TOKYO ELECTRON LIMITED

IR

Q1 FY2020 Earnings Release Conference Q&A

Although delays are being seen in the investments of customers that had been expected in the period from July-September 2019, the TEL financial estimates have been left in place. Do customers have some sort of new investment plans? In addition, the Chinese memory customers are moving their investments earlier, but is this included in the financial estimates?

In logic, investment for 5G will finally begin, and we expect investment to continue to be strong. As for management of memory customers, they are saying that new investment plans are likely to be clear sometime in the next three months. Note that we expect those Chinese memory investments in the next fiscal year, not this one.

Does your expression “memory investment plans are likely to become clearer in the next three months” indicate that estimates of investments by customers will be lowered or that they will recover? What is the outlook for memory investment in CY2020?

In the next three months, memory inventories will fall even further, and we think that the new investment plans of our customers will become clearer along with this. We believe that memory investment is now nearing its bottom and expect that both orders and shipments will increase for the second half of CY2020.

What do you consider the causes to be for the improvement in the memory supply and demand balance? For example, do you think that the power outage at the Toshiba Memory plant or the restrictions on the export of hydrogen fluoride to South Korea will cause a decrease in memory supply?

Demand for data centers and smartphones and the spread of new technologies such as 5G and AI will promote adjustment of memory inventories.

What made you think that investment is now at a bottom? Which application will recover first?

Three months ago, we saw a 50:50 composition ratio of the WFE*1 market for memory and logic, but now the demand for logic is picking up and we believe that it is 40:60 at this point. Although three months ago we saw NAND as the first to recover in memory, we now believe that NAND and DRAM will recover at about the same time.

What is the investment outlook for CY2020 in logic and logic foundries, respectively? Investment is at a high level in CY2019, but will it increase even further after this?

Although we expect investment in logic and logic foundries in CY2020 to increase in the second half, it is too early to forecast the full-year investment amount.

Regarding the WFE market, can we expect that the market size in CY2020 will be as large as in CY2018? I would like to ask about the magnitude of the investment recovery in CY2020.

Although we believe that investment will recover in CY2020, it is too early to say anything about to what extent.

Regarding the FY2020 sales estimates by application shown on page 20 of the presentation materials, the total of the two categories Logic Foundry and Logic & Others is expected to grow by about 50-60% year-on-year. What lies behind this growth being higher than the CY2019 WFE market outlook (+35% year-on-year) for logic and logic foundries?

This is because as the technology node proceeds, the proportion of critical processes also increases, and the areas in which TEL participates will grow even faster than in the WFE Market.

Compared to three months ago, sales estimates by application were revised downward by about 30 billion yen in NAND and 20 billion yen in DRAM, and revised upward by about 50 billion yen for Logic Foundry and Logic & Others. Will gross profit margin decline due to a decrease in sales for NAND that requires more etch systems, which are more profitable than other systems?

TEL’s market share in etch systems is higher for logic than for NAND. However, if we look at TEL as a whole, there is not such a large difference in market share by application, and there will be no major impact on profitability even if the composition of sales by application changes.

Regarding Field Solutions (FS), it was explained that parts sales were favorable. Why are parts sales favorable amid the decline in factory utilization rates of memory customers? Also, what is the outlook for FS sales in Q2?

Sales in FS for Q1 and Q3 tend to be lower than in Q2 and Q4, but in this Q1 sales declined slightly compared to the previous quarter, FY2019 Q4, and also compared to two quarters ago, FY2019 Q3. Sales are declining mainly in modifications, and parts sales have not fallen that far. In Q2, we estimate that progress will be as planned.
TEL ships about 4,000 units of equipment per year, so FS demand is certain to increase.

Why is it that FS business results do not fall off much even though the new equipment business is in a slowdown? In which kinds of fields are customer needs increasing?

As you know, sales of parts improve as the installed base increases. We will continue to make the most of product reliability and support capabilities, which are TEL’s strengths. As technology becomes critical, improving equipment uptime and yields through tool-to-tool and chamber-to-chamber matching become important for customers. TEL is aiming to provide new services for the future such as remote services using AI and advanced field support.

What are the reasons for customer advances increasing greatly quarter-on-quarter in the balance sheets?

TEL recognizes sales at the time of completion of set-up and testing, and records any payments received from customers prior to the recognition of sales as customer advances. In this Q1, there just happened to be many cases in which payment was received during shipment or installation.

How does TEL understand the high level of gross profit margin in Q1? In addition, do you agree that it would have been better if you could have made further cuts in selling, general and administrative expenses?

One of the reasons for being able to maintain a high level of gross profit margin while sales fell sharply is that inventories that are to be sold during Q2 increased, and fixed costs were converted into these inventories. It was not because profits were generated by squeezing expenses.
Regarding selling, general and administrative expenses, among these are also development costs and related capital expenditures that TEL considers to be important over the medium term, and we will not be letting go of these investments this year even though investments of memory customers are slowing down. Even if we were to reduce other costs, we intend to maintain growth investments for the future.

If we assume that revenue will increase in the next fiscal year compared to this fiscal year, will you be able to improve gross profit margin compared to this fiscal year? For example, I assume that initial development costs have been occurring in this fiscal year due to factors such as the acquisition of POR*2 in ALD, but won’t profitability improve as these costs are no longer there in the next fiscal year?

An improvement in gross profit margin is also stated in the Medium-term Management Plan. We will proceed to make improvements by (i) supplying high value-added products, (ii) the increase in sales that accompanies market recovery, and in addition, (iii) the growth of FS sales (target of 380 billion yen after 5 years) which are more profitable than the company average, and (iv) achieving an operating margin of 30% in the FPD business.

Regarding the acquisition of POR for etch and deposition mentioned on page 17 of the presentation materials, what is the outlook for the volume and timing of sales?

We would like to refrain from commenting on the volume and timing of sales because this depends on the investment plans of customers. Still, we were able to acquire the high value-added ALD processes for DRAM and NAND. We are anticipating next year’s memory investment.

What measures are you taking concerning the merger of Applied Materials and KOKUSAI ELECTRIC?

Technology performance of equipment is important, and our strategy does not change due to mergers of other companies. TEL already owns all of the deposition technologies for batch, semi-batch, and single wafer production, and we will ensure that we improve the performance of processes and equipment for the applications that can be expected in the future.

What are your ideas concerning M&A?

We do not rule out M&A if there are advantages for shareholders, if it will lead to raising corporate value, if there are also technological advantages for customers, and a return on investment can be expected. We would like to consider non-M&A measures such as joint development as well.

WFE (Wafer fab equipment): The semiconductor production process is divided into front-end production, in which circuits are formed on wafers and inspected, and back-end production, in which wafers are cut into chips, assembled and inspected again. Wafer fab equipment refers to the production equipment used in front-end production and in wafer-level packaging production

POR (Process of record): Certification of the adoption of equipment in customers' semiconductor production processes

The above content is a summary of question and answers session.