[Non-English content]
Good morning, and good afternoon and evening to those calling in from abroad. This is Park Seong- Hwan, the head of IR at SK hynix. Welcome to the SK hynix 2021 second quarter earnings release conference call.
[Non-English content]
Before starting the conference call, allow me to introduce the executives present here with me today. First, Kevin Noh, CFO and Head of Corporate Center, Jang Hyeok Jun, Head of Finance, Park Myoung-Soo, in charge of DRAM Marketing, and Park Chan-dong, in charge of NAND Marketing.
[Non-English content]
Let me issue a disclaimer that all outlooks presented by the company are subject to change depending on the macroeconomic and market circumstances.
[Non-English content]
With that, we will now begin SK hynix 2021 second quarter earnings release conference call. Kevin will first present the earnings for the second quarter and the company's plan. It will then be followed by a Q&A session with the executives present here today.
[Non-English content]
Good morning, everyone. This is Kevin Noh, CFO and Head of Corporate Center. I will first brief you on the company's financial performance in the second quarter of 2021.
[Non-English content]
Memory market in the second quarter showed rapid improvement thanks to solid demand for PC, graphics and consumer products, as well as recovery in the demand for server products compared to the previous quarter.
[Non-English content]
To respond to the growing demand, bit shipment increased by mid single digit percent and low single digit percent for DRAM and NAND respectively, slightly outperforming the guidance. ASP also grew by a high 10% and approximately 10% in DRAM and NAND respectively, rising across all applications.
[Non-English content]
With DRAM sales growing significantly by high 20% QoQ, MCP sales also grew on the back of stronger sales of high-density products, such as DRAM 8 GB and NAND 256 GB or higher, with higher specifications for mobile products in the Greater China region. Revenue increased by 11% quarter-on-quarter. The portion out of sales fell slightly from the previous quarter.
[Non-English content]
This brings the company's consolidated sales in the second quarter to KRW 10.32 trillion, up 22% QoQ and up 20% YoY. This is meaningful in that quarterly sales topped KRW 10 trillion again for the first time in three years since the second and third quarters of 2018, when the memory market was in a super cycle.
[Non-English content]
Cost of goods sold and SG&A went up by 4% and 17% respectively. There were initial startup costs for the M16 fab, which opened early this year, as well as the annual increase in payroll applied retrospectively and payout of the ESOP.
[Non-English content]
Despite increase in expenses, there was still a slight reduction in unit cost QoQ for both DRAM and NAND, thanks to the growth in sales portion of DRAM 1y nano and 1z nano, as well as the 128-layer NAND.
[Non-English content]
Operating profit in the second quarter was KRW 2.69 trillion, up 103% QoQ. Operating profit margin was 26%, higher by 11 percentage points QoQ.
[Non English content]
Depreciation and amortization in the second quarter was KRW 2.62 trillion, slightly increased from the previous quarter. EBITDA was KRW 5.31 trillion, with EBITDA margin of 51%.
[Non-English content]
There was a net non-operating income of KRW 106.5 billion. There was net interest expense of KRW 63.4 billion, net foreign currency-related gain of KRW 42.2 billion. There was also gain on equity method valuation of KRW 125.5 billion from the building sell-off by SK China, where the company has some equity ownership.
[Non-English content]
Net profit before tax was KRW 2.8 trillion, corporate tax expense was KRW 0.81 trillion. Net profit was KRW 1.99 trillion with a net profit margin of 19%.
[Non-English content]
Consolidated cash balance at the end of Q2 was KRW 6.64 trillion, up by KRW 0.65 trillion from the previous quarter. Interest-bearing debt was KRW 13.85 trillion, increased by KRW 0.25 trillion QoQ. The company's debt-to-equity ratio and net debt-to-equity ratio stood at 25% and 13% respectively, slightly improved from the previous quarter.
[Non-English content]
Next is the company's market outlook and plan.
[Non-English content]
The memory market in the first half improved more swiftly than expected. DRAM demand growth exceeded the projected 20%, despite the shortage of some parts, and is now expected to be in the low 20%.
[Non-English content]
Demand growth was likewise robust for NAND, which was initially expected to improve in the second half of the year, with prices turning around to a steep increase in the second quarter. Annual demand growth is also projected to be higher than initially thought, to be in the mid to high 30%.
[Non-English content]
Breaking down the demand environment by application, the PC market is expected to see over 10% shipment growth for the second year in a row. Following the PC demand for educational PCs, it will be enterprise PCs that will drive demand in the second half.
[Non-English content]
The mobile market initially anticipated to recover quickly with the accelerated transition to 5G and high pent-up demand, was slow due to another surge in COVID-19 in some regions, as well as some component shortages. Demand is likely to recover in the second half, as COVID is coming under control in major markets, including India, and smartphone demand is recovering in the Greater China region. Plans for new product launches and seasonality will further drive demand.
[Non-English content]
As for the memory content per box, there is accelerated growth in demand for high memory density, like 8 GB DRAM and 256 GB NAND, thanks to the ongoing penetration of 5G models and the growth in high-end markets. Especially mobile phones that adopt 12 GB DRAM and 256 GB NAND are expected to be introduced in the market in the second half, which will further drive the mobile memory demand.
[Non-English content]
In the server market, demand from hyperscale customers began to improve from the second quarter with growing investment in new data centers. Solid growth in demand is expected into the second half with the newly launched CPU driving replacement demand and corporate IT spending also picking up. The improved performance by the new CPU will also drive up the memory content per box, likely to increase demand in the second half for high-density server DRAM modules of 64 GB or higher.
[Non-English content]
Given such market environment, the memory market in the second half will continue to see favorable trends, to be further aided by seasonality that will start in the third quarter. For NAND, in particular, improvement in the supply-demand dynamics will accelerate thanks to launch of new high-density mobile products and strong demand for Enterprise SSD.
[Non-English content]
With the stronger-than-expected demand continuing this year, memory suppliers are utilizing their inventory as much as possible. DDR5 products are set to go into production next year, and with die size penalty versus DDR4 due to their on-chip ECC feature, it is expected to affect DRAM supply thereafter.
[Non-English content]
SK hynix is planning for a low single-digit bit shipment growth for DRAM and high 10% growth for NAND in the third quarter. For the year, we maintain the guidance given out in the first quarter that DRAM bit shipment is expected to be in line with market demand growth, and NAND bit shipment growth is planned to far outpace the market growth rate.
[Non-English content]
The company's mainstream product for this year, DRAM 1z nano, went into full-scale ramp-up in the second quarter, helping reduce unit cost. In the second half, the company will solidify its position in the DRAM market by actively addressing the 64 GB or higher density server demand by expanding the production of 1z nanometer-based 16 Gb products.
[Non-English content]
1a nanometer product, which applies EUV technology, recently started volume production for mobile application to be supplied to customers starting in the second half. DDR5, whose demand is expected to grow from next year, will also go into mass production in the second half of the year to ensure timely response to customers' needs and support market's transition to DDR5.
[Non-English content]
For NAND, the company's mainstream product this year, the 128-layer, went over 50% in sales portion in the second quarter, also contributing to unit cost reduction. The next generation, the 176-layer, will go into mass production by the end of this year, at which time the sales portion of 128-layer and 176-layer combined will be close to 80%.
[Non-English content]
The company will aggressively expand the portion of the 128-layer, whose competitiveness is well recognized, to respond to demand for new mobile products and eSSD that are expected to surge in the second half. This will help quicken the pace of improvement in NAND profitability to turn around to profit in the third quarter and lay the basis for sustainable profit generation.
[Non-English content]
Last is the company's ESG management activities and achievements in the previous quarter.
[Non-English content]
In April, the company received the highest rating from the Carbon Disclosure Project, or CDP, a global project for information disclosure in environment. Of the approximately 9,000 participating global companies, SK hynix earned top scores in climate change and water security in 2020 and was selected for the Platinum Club in climate change and awarded first prize for water security by CDP Korea.
[Non-English content]
Particularly for the Platinum Club, the company has maintained its position for eight years in a row since becoming the first member in Korea in 2013. We will keep up our efforts on climate change issues and water resource management to sustain the honor.
[Non-English content]
In the second quarter, we published the 2021 Sustainability Management Report, faithfully reflecting the key standards of the SASB, or the Sustainability Accounting Standards Board, and the recommendations of the TCFD, Task Force on Climate-related Financial Disclosures. We also published the Human Rights Management Report for the first time, detailing the company's human rights related activities such as human rights management policy, risk management system, impact assessment, and improvement activities.
[Non-English content]
The company will keep up our effort to not only improve financial performance, but also ESG management and external communication to keep working towards sustainable growth. Thank you very much for your attention.
[Non-English content]
With that, we will now take your questions.
[Non-English content]
The first question will be presented by S. K. Kim from Daiwa Capital Markets. Please go ahead with your question.
[Non-English content]
First of all, congratulations on your performance. Now my question is pertaining to the DRAM market. Now in the second quarter, we see that the DRAM prices went up quite sharply. Going into the third quarter, of course, there is expectation that the prices will continue to rise. At the same time, there appears to be some resistance against the price increase, especially per certain products. Now for the data centers as well, there are some concerns in the market regarding the inventory level among some of the data center customers. Now, if in the second half, if the inventory continues to rise, then there are some concerns among the investors that the prices might start to fall in the fourth quarter of the year.
With the increased investment by the suppliers, there's also the likelihood that supply will increase. What is the company's view regarding the DRAM market in the second half? Also, if possible, what is your outlook for the DRAM market in next year?
[Non-English content]
First of all, let me say that your questions are quite complex and also includes very important points. Please understand that my response is going to be a bit longer winded than usual so that I will be able to fully respond to the questions. Compared to looking back to the market early this year or even late last year, the most important question was the COVID situation or how we were going to recover from the COVID situation. In response to that, how the suppliers were going to address the growing demand coming out of the COVID situation. At that time, the thinking was that just utilizing the production capacity was not enough, and that we had to tap into the inventory in order to fulfill the growing demand.
[Non-English content]
Looking at it from that perspective, then we see that demand starting this year was much stronger than expected. Especially demand coming from the 5G-based smartphones was stronger than expected, as well as the demand for PCs. The demand in general was stronger than what we had initially expected at the early part of the year. This, of course, resulted in, as expected, a reduction in the inventory level on the supply side.
[Non-English content]
Looking at it from those big trends, then we see that in terms of the supply-demand dynamics in the market, they are stronger than we had initially expected in the market, especially for the consumption demand. Of course, there have been some corrections to the demand because of the parts supply disruptions, but then not so much so that it would alter the fundamental course of the demand growth.
[Non-English content]
Looking at the current trend, which are likely to continue into the end of this year and also all the way into next year, we believe that because of both the demand and supply factors combined in the industry, we believe that for the industry overall, the inventory level will continue to decline.
[Non-English content]
Also from the demand perspective, there are also a number of important factors that we have to watch out for in the second half and also into next year. In the second half, there are a number of commerce events that have been planned in various parts of the world. Into next year, we are likely to move further away from the COVID-19 situation, meaning that the demand recovery coming from the consumer sector will now then move on to the enterprise sector.
[Non-English content]
Overall, we believe that the enterprise demand will be further stimulated moving into next year. Also after the COVID-19, in the post-COVID era, we believe that there will be demand growth coming from the 5G standalone clients. This will further stoke demand for the 5G smartphones as well, also combined with the growth in the 5G-related infrastructure. This means that compared to the pre-COVID situation, the IT industry's time series, I would say, has been pulled up by about two to three years.
[Non-English content]
A bit more about the supply side. Next year, looking at the market and the industry overall, we believe that continuous growth has to come from new and emerging technologies like the LPDDR or DDR5 or the EUV process. This is likely to temporarily pull down the productivity, but also increase the product value. For the market in general, we believe that there is going to be continuous growth in the size of value.
[Non-English content]
[Non-English content]
The next question will be presented by Nicolas Gaudois from UBS. Please go ahead with your question.
Good morning. Thanks for taking my questions. Going back to demand, I think what investors have been worried about is really any variations in demand into the second half of 2021. Maybe if you could broaden the spectrum beyond DRAM to NAND flash as well and outline if you've seen any changes up or down in the second half of the year so far for specific segments. You just commented on inventories for DRAM potentially coming down. Could you maybe provide a little bit more color on what you expect each segments to do in the second half of the year, and maybe NAND flash as well? Secondly, you just talked about DRAM 1a nanometer production starting on the mobile side, which seems to be slightly early. How would you characterize yields so far?
What portion of wafer starts should be at 1a nanometer by Q4 of this year? What's the broad target for Q4 2022 as well? Thank you.
[Non-English content]
[Non-English content]
First, regarding the demand by application in the second half. First, regarding the server side, we see that the build demand continues to remain solid for server, and also the related demand also remains largely unchanged.
[Non-English content]
For the server side, we see the demand likely to remain solid for some time, thanks to the CPUs that have been launched this year, going into maturity, and then also the new CPU slated for launch next year. For PCs, now we actually see that compared to what we had expected, the demand coming from the customers was a bit stronger than our expectation. Overall for the demand for memory, now on the low-end side, for the low-end products, there have been some issues, especially for the Chromebook because of the part supply. We see that the demand actually has then shifted to the high-end segment. We see that for the low-end memory mix, there is likely to be a lower memory mix, while for the high-end, the memory mix is actually going to increase. Overall, this is going to even out the overall demand as well.
[Non-English content]
For the inventory level. Now, given the breadth of the part supply issues across the SCM, it is very difficult for us to speak regarding the inventory for the industry on the whole at this time. Our outlook is that for the industry, so looking at both the demand and supply side, then looking at both factors, then we believe that the inventory level will continue to decline into the end of this year and also into next year.
[Non-English content]
Regarding your question about the 1a nanometer. Yes, as you have also mentioned, we have started the volume production for the 1a nanometer product for a mobile application. As you would know, the 1a nanometer product is the first time that SK hynix is going to volume production utilizing the EUV technology. Having said that, although we are using the EUV technology for the 1a nanometer volume production, it also has somewhat of a testing purpose too. This is also going to be the last product where the existing technology platform will be maintained. Yes, on one hand, we have the volume production and so this also means that in terms of the yield, we have the economically viable yields so that we can go into the volume production.
Also at the same time, in order to reach the cost crossover with the previous generation, it would take some time. Given that we are utilizing this existing technology platform, we believe that we will be able to have a rapid yield catch up.
[Non-English content]
[Non-English content]
The next question will be presented by Ricky Seo from HSBC. Please go ahead with your question.
[Non-English content]
I have some questions about NAND. First, I see that the company's margin for NAND business has been improving much more quickly than expected. You have touched upon some of the factors earlier, but could you elaborate a bit more about the factors that are driving the margin improvement, for example, cost reduction? Also, given the bit growth guidance, I believe that we can expect the margin to continue to improve. What is the appropriate level of margin that the company sees, if you could give us an outlook regarding that? Another question is about DRAM, and there are some concerns about the demand movements in the fourth quarter and early part of next year.
Now, depending on the demand situation, does the company have plans or plans to make changes to the CapEx implementation or maybe some changes to your timeline of the CapEx?
[Non-English content]
First about the NAND market. We see that demand continues to increase in the second half due to the new CPU launches in the server and also for the mobile side, the continuing penetration of 5G and new product launches.
[Non-English content]
Combined with the favorable seasonality, we believe that the demand for NAND in the second half is likely to swiftly increase compared to the second quarter.
[Non-English content]
Especially there is a demand for high-density memory coming from the mobile companies in the Greater China region, which we believe will further drive demand. Overall, the demand for sales for the NAND in general will continue to increase.
[Non-English content]
Regarding your question about the appropriate margin for NAND. Of course, for the company, the higher the margin, the better. Of course, what is important is sustainability as well. As we had explained earlier, we are looking toward a quarterly turnaround in the third quarter of this year. For the year as well, we are expecting a turnaround to profit. For the 128-layer and 176-layer products. Based on these products, you would know the company's competitiveness. Thanks to the competitiveness of these products, we believe that we will be able to maintain profitability next year as well. Having said that, there is the factor of the Intel NAND business acquisition, which we are planning to complete the closing of by the end of this year.
As a result of this, there could be some one-off costs to be incurred next year. With the combination of these two businesses, we believe that the market share is going to increase quite considerably. Also in that process, there would be some synergy effects to be gained in cost as well, which is likely to more than offset the one-off costs related to the combination. All in all, we believe that the NAND turnaround to profit for the medium to longer term will continue.
[Non-English content]
Regarding your question about the DRAM market's implications on the company's CapEx plan. Now, if I may take a step back and explain about the cycles in the market. Now, usually there are various cycles in the memory industry because, compared to the fluctuations in the demand, it's not as easy to make adjustments to the CapEx on the supply side. As a result of this, there have been some fluctuations in the cycle so far. What we are seeing more recently is that addressing the demand changes, there is a much quicker decisions regarding the investment changes on the supply side, meaning that the lead time has been much shortened. This means that the speed to address the changing market needs has also become much faster.
That is why the cycles that we are seeing recently are not the type of conventional or the typical cycles that we are used to. Whenever there is a downtime, then a demand will pick up more quickly than before. Likewise, any boom time is not going to last as long as it used to in the past.
[Non-English content]
We believe that such changes bode well not only for the interest of the memory suppliers, but it is also better for the customer side, meaning that it is beneficial in increasing the overall social welfare. Now, having said that, now for the, let's say, any changes in the market, the company is also able to adjust the CapEx more agilely than before.
At the same time, as we have explained earlier, looking at the overall suppliers, then the inventory level, especially for DRAM, is considerably low at this time. For next year, if the demand continues to grow at this pace, then we would have to make additional CapEx, invest additional CapEx into the production capacity in order to fulfill the demand. Because of this, what we have announced in the last quarter about our CapEx plan to fulfill the demand for next year will continue as planned. Having said that, the CapEx size is not going to change because of the, let's say, growth in demand in the fourth quarter.
[Non-English content]
[Non-English content]
The next question will be presented by Hyunwoo Doh from NH Investment & Securities. Please go ahead with your question.
[Non-English content]
My question is regarding the DDR5. This was mentioned briefly earlier, but I believe that that is one of the most hotly anticipated product in the market in 2022. With the DDR5, of course the die density will grow, but also there will be some die size penalty, as was mentioned. With the launch of the DDR5, what does the company expect in terms of any changes to the demand, supply or the pricings?
[Non-English content]
Regarding the launch of the DDR5 next year, yes, we believe that next year is going to be the opening of the era of the DDR5. Looking at it from this perspective, one is from the demand side. We believe that the DDR5 is also going to drive the density for DRAM as well.
It is probably going to move the DRAM to a higher density. By the end of the year, the 64 GB in the DDR4 is going to reach close to the half line point. We believe that this trend is going to continue in the server segment next year.
[Non-English content]
From the perspective of the product value, now, because the DDR5 offers very clear advantages in terms of performance and also combined with the new CPU to be launched next year, we believe that it will serve to further drive up the overall value of DRAM.
[Non-English content]
To make sure that we can launch the DDR5 products not only for the server but also for the PCs, we are leading the way, working together with the partners and customers in terms of the launch in the market and also for the ramp-up of the production. We believe that the DDR5 will further solidify our competitiveness in the server market.
[Non-English content]
[Non-English content]
The next question will be presented by Dong-je Woo from Bank of America. Please go ahead with your question.
[Non-English content]
Since we are on the topic again regarding the DDR5. Currently, Intel CPU does not support DDR5, and Intel also has the track record of having some delay in the launch of their new CPUs compared to the plan. I wonder whether your outlook for the DDR5 is based on the assumption that the Intel's new CPU, for example, the Sapphire Rapids, will be launched on schedule in the second quarter, or is it related to other, like AMD demand and so forth?
[Non-English content]
Regarding the, let's say, the CPU suppliers or their, let's say, timeline in the past or in the future, I don't think we are in this position to make comments on that. For any factors, of course, we would have to think about the potential impact and be ready for it. For next year, the launch of the DDR5 itself is going to be an event that is going to drive further demand. Of course, if there are some, let's say, unexpected factors that occur, then we believe that still there would be a healthy environment to keep driving demand. For example, the CPU launched this year will be maturing. There can be a combined usage of DDR4 and DDR5 and also the demand mix coming from both the mobile and the server side.
We believe that the demand will continue to remain solid. Of course, what you have mentioned is going to be one catalyst for demand, but we do not believe that is going to turn into a big risk.
[Non-English content]
My follow-up question regarding the M16 fab. The M16 fab has gone into operation, but I understand that at this site there is no more land to build new facilities. In order to build a new fab in the Wonsam-myeon area, I understand that it would take at least a few years. Do you believe that with the existing M16 size, it will be enough to sustain the business for the next three to five years, for example, utilizing more EUV and also moving on to 1b, 1c technology.
Do you believe that you will have enough space for the clean rooms for the continuous tech migration? In addition to that, I would like to gain a better understanding of the company's overall fab operation plan. What we used to call the M10, for their conversion to CMOS and also the facilities in Cheongju, what are your plans for those as well?
[Non-English content]
You're right that the new site that we have secured in the Wonsam-myeon of Yongin, now this is for the company to secure a new site for a new fab. Yes, as you have mentioned in the Icheon site there is not enough land left to build another large scale fab. The M16 fab, now it is much bigger in size than the existing fabs.
For example, other fabs are two layered, whereas for the M16 it is three layered, including one layer for EUV. In terms of capacity, it is much bigger than the existing fabs and it just got started in its operation. Until we have the, let's say, if necessary, until we have a new fab ready in Yongin, we believe that it will provide us with enough capacity for the next few years. That having said that, of course, the memory industry tends to be very dynamic and also subject to rapid changes. In the event of such drastic changes in the demand, then of course we do have some backup options. For example, we could move some of the NAND capacity from M10 to Cheongju, or also the M10, which is slightly outdated, so is not being fully utilized.
We can also put the M10 into more utilization. Of course, there is always the option of utilizing the Wuxi fab more actively. We do have such a backup plan, but our assumption remains the same, that the M16 will provide us with enough capacity for the next few years. Let me make one correction. I mentioned that if such a drastic demand occurs, then we could also move some of the NAND from M10 to I'm sorry, M14 to Cheongju, not M10.
[Non-English content]
[Non-English content]
The next question will be presented by Min Seong Hwang from Samsung Securities. Please go ahead with your question.
[Non-English content]
From the supply side, we see that your peers are increasing supply more than anticipated. When it was announced that some of the investment plan for next year will be pulled up, the market reacted somewhat negatively. What is the company's plan for the capacity for this year and next year for M16 DRAM and M15 NAND? What do you believe is going to be the production capacity? It was mentioned that the industry's inventory in general is going to decline. Do you believe that this merits a higher CapEx investment next year compared to this year?
[Non-English content]
From the supplier side in the market, our perspective is how we can provide enough supply to meet the demand coming from the market. That, I would say, is the most important question for the supplier.
What I mean is that we are not really thinking about or acting out of fear that what is increased CapEx leads to changes in the supply-demand dynamics and also because of the competition among the suppliers. That is not our particular worry. It's more about how we are going to meet the strong demand coming from the market. Regarding the wafer capacity at M15 and M16, I do not believe it is appropriate for me to make comments on them at this point. Again, our basic position is that there has to be sufficient CapEx investment made in order to fill the demand. What I mean by CapEx investment, it's not just about the absolute size, it's more as a percentage out of the sales or EBITDA.
When we think of it in those terms, and I do not believe that it is going to increase dramatically. Especially given the strong demand and the rise in the ASP, I do not believe that the CapEx for next year as a percentage out of the sales or the EBITDA is going to increase all that much. It is probably going to be maintained around the same level.
[Non-English content]
[Non-English content]
The last question will be presented by JJ Park from JPMorgan. Please go ahead with your question.
[Non-English content]
My question first, I believe that the CFO also touched upon this earlier, about the EUV adoption for the 1a nanometer. I understand that the EUV adoption at this point is not significant. In other words, it is only applied for one layer. I also understand that the plan is to start applying the EUV at multiple layers for 1b nanometer products. Of course, I believe that each company would have different strategy for the EUV technology. Let's say for the 1b nanometer product, compared to EUV process and non-EUV process, what do you believe are the advantages of the EUV process? For example, is the die size going to be smaller or the process time going to be much shorter? Just about the advantages of the EUV process when applied to the 1b nanometer.
My second question is a request for update on the Chinese authorities approval for the Intel NAND business acquisition, because I understand that the approval has not been given yet. Any update?
[Non-English content]
For the 1a nanometer, yes, the EUV application is not that many in terms of the number of layers. That is why, as I explained earlier, we are maintaining the existing technology platform, and we are applying the EUV on the 1a nanometer products for testing purpose. As you have mentioned, different companies would have different approaches to the EUV. Some would try to put their existing EUV equipment to more utilization, while some might try to extend the use of the immersion technology as much as possible. I would say that we probably stand somewhere in the middle. I believe that the different companies would have their own EUV adoption pace according to their circumstances.
[Non-English content]
Now, without going into much technical details, if I may just give you a high level explanation, then for what's called the tech shrink. Going from 20 nanometer to 10 nanometer, then to under 10 nanometer, means that we get more chips out of one sheet of wafer. This also means that we can further reduce cost. Now between the EUV and the immersion technology, now EUV's wavelength is shorter, meaning that it is more conducive to enabling more detailed design on the wafer. For the immersion technology, the wavelength is longer and also because of the inherent technical limitations, for certain designs, it had to go through the process three or four times in order to get the right design.
For the same design, EUV can do this in one iteration, meaning that the number of process steps can be reduced quite a lot, which also is much helpful for cost reduction, too. Having said that, EUV also has some challenges, especially when you're adopting it for the first time, there could be some unexpected technical difficulties. Depending on who can overcome such difficulties, how quickly would also spot the difference in securing the competitiveness.
[Non-English content]
Your second question was about the approval for the Intel NAND business acquisition. Now, including China, we went through the antitrust review in eight countries, and we have received the unconditional approval from seven countries, including the U.S. and the EU. The last in the approval process is to come from China. For this, what we are hoping for is to receive the approval in an appropriate time in the second half of the year, so that we will be able to have the closing of the deal by the end of this year as we are planning to. Right now the process in China is in the final review phase.
[Non-English content]
Thank you very much. That concludes the SK hynix 2021 second quarter earnings release conference call. Thank you very much for your participation.