Tokyo Electron Ltd

TSE:8035

Earnings Call Transcript 2019-Q1

[Kenichi Hirayama, executive]

So it's time for us to start Tokyo Electron's financial announcement for the first quarter.

Thank you very much for joining us despite your very busy schedule and very, very hot weather. I am Hirayama of IR Department acting as today's moderator.

Now, we'd like to introduce the attendees on our side. Mr. Tetsuo Tsuneishi, Representative Director, Chairman of the Board; Mr. Toshiki Kawai, Representative, Director, President and CEO; and next, Mr. Ken Sasagawa, Vice President, Accounting Department.

Now first of all, Mr. Sasagawa, Vice President, Accounting Department, will make announcement to our presentation about the consolidated financial summary. So Mr. Sasagawa, please.

[Ken Sasagawa, executive]

Good afternoon, everybody. Thank you very much for joining us despite your very busy schedule. I am Sasagawa, Vice President of Accounting Department. I'd like to talk about the consolidated financial summary of the first quarter of the fiscal year, ending in March 2019.

First of all, you can see the highlights of the first quarter. As you can see over here, the Q1 results showed good progress, in line with the plan for the first half of this fiscal year.

For the second quarter, new equipment sales plan, more than 60% of SPE and 100% of FPD tools have been shipped, waiting for the installation. Therefore, we are on the very good progress as well.

Second point, the SPE sales remained very strong due to increasing demand for data centers. Similarly, for FPD as well, the investment for G10.5, in particular, has been very strong, and we can see the very good progress as well.

Let's go to the next slide. So this shows the quarterly financial summary. On the right-hand side, you can see the first quarter results. The net sales was JPY 295.5 billion, decreased by 17% from the fourth quarter. SPE sales was JPY 280.4 billion, and FPD sales was JPY 15.1 billion. The corresponding gross profit was JPY 122.4 billion. Gross profit margin was 41.4%. The margin was decreased by 1.7 percentage points. Because of the decrease in net sales, the proportion of the production cost was increased. That is the reason why we have the decrease of 1.7 percentage points.

And SG&A expenses, JPY 50.0 billion, and operating income is JPY 72.4 billion. Operating margin was 24.5%. Income before income tax is JPY 75.2 billion, and net income was JPY 55.7 billion. And R&D expenses, capital expenditures and depreciation are shown on the slide.

Now you can see the financial performance on the graphical basis. As you can see on quarter-to-quarter basis, both net sales and income goes up and down. However, when you look at every half year or every year, you can see the steady increase in trend overall.

Next slide, please. Here, you can see the segment information. Our segment income is based on income before income taxes.

Let me start with SPE. SPE sales was JPY 280.4 billion on the first quarter. The segment income was JPY 78.3 billion. Segment profit margin was 27.9% compared with third -- fourth quarter. Marginal profit ratio product increased. However, as I said earlier, because of the proportion of fixed cost increases against sales, segment profit margin declined. [ Absent to ] last year, the R&D expenses, some of them are included in the adjustment there as transferred to the SPE segment. That's the reason why slight decline was observed.

Next, FPD segment. Sales was JPY 15.1 billion. Segment income was JPY 2.4 billion and profit margin was 16.2%. Again, just like SPE, because of the reduction in sales, proportion of fixed cost increased. Because of that, profit margin declined from the fourth quarter.

For FPD over here, we are providing high value-added products only. So in this fiscal year, we are going to achieve the operating profit margin target of 20% for the midterm business plan. And composition sales is shown on the right-hand side.

Next slide, please. This shows SPE sales by region. As you can see over here, for the first quarter, as for the volume for sales, the first one is in Korea because of the very active investment in memory devices. And increase in sales is observed in China as well. Growing ratio is also big in China, the pink color on the bottom. In Japan, you can see increasing trend of sales as well. Both of them, memory investment are the primary reasons.

Next slide shows the new equipment sales by application. As I said earlier, the Korea, China and Japanese memory customers are major investors, so the proportion for nonvolatile memory was 49%. DRAM accounted for 30%. Memory proportion was rather high in the first quarter compared with others.

Next slide shows the Field Solutions sales every quarter. As for the first quarter, JPY 63.2 billion; SPE, JPY 60.6 billion; and FPD sales was JPY 2.5 billion. Again, you can see some up and down by quarter. But when you look every 6 months, you can see steady increasing trend for the Field Solutions sales.

Next shows balance sheet. On the left, you can see assets. The total assets was JPY 1,185,000,000,000. Compared with previous quarter, it declined by JPY 17.7 billion. On the top, cash and cash equivalents declined by JPY 18.2 billion from the fourth quarter.

The major breakdown should -- there are quite a few breakdown for that, and please see the financial review for the details. But major one is payment of the dividend, JPY 56.9 billion; and accounts receivable, almost the same level as the previous quarter.

Inventories, compared with the end of fourth quarter, JPY 381.6 billion, declined (sic) [ increased ] by JPY 37.5 billion because we are going to prepare for the second quarter. And you can see, we expect a high level of sales. Because of that, we have the -- some increase in inventories.

Next, on the right-hand side, you can see liabilities and net assets. For liabilities, in total, JPY 412.7 billion, declined from the fourth quarter by JPY 18.5 billion. The net assets was JPY 772.3 billion. The first quarter equity ratio was 64.5%.

Next slide shows inventory turnover and accounts receivable turnover. The bar graph shows the sales, and inventory turnover was 117 days on the first quarter, increased by 6 days from the end of the previous quarter. Once again, this is because of the increase of the inventory waiting for the installation at the customer sites. Accounts receivable turnover was 46 days, decreased by 6 days from the previous quarter.

This slide shows cash flow [ obviously ] as shown on the slide. The cash flow from operating activity was JPY 51.4 billion. Next, cash flow from investing activity was minus JPY 12.1 billion, mainly for the capital investment. And cash flow from the financing activities, minus JPY 56.9 billion; again, the payment of the dividend was a major factor. Free cash flow was JPY 39.3 billion. And finally, the cash on hand was declined by JPY 18.2 billion from the previous quarter, amounting to JPY 355.5 billion.

Thank you very much. That's all from me.

[Kenichi Hirayama, executive]

Now next, Mr. Kawai, our CEO, will make a presentation on the business environment and financial estimates. So Mr. Kawai, please.

[Toshiki Kawai, executive]

Thank you very much, and good afternoon -- good evening, I should say. Once again, thank you very much for joining us despite the very tight schedule and hot weather. As Mr. Hirayama said earlier, I'd like to make a presentation on business environment and financial estimates.

This slide shows the future market outlook of WFE by application. In May, when we make a presentation of the midterm management plan, we talked about very similar issue. The new applications, having semiconductors and the servers for data storage and high speed calculation, are increasing rapidly. Therefore, WFE market is now exceeding $60 billion. That's a new phase that we are getting into.

This shows business environment as of July 2018. Let me start with WFE CapEx. The WFE investment in calendar year 2018 is expected to grow by 10% to 15% on the year-on-year basis, driven by active demand for memory devices for data centers.

Next, flat panel production equipment CapEx for the TFT array process. For this, the FPD production equipment investment is expected to remain as high as last year due to significant growth of CapEx for G10.5 despite adjustments to the investment in small/medium panels for mobile devices. In the previous time, I talked about 10% increase for flat panel display CapEx. However, this time, we just express it as high as last year just because a part of [ appreciate ] of the G10.5.

Now we can see the WFE market growth and business opportunities by application for this calendar year. For DRAM, last year, it increased by 70%. On top of that, the CapEx for DRAM is expected to grow by 60% to 65% on the year-on-year basis.

For DRAM, actually, there is a tight supply situation. Supply cannot catch up with demand, therefore, 70% of the investment is to increase capacity. Also, in order to increase the big capacity for wafer, 80% of the investment is for the leading-edge devices. For DRAM, it's driven by the devices for servers.

As for the nonvolatile memories, again, last year, about 70% increase was recorded. For this year, CapEx for nonvolatile memory is expected to be similar to the last year level or increase by 5% on a year-on-year basis. For NAND investment, the leading-edge 9X layer generation is the major area for investment. In particular, demand of SSDs for data centers and PCs will grow farther more.

For business opportunity, high value-added etching and cleaning processes are the area that we can differentiate ourselves from the competitors. You can see a good progress here. At the same time, for film deposition and high-speed testing system, we are, in all products, we are getting higher positions. And we can have good progress, and we are enhancing our efforts to further increase our position.

For logic and foundry, the CapEx for logic/foundry will drop by 10% or remain unchanged from previous year. For leading-edge 10-nanometer or beyond devices, accounts for about 50% of the total investment. For this year, for 7-nanometer node, high-volume production will start and also investment to the pilot production of 5-nanometer node also starts. Compared with 3 months ago, there are slight push-out trend in the investment. However, semiconductor demand does not get decreased, but we should say that it's because of the technology migration. That's how we understand the trend of the slight pushout.

Under those circumstances, when you look at -- by each application on the quarterly basis, you can see some increase or decrease, up and down trends. But when you look at the entire big picture, there is no change as semiconductor demand continued very strong. So midterm or long-term, demand remains unchanged from the previous report announcement.

Now you can see the business progress of this fiscal year, fiscal year ending in March 2019. The SPE business strategy is being implemented as planned. For etching, 3D NAND slit process. For DRAM, combined patterning and interconnect process are very strong. We can increase share, and we are going to increase share -- sales as well. For film deposition as well, this year, we are focusing on [ same-batch LOD ]. We have won several certification of the POR for the volume production. You can make some contribution to the next year's sales as well. For cleaning process as well, the process for the pattern collapse prevention, we can get the POR as well.

As for the flat panel display production equipment, as Mr. Sasagawa said earlier a little bit, high value-added products, PICP etching products, is now getting into the general -- Generation 8.5. Last year, G10.5 was a major area. We increased our position by gaining some PORs. But this year, the mass production of G10.5 started, So FPD production equipment profit margin is now increasing right now.

For Miyagi factory, for etching equipment, we have the automated warehouse system, which was supposed to start this operation in June, and actually, operations started in June. And second new development building is supposed to be completed in September. And this construction of the new development building is under progress. And new production line is planned to start operation in November this year. Not only this year, but also 2019 and 2020, you can address the demand for the future.

Now I'd like to talk about the financial estimates for the year ending in March 2019. There is no changes from the previous announcement. As I said earlier, by application, you can see some increase or decrease. However, by and large, there is no major changes. There is no changes at all. We expect the sales growth to exceed market growth and generating record-high profit for the third consecutive year.

Next, you can see SPE Division new equipment sales forecast, sales by application. For Q1, we -- sales of Q1 were in line with the plan, and more than 60% of Q2 sales target has been shipped. As you can see, beyond Q2 and second half of this year, we can expect high-level investment for memory devices.

R&D expenses and CapEx plan. Again, there is no change from the previous announcement. Midterm, we have JPY 60 billion-plus market trend, so we are going to be active in investment for further growth of our company.

Next slide, the final slide. The last slide shows dividend forecast. Again, there is no changes. And we don't have any change in the dividend strategy, so the plan is just the same as previous announcement.

Thank you very much for your kind attention.

[Kenichi Hirayama, executive]

Now I'd like to entertain questions from the floor up to -- until 6:00. I'd like you to limit the question by 1 per person with a follow-up question. Could you identify yourself by your name and affiliation? And please make it sure to speak accurately and briefly because today's financial announcement will be uploaded to our website. So could you raise your hand if you have any questions, please? Yes, the gentleman in front, please.

[Tetsuya Wadaki, analyst]

I'm Wadaki from Nomura Securities. The first question, only one question, so actually, I am -- I think the -- your company is very bullish, and I am very happy to see there is no change to the financial estimates, but there are some concerns. Therefore, I'd like to ask one question. Actually, I interviewed about 10 suppliers. And actually, overall, the suppliers said the number of orders, amount of order have decreased by half. But when I look at Samsung, Micron and Intel, there are some pushout announcements. And when you look at the entire pushout plan, it's within this fiscal year. So there should be no problem as far as this fiscal year is concerned. But actually, you haven't changed your financial estimates. You said there is no problem. And could you let us know about the reasons why you didn't change the financial estimates. First, I'd like you, Mr. Kawai, to let me know your opinion. After that, I also want to get some comments from Mr. Tsuneishi as well.

[Toshiki Kawai, executive]

Thank you very much. I am Kawai. I'd like to give you my answer first. For this year's growth, for SPE, the 15% increase is expected in the past, but this time, it's a 10% to 15% is the growth rate for SPE in today's announcement. For -- we have been getting questions that our focus is too bullish. That's the question I myself received in the past. But actually, I am talking with customers. Based on that discussion with customer, we announce our financial estimates. When you look at the recent information, I do understand there's some trend information coming up from the industry. However, in the shareholders' meeting, General Shareholders Meeting, I got the question. The question was what are the risks for our company? I answered, there are financial risks, geopolitical risks and also the trade risk between United States and China. Also, some potential risk is in the investment for scaling, whether the investment will progress as planned. That's another potential risk. So as for your question, Mr. Wadaki, let me answer to your question, there are some pushouts trend, by and large. That is the migration to the scaling, a bit delayed compared with the plan. And I think that review of the plan has already considered. That's how I view.

[Tetsuo Tsuneishi, executive]

And I am Tsuneishi. And I quite agree with Mr. Kawai when it comes to financial estimates until March next year, the end of our fiscal year. So when you look at WFE for calendar year, but also when you look at the end of March 2019, of course. When you look at short-term trend, quarter-to-quarter, you can see some increase or decrease. However, there is no situation to force us to change the financial estimate for this fiscal year. Up to until next year, I think the market trend is rather good. Market is active in big picture. That's the reason why we didn't change our financial estimates.

[Tetsuya Wadaki, analyst]

One follow-up question. When you -- other companies often say that every 3 months, it's so difficult to change or revise the financial estimates. So we need to wait for 6 months to come up with a more precise financial estimate. Is that how you also think about the financial estimates? Or do you really think there is no need to change financial estimates now?

[Toshiki Kawai, executive]

For midterm, long-term period -- and there was SEMICON West, and I talked with many people at SEMICON West the other day, 2019 and 2020, positive messages were heard in SEMICON West as well. So when you look at big picture, there have been no changes at all. So very strong demand continue. That's what I want to report to you. As I said earlier, the plan is not because of the decrease in demand, but just because of the migration plan revision. So pushouts exist, at the same time, there are some pull-ins existing as well. In that sense, we have very strong demand, and there are some increase or decrease with pushouts and pull-ins. That trend are also incorporated.

[Tetsuo Tsuneishi, executive]

May I? I am Tsuneishi. Even in the past, as you know, we revised our financial estimate even in the first quarter. That happened in the past. In principle, essentially, as necessary, we revise or change the financial estimate if it's necessary. If there is no need, we don't change our financial estimate. That is our basic approach. So there is no such cycle of every 6 months.

[Kenichi Hirayama, executive]

Next question, please. Yes, the gentleman in front row, please.

[Yu Yoshida, analyst]

I am Yoshida from Deutsche Securities. About SPE new equipment sales forecast, the second quarter, 60% of the sales plan has been already shipped. But there is a big difference between first quarter and second quarter. Similarly, in the second half of your -- there should be big difference between the third and fourth quarter as well? That's my question.

[Toshiki Kawai, executive]

So your question is about whether there are any difference between first, third and fourth quarter? Is that your question? Is that correct? In particular, for this fiscal year, actually, there should be no major difference from the previous trends. So on the budget approval by the customer, it might have some impact. And because of that, there are some ups and downs. But in some cases, customer places order, and right away, they want to get delivery of the products. But for this fiscal year, there is no major difference from the previous years. That's how I view the trends. So maybe the January-March in the second half of this year, compared with October to December, there are some increase of the sales in January to March compared with the October to December, is that how I can view that. There might be a slight trim as such, but we don't change our plan at all. And when you look at closely the third quarter and fourth quarter, we haven't announced any breakdown between third quarter and fourth quarter. But this year is not different from the previous years.

[Yu Yoshida, analyst]

My follow-up question for FPD. Actually, first and second quarter, there are some big difference between first and second quarter as well. And you said, the second quarter sales plan has already been achieved. So very similarly, on the second half of this year, how do you view the trends of the sales for this FPD?

[Toshiki Kawai, executive]

So the FPD is -- rather requires very big production equipment. And we don't -- I am not able to make any comments on the details of the third and fourth quarter. But when you look at the overall market of FPD this year, next year and the year after, there should be very similar trend for 3 years to come in our company. As Mr. Sasagawa said earlier, we are trying to increase our profit margin. So 20% of operating profit margin is to be achieved by focusing on the high value-added products. Because of that, there might be some increase or decrease in the future. That's how I would like you to understand our future trend. Thank you very much for your question.

[Kenichi Hirayama, executive]

Any other questions? Yes, the gentleman next to the previous gentlemen, please.

[Satoru Ogawa, analyst]

I am Ogawa from Goldman Sachs, Japan. Again, I have a question about this fiscal year plan. For WFE, you said 10% to 15% growth expected, slight decreasing trend. On the other hand, you haven't changed the full year plan because you have -- when you -- there is conversion between the calendar year, the fiscal year, there is no change? Or even for your fiscal year, just like WFE, 5 point fluctuation factors incorporated. That's what I want to understand. So application, which might affect your financial estimate, logic is the important driver or memory also drive the changes.

[Toshiki Kawai, executive]

So demand is rather big, significant. And I mean, those are some circumstances. There are quite a few pull-ins and pull-out in all applications. That's the reason why we haven't changed. So for your second question, could you just repeat your second question, please? So you're talking about memory or logic?

[Satoru Ogawa, analyst]

Right.

[Toshiki Kawai, executive]

So your question is whether pull-in and pullouts are driven by memory devices or logic devices?

[Satoru Ogawa, analyst]

By application, actually.

[Toshiki Kawai, executive]

Right. For 3D NAND, DRAM or logic devices, the pull-in, pullouts happens in every application. Thank you.

[Satoru Ogawa, analyst]

So it's not because of demand, just because of scaling, driven by the leading-edge technology. So pushout, not for this fiscal year. Maybe next fiscal year we can see the realization? Or are -- they're something which just disappear? But it happens for pull-in or pullout. That's on the basis of 3 months or 6 months period?

[Toshiki Kawai, executive]

In particular, so when it comes to scaling, the maybe demanding device structure, maybe the yield should be one of the factors. As I said earlier, the high-volume production of existing or previous nodes, that's not the area that capacity is increasing. Rather, capacity is trying to be added for the leading-edge nodes to increase the big capacity, the wafer. So that's where the investment plan, in line with the scaling, causes some pull-ins or pullouts. Thank you very much.

[Kenichi Hirayama, executive]

Next question, please? Again, the gentleman in the front row, please.

[Mikio Hirakawa, analyst]

I am Hirakawa from Merrill Lynch Japan Securities. My question is very similar to the previous ones for pushout. I think I understand why pullout -- pushout takes place. But I want us -- I am not able to understand why pull-in takes place. Could you elaborate why pull-in takes place in the market?

[Toshiki Kawai, executive]

I cannot give you something specific. But, for example, when yield is improving rapidly, then maybe we can do some pull-in. That is one of the reasons why pull-in takes place. For IoT, Internet of Things, application as a whole is expanding. At the same time, application is diversified. Therefore, the device manufacturer [ trust us ] for expanding or diversifying applications. That's the reason why pull-in takes place. Not only the yield increase, but also the expansion of the IoT application is one of the reasons why pull-in takes place. Thank you very much.

[Kenichi Hirayama, executive]

Next question, please? So there is a gentleman raising a hand in the middle of this room.

[Yoshitsugu Yamamoto, analyst]

I am Yamamoto from Mizuho Securities. I have one question. I have a question to Mr. Sasagawa. So Slide 9, a very similar question was asked earlier, so I'm sorry for that. But your quarterly sales, when I built it, so first and third quarter, sales is rather low. And second and fourth quarter, sales are rather higher. When I look at American 2 vendors, their sales are very close to the final demands in trends. But why, in the case of your company, there is some seasonality in your sales? The second and fourth quarters sales, rather higher than the other 2 quarters. You are now following using the CST rather than shipment-based sales recognition that tells trend might be very close to American companies. But I think you define your own CST, so you can control the timing of sales recognition. Could you just elaborate the reason why this kind of trend happens in your company, Mr. Sasagawa, please?

[Ken Sasagawa, executive]

So I am Sasagawa. So from the accounting viewpoints, actually, I have no accurate idea about the other companies' situation. But generally speaking, we can look at the American competitors. They are recognizing sales very close to the shipment-based sales recognition. So shipment itself doesn't change so drastically from month to another or quarter-to-quarter. So I think the American competitors, sales are increasing linearly. That could be the style of accounting system of American companies. In the case of our company, on the other hand, as you said earlier, we are not based on shipment, but we are using CST, completion of start-up and testing. Of course, we are shipping our product on a linear basis, but we recognize ourselves for each line of the client altogether although those tools are shipped in different timing. Therefore, a certain week or in certain month, sales tend to be recognized altogether. That's the reason why you can see some increase or decrease in our sales from the viewpoint of accounting. So there is no intention at all, but sales recognition is increased in second and fourth quarter.

[Yoshitsugu Yamamoto, analyst]

Are there any particular reasons?

[Ken Sasagawa, executive]

No. There are no particular reasons for that. Thank you very much.

[Kenichi Hirayama, executive]

Yes? The gentleman just behind, 3 rows behind from the gentleman who asked the question earlier.

[Masahiro Nakanomyo, analyst]

I'm Nakanomyo from Jefferies Japan. My question is rather vague, I'm sorry for that. Earlier, so we are now observing some pushout because of the scaling. It's not because of the supply/demand relationship. That's what you said, and that's what's happening. However, when you look at the future trends, the scaling problem is albeit soft, and that might be some problem of supply and demands in the future, maybe next year. Don't you have such kind of a concern for the future? In other words, the supply/demand relationship will not improve so rapidly, and we don't have to worry about that particular issue. Could you just give us some quantitative answer to my question, please?

[Toshiki Kawai, executive]

So hyperscale data centers and based on IoT, big data era is now unfolding, and there is very strong demand coming out. So high-capacity device, high speed, high-reliability and low power consumption. Those devices will be required in the future. So [ robust ] demand increase, but technology innovation is necessary for the future. And that technology, leading-edge technology, I think there is only limited number of clients who are able to address leading-edge technologies. And those clients who can address leading-edge technologies are rather dominant in the market because they have very excellent technologies. In other words, those limited number of customers can have a very good command or view of the market. They understand the market. So technology innovation is required. There are only limited number of clients who can address the high level of technology innovation. Those limited number of customers are having very good command or view of the market. So therefore, customers do not produce the same thing at the same time. That happened in the past, but it doesn't happen right now because every customer -- leading customer can see the market trend very well. That is the background. And now we have the forecasts, so there should be no change expected for the future. There is no need for us to worry about the major change in the future. Thank you very much.

[Masahiro Nakanomyo, analyst]

Once again, actually, this -- today, Hynix made the financial announcement. And this year, next year, Hynix is going to increase the CapEx capital investment. That's what they say. So under those circumstances, still, you -- your current comment or your current view doesn't change. So limited number of clients can see the trend of demands, can see -- understand the market trend? Is that what you want to say?

[Toshiki Kawai, executive]

Yes. That's correct. Thank you very much for your question.

[Kenichi Hirayama, executive]

Again, the gentleman in the middle, please?

[Takeo Miyamoto, analyst]

I am Miyamoto from Mitsubishi UFJ Morgan Stanley Securities. On Page 15, Slide 15, I'd like you to give us some extra explanation. For outlook of WFE, so for this year, JPY 58 billion has been replaced by range from JPY 56 billion to JPY 58 billion. It looks like some decline. So that JPY 2 billion will disappear or the figure for 2019 will be increased from JPY 61 billion to JPY 63 billion? So I want to see where does it go, the $2 billion decrease. And I would like you to give us your comments on that issue.

[Toshiki Kawai, executive]

I understand your question. For this financial announcement, we just focus on this year's values -- figures. As I said earlier, the scaling, the migration to the scaling causes pushout. Because of that, when there is a strong pushout due to the migration for scaling, we just include $56 billion to $57 billion. But we haven't revised the figure for calendar year 2019. That's the reason why those figures just remain. In principal, pull-in and pushouts take place, but that $2 billion will not disappear at all. That is my answer to your question.

[Takeo Miyamoto, analyst]

So I have one follow-up question. When you talk about scaling, is that scaling for logic and DRAM? But when it comes to 3D NAND, there is no further scaling. So 3D NAND is not relevant to your story. Is that a correct understanding?

[Toshiki Kawai, executive]

No. For 3D NAND, in the structure of 3D NAND, the leading-edge 3D NAND, some revision takes place.

[Takeo Miyamoto, analyst]

But you said scaling, not this. Not only device shrink, but also the layer stacking is included. It's the practice. Is that correct?

[Toshiki Kawai, executive]

Yes. That's correct. Thank you very much for your question.

[Kenichi Hirayama, executive]

Yes, the gentleman in the front row, please?

[Damian Thong, analyst]

I'm Damian Thong from Macquarie Capital Securities. I have some additional questions for this fiscal year. WFE market outlook, back in April, for DRAM, you said 60% increase is expected. But this time, you said 60% to 65% increase. NAND flash, the same as before. That's what you said in April. But this time, you said 5% increase. But when it comes to this increase for the memory, are there any reasons for that? As a -- I think 3D NAND flash, the scaling trend is slowing down. So why do you think memory market is growing? Could you give me some reasons for that?

[Toshiki Kawai, executive]

Actually, the 3D NAND investment is actually stronger than our expectation. That's the reason why.

[Kenichi Hirayama, executive]

Next question, please? Yes, the gentleman next to the previous gentlemen, please.

[Toru Sugiura, analyst]

I am Sugiura from Daiwa Securities. So you said there is no change in the big picture. And I think I understand it very well, but you said 3 months or 6 months pushout taking place. And if it is just because of the yield program at the customer sites, do you think that yield program will be corrected or solved in 3 months or 6 months? Do you think this is rather accurate, the pushout will be completed within 3 months or 6 months? Or depending on the situation, there might be some possibility the push-out period will be extended farther more? Could you give me your comments, please?

[Toshiki Kawai, executive]

As I said earlier, 1 quarter or 2 quarters, 3 to 6 months should be the push-out period. That's the image I have. Whether it will be extended farther more or not, actually it's up to customers, the customers' progress. So I'm not in the position to make some comments on that issue.

[Toru Sugiura, analyst]

In that sense, so now you are communicating with customers and you think maybe yield problem can be solved within the 3 or 6 months? Is that how you view the situation?

[Toshiki Kawai, executive]

Yes. That's correct.

[Toru Sugiura, analyst]

I have a follow-up question on Page 21, Slide 21. Now we can see the sales by application compared with previous announcement. When I look at the composition, the nonvolatile proportion increases in this composition. So I wonder what is the background. For example, which generation of NAND increases in composition or proportion? Could you give us the reason why the proportion of NAND device is increasing?

[Toshiki Kawai, executive]

So this is the composition of our sales. So this is not the composition of the market itself. I don't have any appropriate figures, but DRAM composition in the market is in -- proportion in the market is increasing. But as far as our sales is concerned, since last year, we have been focusing on the etching and cleaning process, which are to be introducing 3D NAND processes. So we have added that bit. That's the reason why the composition of 3D NAND increases. So 3D NAND and DRAM, I think they are rather [ complement ] each other, almost the same level, but we have been working hard for the 3D NAND. Now we can see successful results, and that 3D NAND success contribute to our sales. That is the meaning of this purple color. So this is -- this reflect our efforts, and we are very happy to see that.

[Kenichi Hirayama, executive]

The gentleman on the second row, please.

[Takeru Hanaya, analyst]

I am Hanaya from SMBC Nikko Securities. My question is very similar to the previous question. As for the new equipment sales forecast shown on the Page 21, so when I just roughly calculated it, from the 3 months ago, actually, sales for DRAM has been decreasing according to my calculation. However, for WFE, growth rate for DRAM was increased from 60% to 65%. So -- but the sales for DRAM is not increased. Why does that happen? Could you explain the reason why, please?

[Toshiki Kawai, executive]

Just because of the changes in mix. There is some slight change in DRAM plan, and there are some factors to replenish it. But there have been some changes.

[Takeru Hanaya, analyst]

As for nonvolatile memory. Again, WFE, almost the same level to plus 5%, 5% increase. But when you have the breakdown NAND and other nonvolatile memory, for example, 3D NAND might be decreasing, while the other is increasing. Could you give us some breakdown between NAND and others in nonvolatile memory? Could you give us some comments, please?

[Toshiki Kawai, executive]

The composition of nonvolatile memory sales haven't changed. Thank you very much.

[Kenichi Hirayama, executive]

Next question, please? We do have 10 more minutes. So maybe if you can come up with second question. I'm sorry. We have a gentleman in the rear of this room. Could you ask your question, please?

[Masahiko Ishino, analyst]

I am Ishino from Tokai Tokyo Research Center. There are some technological hurdles, and there are some pushouts. That's what I heard from manufacturers. But for DRAM, there was the problem in 18-nanometer node. Also, in the case of NAND, 96 layers technology doesn't go so well. In the case of logic device, 7-nano node might be skipped, strictly go to 5-nanometer node. That's what is happening for each application. From your viewpoint, what are the technological hurdles to cause pushouts? And I think that several problems are being solved. But do you make some proposal to solve those problems causing pushouts? Could you give us some comments on that technology issue, please?

[Toshiki Kawai, executive]

It's a bit difficult for me to give you the specific answer to that. I am not in a position to give you the technology issues. And for the details, I'm not able to get the information. That's the reality. However, it is just a matter of process tuning. There is no need to change the materials or change the device structure. The problem is not that significant. That's what I heard.

[Masahiko Ishino, analyst]

If that is the case, in 3 months or in 6 months, pushout will be completed, and you can solve the problems by carrying out some process tuning. That's the current status of the problem according to your understanding. Is that correct?

[Toshiki Kawai, executive]

Yes. That's how I understand the situation.

[Masahiko Ishino, analyst]

I have one follow-up question. As for the flat panel display, G10.5, there are some pushouts. That's what you said earlier. So what is, again, the technology issue make the pushout -- is making the pushouts? Or I just heard, because of several issues, construction of factories is delayed. That is the reason why there are some pushouts. Is that correct understanding?

[Toshiki Kawai, executive]

Well, some of the customers are now revisiting their CapEx.

[Kenichi Hirayama, executive]

Any other questions from the floor? Maybe second question will do. Yes, coming back to the other gentleman, the gentleman in the front row, please.

[Satoru Ogawa, analyst]

I am Ogawa from Goldman Sachs Japan. 2019, 2020, you said there is no change in big picture. But in the past, 2019 NAND investment will be increasing. That's what you said earlier. And Mr. Kawai, when you discuss with clients, what sort of message you got from the customer? Next year in-all, if the NAND price and NAND profitability, what sort of comments do your customers have? At present, NAND price is going down. But still, customers are willing to invest in the NAND.

[Toshiki Kawai, executive]

Actually, there is a big trend -- demands in principle. But for NAND, the scaling investment -- investment for scaling is a major driver for NAND flash. The investment to the 9X layer is the dominant -- is dominant. I'm sorry, this is not scaling, the layer stacking, I should say.

[Satoru Ogawa, analyst]

For beyond 2019, for the purpose of the layer stacking, not for the greenfield, the investment will be continuing for the purpose of this layer stacking. Is that correct understanding for 3D NAND clients?

[Toshiki Kawai, executive]

They are mainly investing to introduce 9X generation. This fiscal year, some clients already started the investment for 9X layer generation. But there are some customers to shift the existing production line to produce 9X layer generation or some other customers want to invest in new production lines.

[Satoru Ogawa, analyst]

If that's the case, for NAND manufacturer, they are concerned about the profitability when they make an investment decision. Is that correct?

[Toshiki Kawai, executive]

Yes. Every customer is always concerned about profitability. That's the reason why we are observing continuous, stable growth in the market. The demand itself is rather big.

[Tetsuo Tsuneishi, executive]

And I am Tsuneishi. Let me answer your question. So when you look at the market view, I don't know whether my view is correct or not. But now, we have the data center, and big data drives NAND, DRAM and logic because they are servers. There is a drastic growth in that area. When you look at customer CapEx timing, in some cases, they are making investment decisions in order to improve yield or to introduce new technologies. But in reality, for each quarter, customers are looking at the EPS and operating profit margin. And many clients, which are very much concerned about EPS and operating profit margin, are [ now ] investing. So customers do not like to see any decline in EPS or profitability. And that's how they make a decision when to invest. Maybe that should be the appropriate way to look at the market at present. So demand itself, for example, data centers, probably, I personally think, are limited. Big demand exist in the market. And in the future, when 5G is introduced, we will see more and more demand and more data centers are to be established. So beyond 2020, we will see much data, and data traffic will be increasing rapidly in every part of the world. As I said earlier, in the previous meeting, when it comes to CapEx or investment for the customer, so when they invest some money, they want to see the solid returns, and they are making decisions to invest when they are sure they can get a return from the investment. For Chinese emerging customers, you may think differently. However, the major DRAM vendors, NAND flash vendors and logic device vendors, all those major players are essentially looking at profitability, EPS as well as the operating profit margin. Those 3 parameters are the core when they make a plan for investment. And there is no concern about declining big demands for a few years to come. There is not a major concern. However, when the price dropped so suddenly or drastically in the future and if customers' EPS will be declining, then they may decide not to invest so much. However, when it comes to big demands, there should be no concern. There is no need to concern about as while as big data era takes place. That's how I view the market right now. Thank you very much.

[Toshiki Kawai, executive]

I am Kawai, and I would like to add some more comments. Recently, when I talk with our customers, actually customers are always looking at their business strategy when we discuss with customers. Actually, I've been working with -- in Tokyo Electron for over the past 30 years. In the past, customers are just focusing on increasing their market share. But right now, currently, customers always think about their business strategy or business plan when we -- when they talk with us. So as Mr. Tsuneishi said earlier, I quite agree with him. Thank you very much.

[Kenichi Hirayama, executive]

Thank you very much for your questions. It's time for us to close this financial announcement. Once again, thank you very much for joining us despite your very busy schedule.