Earnings Call: Q3 2019

Jan 30, 2019

Thank you very much for your participation today, despite your busy schedule. It is now scheduled time So let us begin the fiscal year 2018 third quarter information meeting of Adventist Corporation. First, let me introduce our attendees starting from next to the center screen, President and CEO Yoshida. Corporate Vice President Mion, Managing Executive Officer Fujita, and IIM Nagumo, I will be your facilitator today. To start off, our Management Executive Officer, Fujita, will explain the financial results for fiscal year 2018 third quarter. Then President and CEO Yoshida will explain our fiscal year 2018 business outlook. After that, we will entertain your questions. This meeting is scheduled to be finished at 5. Now Mr. Fujita, please. Good afternoon, everyone. I am Fujita. I am going to outline the financial results for FY2018 third quarter. This slide shows summary of our results percent quarter on quarter and increased 10.5 percent year on year to JPY 62,700,000,000. Sales rose 3.1% quarter on quarter and grew 47.3% year on year. To JPY 74,900,000,000. Operating income was up 15.4 percent quarter on quarter and increased 6.4 fold year on year to JPY 20,700,000,000. Net income grew 10.1 percent quarter on quarter and increased just sevenfold year on year to JPY 17,900,000,000. At the previous results briefing in October 2018, we had a cautious forward outlook in view of the possibility that U. S.-China trade friction would affect our business. And currently, uncertainty surrounding the trade friction between U. S. And China is considerably dampening enthusiasm for Tesla investment. And the entire same conductor industry is undertaking inventory adjustments. With that trend, Orders for memory testers declined considerably from the previous quarter and overall orders decreased year on year. On the other hand, with respect to SoC testers, chip makers actively invested in reinforcement of Versum Conductor capability in Q3 against a backdrop of advancing smartphone performance. Due to this continued high demand for SOC testers, both sales and profits exceeded our assumptions 3 months ago. This slide is about FY2018 3rd quarter orders by segment. Instant conductor and component test systems, orders declined 16.2% quarter on quarter to JPY 48,800,000,000. SoC testers continued to see a high level of orders particularly for application processes and display driver ICs, which are becoming more sophisticated. Leading to orders of JPY 43,100,000,000. Memory test orders declined sharply as major customers proceeded with memory chip inventory adjustment and stood at only JPY 5,700,000,000. Macatronics systems decreased 45.4 percent quarter on quarter to JPY 6,800,000,000. Device interface orders declined along with memory testers. Services, support and others rose 30.7 percent quarter on quarter to JPY 7,100,000,000 as annual maintenance contract orders recovered. This slide shows FY2018 Q3 sales by segment. Semiconductor And Component Test Systems grew 4.8 percent quarter on quarter to JPY 56,800,000,000. Thanks to growth in smartphone related areas, associate testers could achieve another quarterly sales record of JPY 39,500,000,000. Memory tester sales plateaued at JPY 17,300,000,000 as we worked through our order backlog. Echatronics Systems were flat quarteronquarter as JPY 10,600,000,000. Services, support, and others were also flat quarter on quarter to post JPY 7,500,000,000. As for orders and sales by region, let me go through orders trend in major regions. In Korea, memory testers have been sluggish since the second quarter, but orders for SoC testers such as display driver IC and image sensors increased. In Taiwan and China, memory testers and testers for display driver IC, which was strong in the 2nd quarter, suffered reactive downturn. However, orders for testers for display driver IC will come back in the 4th quarter. As for sales by region, In Asian countries such as Korea, Taiwan and China, where semiconductor volume production processes are concentrated, high level sales continued. In Taiwan, where many customers which perform high end SoC tests are located, sales accelerated further. Overseas sales ratio was 95.6 percent. As for operating income in the third quarter, Gross margin was 56.8 percent. Improved product mix contributed to achieve high profitability. SG and A was JPY 24,400,000,000, up JPY 2,500,000,000 quarter on quarter. It is mainly due to the increasing accrued performance based bonus that is backed by the growth in sales and profit in the third quarter. Operating income was JPY 20,700,000,000, which includes the settlement gain of JPY 2,500,000,000 from the transition our part of Adventist Corporation pension plan for employees in Japan to a defined contribution plan. Operating margin was 27.6 percent, up 2.9 points quarter on quarter. R and D in the third quarter was 1,000,000,000, and this is due to the increased human resources investment into R and D sections. R and D to sales ratio was 13.7 percent. CapEx was JPY 1,200,000,000, and the depreciation and amortization was also JPY 1 point 1,000,000,000. As for cash flow, in the last 2 years, Sears had been expanding without the major production capacity expansion. In the third quarter, we had a net cash inflow of JPY 15,200,000,000. As for the balance sheet as of the end of December 2018, total asset was 1,000,000,000. Cash and cash equivalents were JPY 124,800,000,000. Bonds were JPY 5,900,000,000, and it was almost unchanged from the end of September. As of the end of December, conversion rate of convertible bond exercisable by February 2019 was approximately 80%. Equity attributable to owners of the parent was JPY 188,100,000,000. The ratio of equity went up 1.5. Quarter on quarter to 63.4 percent. With this, I conclude the presentation on financial results for the third quarter FY2018. Thank you for your attention. Next, President Yoshida, please. Good afternoon, everyone. I am Yoshida, I'd like to explain FY2018 business outlook. This slide describes FY2018 forecast. With SoC testers compensating for the deceleration in memory testers, Q3 orders and sales progress considerably better than we expected 3 months ago. Based on this, we will again raise our annual forecast for Sinconductor And Component Test Systems segment as well as our company wide FY2018 financial forecast. We anticipate Q4 orders of about 1,000,000,000 and Q4 sales of about 1,000,000,000. Our full year forecast for orders, sales and net profit will be the highest ever with JPY 278,000,000,000 for sales and JPY 54,500,000,000 for net income. Setting a new record for the our dividend policy calls for a half year consolidated payout ratio of 30%. Our forecast 3 months ago anticipated year end dividend of which would make a total dividend of for the full year. We now expect our year end dividend of JPY 38, making a total dividend of JPY 88 for the full year. This will be an increase of JPY 56 year on year. We anticipate earnings per share for the full year to be slightly less than 190 yen and ROE to be over 30%. As such calendar year 2018 was a great year, but I think you're wondering how about 2019 So let me touch upon calendar year 2019 market forecast. In 2018, the performance was improved in a variety of SIM conductors used in various applications such as data centers, smartphones, industrial equipment, automobiles and displays. We have closely monitored trade friction between the US and China for potential negative impact on semiconductor demand. However, despite a sense that it was running out of steam in the second half, the semiconductor market was basically steady in 2018. As a result, various semiconductor manufacturers reinforced their test capabilities in 2018. Driving very good results for us. However, as US China trade friction intensifies uncertainty in the global economy is increasing. This has affected customer sentiment and customers corporate activities, including inventory control and capital investment. Many product supply chains and customers are currently undergoing inventory adjustments. Demand for new testers is expected to decrease until this inventory adjustment cycle closes. It is very difficult to make predictions at this time, but we think that demand will bottom out in the first half of the calendar year and will recover in the second half. Tester investment should pick up again from 2020 onward. The SOC tester market increased in 2018 to approximately $2,400,000,000 up 9% year on year, but it is estimated to be approximately $2,000,000,000 in 2019. Down about 15% year on year. The memory tester market grew 33% year on year, to approximately $1,000,000,000 in 2018. But now we think that it will shrink to approximately $700,000,000 to $800,000,000 in 2019, down about 20% to 30% year on year. These are our key measures. Demand for testers is expected to slow down in the short term. But on the other hand, we and our customers are already preparing for the next growth phase. While inventory adjustments and capital investment restraint made temporarily depressed demand, In the larger wave of the digital transformation, customers are working hard to develop the next generation of devices in order to achieve higher performance symptom doctors than ever before. To enable continuous innovation, we collaborate with customers from the device development stage by developing and providing new test technologies and solutions. In fact, our collaboration with leading customers in the semiconductor market on technologies including AI, 5g ADAS, and high end memory has been successful This led to our growth in FY2018. From the second half of twenty nineteen onward, volume production of new devices is expected to start in a more sophisticated and reliable way. This will be affected by the length of the inventory adjustment cycle, but we are confident that the Tesla market will be driven by the adoption of these new devices and will continue to grow. This is in line with our mid to long term management plan announced in April 2018. It is a view that the tester market will grow over the medium to long term despite repeated fluctuations, therefore, and daunted by the current sense of deceleration we'll continue to take necessary measures for future growth. As one early step for this, we announced the acquisition of the system level test business of Astronics And American company in November 2018. Unfortunately, we have not closed a deal yet, but the 2 companies are proceeding towards closing it. We'll make an announcement as soon as these procedures are completed. FY18 outlook by segment, First of all, SoC tester's outlook for FY18. In FY18, amid improved performance increased complexity. And the miniaturization of semiconductor for smartphone processors and display panels. Many customers are expanding test capacity. Robust test demand for display driver IC will continue in 2019 following the prior year with the sustained momentum. The growth of our SOC tester business was backed by the advancement of Semiconductors, but we were benefited by the test expansion of many customers because of extensive customer base enhanced product portfolio and the strengthened global support network. Partly affected by slow growth of smartphones and other finer products. 4th quarter were faced some slowdowns. However, series of SLC tester in this fiscal year will clearly exceed the initial expectations. Supported by share increase among others. Talking about the market share, usually it is based on the calendar year rather than the fiscal year. And in current year 2018, we'll be increasing strongly from slightly above 30% in 2017. To approximately 50% in 2018. Memory testers in memory market since last autumn customers have been increasingly adjusting inventories. Many factors have been impacting Ripper effects of U. S. China trade frictions slowdown in data center investment and shortage of CPU and others. Due to memory inventory adjustment, Customer's appetite for investment has been weakening, and we lowered the Tesla sales outlook from that of October. That said, investment adjustment in memory tests will not be as substantial and serious as before compared with the IT bubble burst or Lima crops. Because memory capacity expansion will drive the investment to expand testing capacity. Memory sales will go down, but that will be more than offset by SoC. Therefore, as the semiconductor and component test system as a whole, we make substantial upward revision. Megatreonic systems, services, support and others outlook. Some megatronic systems are closely connected to memory testers. Though device interface business was robust in the first half, is slowed down in the second half. Therefore, we revised down by JPY 2,000,000,000. Services support and others have been in line with the projection, and we didn't make any revisions from October. As for summary, Although it turned more uncertain in the second half of the year, calendar year 2018 was an excellent year. Which successfully captured the growth opportunity. And in the fiscal year 2018, we will mark the record high orders net sales and net profit. They were boosted by structural changes in tester market in the last few years, Semiconductor Application has been expanding and the social needs for ensuring reliability of Semiconductors has been mounting and the customers have been continuously strengthening the initiatives to that end. And as such circumstances, strengthening global operation for global customers delivered results FI18 results are supported by consistent development investment, wide product portfolio, global network of sales development and after sales support and global management system. And in this 1 year, in particular, the production network that helped to respond flexibly to the drastic demand increase supported the FY18 result. Current year 2019, Tesla market will slow down for some time, and that is already counted in our midterm business plan. Structure changes of tester market will continue in the calendar year 2019 and onward. We continue to prepare for sustainable growth without fear for the current weakening sentiment. To be more specific for the upcoming 5G and AI, we will enhance the development structures. We will spend more for human resources and address the shortage of manpower. Further down the road, we explore new businesses leveraging our technologies. For those objectives, we take necessary initiatives. That concludes my presentation. Thank you very much for your attention. Now we will entertain your questions from the floor. Please raise your hand if you have any questions, and please state your name and affiliation before you ask questions. Yes, please. I am Wadaki from Nomura. Thank you for presenting your great financial results. I have a few questions. My first question is about your assumptions used for 2019 market forecast. What do you think about the impact of U. S.-China trade conflict on the Tesla demand, especially for SOC testers which have been strong so far. Also, regarding memory testers in which Korea plays an important role, When do you think the memory tester market will recover? We still cannot have a clear view about the level of impact of the U. S.-China relationship on the SoC business. But based growing demand for high performance smartphones with 7 nanometer node application processors or APs. In China, new middle range smartphones are adopting advanced functionalities. And this makes test time longer. We expect demand for new smartphones related testers will continue in 2019, although it will not be as strong as 2018. As for memory tests, we estimate major manufacturers would finish their inventory adjustment by June, and we assume capital investment will resume in June September timeframe as a result of various adjustment measures such as managing inventory with shorter times than the current 3 months. Regarding the assumption of Test and Market forecast I mentioned earlier, we expect that trade conflict between U. S. And China will not be resolved easily. However, it does not consider cases the U. S. Administration will apply tougher measures on Chinese Semiconductor Companies such as suspension of business transactions. If this happens in macroeconomy deteriorates, we assume market situation could be more miserable than earlier mentioned forecast. Let me ask my second question. You revised SoC tester sales for this fiscal year upward by JPY 22,000,000,000. Can you tell me the breakdown by product? This upward revision comes from V 93,000 and T2000 series, including display drivers and CMOS image sensors. This is my last question. I heard from a friend of mine that present you shooter is very much interested in micro LED. So LED driver IC testers could be promising. Do you have any comments on prospects of LED driver as it test us? The rate of TDDI is still low. We anticipate test time would be doubled or tripled by the market growth for 4K TVs and growth of COF, which currently accounts for 30% of the market and still smaller than COG. Based on this assumption, with full deployment of 5G system in the market, By shifting from 4K to 8K TV in 2019, we expect test demand for this play driver ICs will continue for another 2 years. How about LED driver IC testers? Do you mean testing micro LED itself? I heard that a new driver IC will be required for micro LEDs. We think that driver circuit will not be so different from existing ones. However, the technologies to check colors on LED display by panels with them. Manufact are thinking hard, how it can be checked visually or automatically. So that is still under study about how we should build algorithm on testers for checking colors. We'll focus on this matter going forward. Thank you. I am Yamamoto from Mizuho. My question is about the background of market share increase of SoC testers. I can think of some reasons such as the growth in the market sector in which you are already strong, for instance, driver IC testers or you could grab the market share from your competitors such as Theradyne. Can you tell me the background of the market share increase in 2018. And based on that, I appreciate if you comment on the direction in 2019. There was a significant change of market share in 2018. We announced in the midterm management plan that our strategy was to increase the share steadily, still we did not expect such a drastic change in a short term. The largest reason for it was that major customers of our competitors did not execute much investments. As a result, their market shares dropped sharply. However, as I said earlier, associate market was larger in 2018 than that of 2017, and the entire growth was taken by us. In house their test capabilities. I hear a competitor made a comment in their information meeting that special demand was created for us by a change of supply chain of a certain semiconductor company. But there was no such an event. Demand for testers increased in CMOS image sensors, analogs, and vehicles in which we had been relatively weak before, a wide product portfolio could address the demand. And this increased our market share. Having said that, impact caused by an individual company is getting larger. Demand fluctuation of major mobile phone companies would affect market share significantly. So we understand if demand from one of our major customers suddenly become 0, it could change our market share. Still, our strength lies in the wide customer base and sales is distributed among them. We are determined to maintain this market share by orchestrating global efforts by our sales, support, and development teams. Do you see or expect any changes in market share for CPU, GPU and FPGA for PC and servers. CPU and GPU are slightly different from the situation of APs. For system operations, AP needs the largest number of equipment and CPU and GPU needs slightly a smaller number of equipment than APs. Currently, most major GPU manufacturers are using our products. And we do not expect a significant change Thank you for your explanation. I'm Yoshida from Deutsche. My first question is about your market outlook for calendar year 2019, what do you estimate for your market share growth? Will it be higher than the market growth or equal to the market growth. We intend to maintain 2018 market share of SoC testers. As for memory testers, although I did not mention this earlier, we already have market share of over 60% We expect newly generated demand will be more for high end testers and higher the demand for high end testers more opportunities for us to increase market share. So we intend to maintain the current market share for SOC testers and to gain more share sales decrease will be smaller than the shrinking of the market. Still in next fiscal year, if sales decline by 15% or 20%, considering your high marginal profit ratio, which would affect operating income to be decreased by JPY 20,000,000,000 or JPY 30,000,000,000. I understand you will continue investment but do you have any measures to reduce expenses to compensate for the decline of marginal profit I guess you mean expenses rather than marginal profit. In our system, labor cost decreases in conjunction with sales decline. Having said that, if you are whether we will contain cost in order to increase profits. Honestly speaking, we don't think it is not an appropriate time to do so. In FY 2019, even though sales are expected to decline slightly, We think it is the phase 4 investment. Of course, this does not mean we will do wasteful spending. And if production decline associated expenses will get lower by themselves. So we assure our system will reduce expenses appropriately for sales decline. Based on this assumption, how can we restore the decline sales? For instance, there might be some contributions from earlier mentioned M and A, and that is not necessarily the only M and A deal for us. We'll continue to strive for sales increase and business growth by various measures. This is my last question. As described in the presentation slide, if 2019 is a one step in the cyclical growth cycle, what can be expected in the market of 2020 For memory testers, it can be expected to recover demand as inventory adjustment will be finished and 5G investment will be implemented. But for SOC testers, demand for application processes with AI function or TDDI may die down. What do you expect for new demand in this area? As for SoC, full scale implementation of 5G will be the market driver in 2020. With this trend, memory density will be 2 or 3 fold, and the data transfer speed will be faster. Based on this perspective base station business is expected to start to grow in 2020. An SOC tester market will enter into growth stage. Especially 5G's base station related business is expected to be quite large. And when full scale investment starts for 5G system, demand for testers in this area will rise accordingly. Would you comment on the scale of the market of base stations? For 5G, there will be progress from 7 nano into 5 nano in 2020 and onwards. And EU's needs to be addressed and testing time will be longer with the improvement of performance. In 2018, our SoC series marked the peak, and I think that to be achieved again. I'm Hiroa Kala from Merrill Lynch Japan Securities. I have two questions. First, as a follow-up for the previous question, orders focused were revised up from 255,000,000,000 to 265,000,000,000 the full year, would you give us a breakdown? Does it simply reflect to upside in the third quarter and does your forecast for the 4th quarter remain unchanged. In October, our forecast for the side and 4th quarter were almost the same. In the third quarter, actually orders grew. And as mentioned before, it was mainly driven by SoC. Every year, both of orders in series tend to be stronger in the 4th quarter, but that wouldn't be likely in this 4th quarter. Is that correct? You may have the impression over years that orders and sales are strong in the fourth quarter, but this year, the capital investment in memory was substantially controlled, and it is not likely that we received big orders in the 4th quarter. Especially around this time last year, exactly 1 year ago, situation was completely different. It was hard to deliver in time and lead time was stretching to 6 months from the usual 3 months. Anticipating the delays in delivery customers who are placing orders earlier and that led to the substantial orders in the fourth quarter of last year. But this year, no one is making orders display drivers, we are still not able to deliver in time. Having that said, with the increased uncertainty of macroeconomy customers are not willing to place orders earlier. That is the reason why we cannot have aggressive outlook for the 4th quarter. I would like to have more information about 5g, would you comment on the timing when the demand for 5g Handset and 5g stations where you make full contribution. And also, would you comment how large was the impact by the demand for the 4 g base station in the past? For 5G base station, chip tester is mainly V93K, and for 4 G base station, V93 K was used substantially around 2015 and 'sixteen. And it was a peak time. But for 5G smartphone and 5G networks, it has just started and technology is not yet well established, and it is still in nascent state. And fruit commercialization will be in 2020 according to Makers. And the peak time will be in late 2020. I'm Miyamoto from MUFG Morgan Stanley Securities. I have two questions. First, as for the upside of SoC in the South Quarter, would you comment on the strong regions? For application processors, foundry's demand for making AP with 7 nano was strong. And for the display driver, Taiwan was Zhong with OSAT and Korea also. And in China, also, the OSAT has started to use the display driver testing These three countries were the major contributors. You presented the outlook for the 2019 market size. Based on that, can you sustain the gross margin of 55 percent of 2018 with the contraction of the market, or will it fall to some extent? As you may know already, currently, the profitability of SoC is better than the memory per unit. And proportionary memory will be shrinking. So gross margin will be sustained or may go up slightly compared with this year. I'm Segura from Daiwa Securities. I have one confirmation on 2019 market. SOC Test And Market is shown as -15 percent. Would you give us the balance how it is spaced between the 1st and the second half? If market recovers in the second half, then the balance will be 4060 or 3070 how strong the second half will be? And what would be the key driver for recovery? Is it HPC or application Processors? Well, that's a tough question. It's up to the launch time of new smartphones. And the timing of when the year will be high enough and a stable around 90% for 7 nano needs to be monitored. In 2018 third quarter, substantial capital investment was made. And given the utilization of debt capacity, As for SoC, the balance will be 4060 between the first and the second half of the year. In 2018, TDDI and AI chip contributed to extend that testing time. In 2019, what's your prospect for extending testing time? For example, GPU circuit area will be wider in AP. And will that be beneficial? What matters us most for series of testers is manufacturing process change and extension of testing time. From customer's perspective to cut cost, they are eager to stabilize yield and shorten testing time. In 2019, EUV will be rightly adapted, but the yield of 7 nano EUV is not stable yet. And this is where we have some expectation. As for TDDI, testing time is too long. And the chip on film vis a vis the chip on glass extends the testing time to two or three folds. And due to technological constraint for packaging, currently, the power testing limit is 4. With further use of COF, the display driver demands early increasing. I'm Damien Ton of Macquarie Capital Securities. What will be the impact of DDR5 on your company after 2020? Especially for your memory tester business. And Theradigm was saying that there will be coming into the high speed DRAM tester market Do you have any comment on this? DDR5 law in Turbby at the end of 2019 after shifting 3 months. Application of DDR5, ruby Sabre CPU and Smartphones, and AirP DDR5 is also relevant to new smartphones by 3 major companies, including Chinese. Presumably, there will be a in the DDR5 in 2020, given that mass production will be at the end of this year, 2019. Memory makers were shipped samples the CS and ES around June. Advantest has already shipped solutions. Why our competition also shipped solutions, considering DDR5 specification of 6.5 gigabytes and the recruiting margin 8 gigabps3b, the appropriate one. In high speed, the tester market, because of Adventist technological edge, we think we can gain market share. With this, I'd like to close the Advent Test Information Meeting for FY2018 sad quarter. Thank you very much for joining us despite your busy schedule.