal profit impact of the JPY10 billion drop at the top line. We expect Q3 OP to fall JPY9.8 billion Q-on-Q.

Looking at the analysis of year-on-year change for nine month OP on the right. The structure is similar to the sequential change. Costs are expected to rise JPY20 billion year-on-year. This breaks down into JPY8 billion from materials costs. Another JPY 8 billion from electricity costs, JPY2 billion plus from labor costs and JPY1 billion plus in R&D costs. Depreciation is also expected to increase year-on-year. For sales related variance, the impact of higher prices on LTA kicking in from this fiscal year is around JPY 18 billion. But the impact of volume declines is substantial for an expected negative year-on-your impact. On a year-on-year basis, Forex impact is projected to be a positive JPY11 billion on the weaker.

Page 23, please, the trend is as shown here. Sales have fallen on the significant impact of lower volumes. We are also being impacted by factors pushing up costs. Since the beginning of the year, we have seen costs rise by more than JPY20 billion , hence the EBITDA decline as shown here. This completes my remarks. Mr. Komori?

[Takayuki Komori]

Thank you. We will now open the floor to questions.

[Operator]

[Operator Instructions]

[Takayuki Komori]

We will start with Mr. Enomoto of BofA securities.

[Takashi Enomoto, BofA Securities]

I would like to ask about the investment scale of the new plant where you will receive a subsidy from METI. How did you decide on the scale of investment? I understand the subsidy is around one third of the investment amount. So I feel that you could have chosen to go for a larger scale investment. Also, I know that 2029 is a long way off. But what is your view for demand at that time, or the potential for locking in customer contracts in relation to this plant? Please describe the background to your decision.

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

First of all, fundamentally, our industry is a growth industry as you know. So there is an ongoing need to expand capacity at some point. The Kubara site is already full with no room to expand with another physical structure. So we had already been looking for potential sites. So we were already looking when the possibility of a subsidy emerged. In terms of scale, the market is expected to grow over time, but there is no point in receiving financial support for growth too far in the future. We believe we should base our view on a reasonable volume level. Our decision on scale is a function of what we felt to be a good balance of various factors. It's true that a bigger subsidy would be nice, but we felt it would be prudent to make reasonable assumptions on what we know now.

[Takayuki Komori]

Next is Mr. Ikeda of Goldman Sachs.

[Atsushi Ikeda, Goldman Sachs]

I have a question about your medium term view. As we see progress on generative AI, I am interested in the impact this will have on the market and some gross earnings. It would be helpful if you can comment from both a quantitative and qualitative perspective. With regard to logic epi wafers because of TSMC’s constrained on capacity due to hiring this year, GPU capacity is probably around 8k. However, under Goldman's best case scenario, this could rise roughly tenfold to 70k by 2025. Obviously, CPUs would also be necessary and if you include silicon interposers, then scale could be quite significant. With regard to memory, we also expect server requirements for DRAM to increase eightfold compared to normal servers. Recently, some equipment makers have raised their forecast. It is still quite challenging to project but given that Sumco is actively using AI itself and you are the leading player in logic epi with a share of close to 70% in leading edge, what is your view and what are you hearing from your customers? What do you think will be required for leading edge epi at two nanometer and below from GAA? Also, can you discuss the characteristics required for silicon interposers?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

That is a very tough question. If we look at the H100 GPU chip for AI, it is quite large and costs about JPY 4 million to JPY 5 million per chip. If we think of this from the perspective of wafers, assuming a single data center needs 10,000 servers and new builds for data centers are running at about 100 per year. This implies a need for 3 to 4 million chips per year, which translates into a modest 100k wafers per year. It is surprisingly small. However, this is only for GPUs. If you include interposers and other things, then it should be more. So we do have expectations, but visibility is still poor, specs for interposers and carrier wafers are still not determined. I do think they will be used but it is difficult to quantify at this stage. There is also talk of wafer bonding and memory which would require two wafers. So there is a lot of talk with some makers expressing concern about capacity in the future given the current downturn.

At this stage, it is still difficult to quantify. Qualitatively, the number of wafers should rise on AI and wafer bonding in both logic and memory. But timing and volume is not clear at this stage. Given we are currently in a recession, this is not the time to be thinking about a substantial increase in production capacity. Hope this answers your question.

[Atsushi Ikeda, Goldman Sachs]

What about Crystaline alignment at two nanometers for GAA? Your peer is trying to win back some ground. But how are you doing?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

It's likely the GAA will come but probably further out. There are differing degrees of difficulty, the 110 is the toughest. Our understanding is that everyone is struggling and that it is a long way away. We do expect GAA to emerge at two nanometer and below. Of the three leaders probably the well-known player is likely to be the front runner.

[Takayuki Komori]

Next is Mr. Nishiyama of Citigroup Securities.

[Yuta Nishiyama, Citi Group]

Looking at Page five the overshoot versus your forecast for Q2 OP was larger than the overshoot for sales at JPY3.8 billion versus JPY 2.7 billion. Can you explain the overshoot using the waterfall chart? The magnitude of the cost overshoot seems large, especially since electricity costs are likely to be high in the summer. Can you also discuss the sustainability of your cost reductions.

[Michiharu Takii, Chief Financial Officer and Vice Chairman]

With regard to the JPY 3.8 billion overshoot to Q2 OP costs came in JPY3 billion better than expected and depreciation was JPY 0.6 billion lower than projected. In addition, the weekend had a positive impact of slightly less than JPY 1 billion. The negatives were a lower than expected production level. So there were some shipments from inventory underpinning the overshoot on sales. So shipments from inventory were up, but the production level was down.

[Yuta Nishiyama, Citi Group]

Thank you. You say Q2 depreciation was JPY 0.6 billion lower than expected. For Q3, you had previously indicated the Q-on-Q depreciation was likely to be up around JPY 5 billion. But you are now projecting a JPY 3.8 billion Q-on-Q increase. So it appears you expect to see lower than expected depreciation in Q3 as well. Does this reflect a slowdown in the pace of ramping up new facilities? Some peers have admitted that they are adjusting timing to reflect current demand conditions. Please comment.

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

We are investing in two locations, FST where the main focus is PW and at Kubara, which is focused on leading edge epi. Progress on the Japan plant is tracking in line with schedule and, in fact some of the facilities are already in the process of ramping up. The decline in depreciation reflects only a very slight delay in the timing of completion and is not an intentional move on our part. On PW, however, we are intentionally slowing down the expansion process to match market conditions where possible.

[Yuta Nishiyama, Citi Group]

In terms of the ramp ups at Omura, which is slated for full capacity operations by the end of 2023 and Imari, which is expected to reach full capacity in the June quarter of 2024. Should we expect a linear increase in production?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

At a minimum what we can say is that we will be starting up the operation of new equipment. As to whether it will be linear or not. It will depend on the customer. Customer certification plans are already in place, and we are making steady progress on the certification process.

[Takayuki Komori]

Next is Mr. Okazaki of Nomura.

[Shigeki Okazaki, Nomura Securities]

I would like to ask about costs, of which the key elements are materials, electricity, labor and R&D costs. Can you talk about near term conditions and your thinking? I believe electricity costs were likely to fall slightly. On materials costs, am I correct to assume that while there was a slight decline in Q2, you are expecting an increase in Q3. On depreciation, do you expect to come in line with your stated plan for the full year? How are you thinking about 2024? I do understand there may be things that are still up in the air such as at FST. But if you can talk about your current image for 2024, it would be helpful.

[Michiharu Takii, Chief Financial Officer and Vice Chairman]

On materials costs, we had received a number of requests for higher prices for a variety of materials. So the sequential increase in Q3 may simply be a push out from Q2. In other words, Q2 was better than expected, but there will be a sequential increase into Q3. We believe electric power unit prices are unlikely to rise much from here, as you suggested. On labor costs, we are doing some hiring for the greenfield expansion, but the increase does not reflect higher labor unit costs. As such relative to first half, I wouldn't expect to see a huge increase. For depreciation, typically Q4 tends to increase on the back of completion of inspections ahead of the end of the fiscal year. So we expect to see a JPY 5 billion increase from Q3, although it's possible that some of that could get pushed out. Based on the JPY 5 billion increase in Q4 that would bring the full year total for 2023 to JPY 75 billion. In 2024, we will still be making investments for greenfield so depreciation is likely to rise to perhaps around JPY 100 billion for the full year, even reflecting the moderation of progress at FST.

[Takayuki Komori]

Next is Mr. Yoshida of CLSA securities.

[Yu Yoshida, CLSA Securities]

Based on today's comments, it appears that the environment is likely to be tough through 2024, particularly with depreciation expected to be up year-on-year. It is a little early, but what is your view of 2025? And how the environment could change? What is your current outlook for 2025 in terms of volumes, price and depreciation.

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

I expect volume to recover in 2025. Depreciation in 2025 is likely to be flat, but should decline sharply in 2026. Prices should remain flat reflecting the impact of LTAs.

[Michiharu Takii, Chief Financial Officer and Vice Chairman]

One additional comment on depreciation, given the bulk of greenfield investments are expected in 2024. Depreciation will kick in during the 2024 fiscal year. So it's possible that depreciation in 2025 could be up slightly.

[Yu Yoshida, CLSA Securities]

I see. So in 2025, the rebound in volumes should offset a slight increase in depreciation.

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

Yes, I hope so. I don't think the recession will persist into 2025. So I think it is reasonable to expect a recovery in 2025.

[Takayuki Komori]

Next is Mr. Miyamoto of SMBC Nikko.

[Go Miyamoto, SMBC Nikko]

I have a question about price trends. You indicated on page 7 that LTA prices were being respected. A peer has indicated that spot prices have fallen. I know that Sumco does not have much spot business in 300 millimeter in Japan. But what is happening with FSTs 300 millimeter and 200 millimeter spot at Sumco?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

Our non-LTA business in Epi there isn't really a market for spot and 300 millimeters right now. With LTA customers now negotiating to push out delivery times for wafers under LTIs, there isn't a need to buy in the spot market even if prices were cheap. To the extent there is a spot market for 300 millimeter transaction volumes are extremely thin to the point where it really doesn't warrant discussion. We do have some non-LTA business in 200 millimeters, we are seeing some price erosion in the spot market here. However, there are differences by application. For instance, the spot market for automotive use 200 millimeter is not down very much. But there have been price declines on simple PW wafers in the spot market.

[Go Miyamoto, SMBC Nikko]

I see. On the 200 millimeter wafer price declines. Do you expect prices to start falling in the September quarter? And how long do you expect the downtrend to continue?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

Prices are already falling, but actually it varies by product. There are some 200 millimeter applications where there are wafer shortages. For these products, there are instances where prices have actually gone up. So it's difficult to generalize. However, prices on the simpler products are falling. In terms of when we might see a change, we don't know, it will be up to the market. In the 200 millimeter market., generally I would expect a recovery to happen at around the same time as the 300 millimeter market. Up until that point, I would expect prices on simply PW to fall and they are already falling. That said the magnitude of the price decline is not huge to begin with market capacities fixed and the likelihood of the customer business in 200 millimeter going away is limited. 150 millimeter business is likely to go away but not 200 millimeters. The customer's expectation is that it will be challenging to source in demand wafers. So there isn't much negotiation on price for such items. In some instances, customers have accepted price hikes. So it's a tough market to generalize.

[Takayuki Komori]

Next is Mr. Yamada of Mizuho Securities.

[Mikiya Yamada, Mizuho Securities]

Earlier you mentioned that many players were struggling with GAA 110. Do you think we will see the adoption of true GAA with layers of germanium and silicon in 2025 or ‘26? I believe that Sumco is the front runner in terms of wafers. If Sumco can supply wafers, I think we could see the market emerge. What is your view of timing? Also on [inaudible] such as silicon interposers, you commented the specs were not set yet. But I believe that even if the specs are not fixed, it is likely that they will be low spec. I think the emergence of interposers could be supportive in terms of improving wafer supply demand. But is that the right way to think about this?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

My view would be that GAA 100 is coming soon. For GAA 110, tough to say although the plans are for 2025. But I have also heard that some projects have gone back to the drawing board. So progress has been slow. That is the case even for the front runner. So my view is that GAA 110 in 2025, or 2026 would be tough. 110 is very tough for Sumco as well. We haven't seen much in terms of sample shipments to this point. My sense is that it will take time. However, compared to 100, theoretically 110 is much more efficient. So I think work will continue. The bar for wafers in terms of Crystaline alignment and planarization is very challenging. We have managed to fabricate wafers, but it is extremely difficult. On interposers carrier wafers and wafer bonding, I think all of these technologies are coming at some point, but at this stage, there has been virtually nothing. I do think that wafer bonding is on the verge, but with interposers there have been lots of talk to this point. But the front runner is not keeping up with its plan and there is no clarity on carrier wafer specs. I have thought that ultimately interposers would be easier. Wafer bonding is easy. So I would expect this to see the light of day. In fact, some customers have already asked if we would be able to cope when this happens, given that wafer volumes would double. This suggests this technology is coming. My sense is that there will be a certain level of volume in 2024 or 2025. There is a lot of R&D in many areas happening at the same time. So beyond a recovery in the underlying market, we are seeing a shift to structures that will require more wafers. Given this my view is that 2024 will be the bottom for wafers.

[Mikiya Yamada, Mizuho Securities]

If interposers come as [inaudible] rather than EMIB then I would expect to see high volumes. I have high expectations for Sumco.

[Takayuki Komori]

Our final question will be a second question from Mr. Nishiyama from Citi Group.

[Yuta Nishiyama, Citi Group]

I have a question about the JPY 225 billion investment in a new plant in Saga Prefecture. I understand you are receiving a JPY 75 million subsidy. As you negotiate LTAs for the new plant going forward, will you have to discount for the subsidy and therefore accept a lower price point?

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

To reiterate, the timeframe for the new plant is 2029. When we looked at the historical trend for wafer growth, we felt entering into a huge investment would be time consuming. Therefore, we felt it would be reasonable at this stage to focus on a physical shell and the equipment to populate it for an initial volume level, as well as facilities for [inaudible] raising. This was our thinking in determining investment size. We felt this was a reasonable approach rather than trying to factor in developments that were significantly further out in the future in accepting the subsidy. Also, although the subsidy we receive will reduce our own investment, it will not have an impact on our selling prices. Our peers are eligible for hefty subsidies in the US under the Chips Act or in Singapore. Market price is market price. Our investment is not that large and depreciation as a percentage of cost is not huge. We apply the declining balance method over five years. So anything after that is running costs. So there is no need to accept cheaper prices, just because we received a subsidy.

[Takayuki Komori]

Thank you, everyone for joining the Q2 2023 results briefing. We are grateful for your participation today. We will end the meeting here.

[Mayuki Hashimoto, Chief Executive Officer and Chairman]

Thank you.