Why is Japan’s share of front-end process equipment declining?

The share of front-end equipment across Japan has been declining since around 2013.

Figure 1: Changes in front-end process equipment shipments and regional share (until 2021) Source: Author based on Nomura Securities data

The article seems to have sent ripples in all directions. That’s because, on July 20, Nikkei Crosstech published an article titled “Japan’s Semiconductor Manufacturing Equipment, Responding to Declining Global Market Share with Next-Generation Technology”. Separately, on July 28, the Nikkei Electronic Edition published an article with the exact same content, titled “Japan’s semiconductor manufacturing equipment, next-generation technology regains market share.”

But why is Japan’s share of front-end process equipment declining?

In my own previous article, I pointed out that the share of all front-end process equipment (excluding mask inspection equipment) will decline over the decade from 2011 to 2021 (Figure 2). I also think that this is directly related to the decline in the share of front-end equipment in Japan.

Figure 2: Front-end equipment market share (2011 and 2021) Source: Author based on Nomura Securities data

However, I feel that this analysis alone is not enough. Therefore, in this article, we will once again look at the declining share of front-end equipment in Japan. The analysis results show that European and American companies are concentrated in the equipment field with a large market size, and their market share has increased.

This problem will not be solved unless Japanese companies grow at a rate that matches and exceeds that of Western and South Korean companies. After all, it can be said that the situation is serious.

Strategies and Concentrations of Western Companies

Figure 3 shows the market share of each company’s front-end process, the share of Europe, the United States and Japan, and the market size in 2021. Japan has a high market share in the equipment field, but the market size is not large. Specifically, the 90.6% share of the coating machine/developing machine is 3.3 billion US dollars, the 94.7% share of the heat treatment equipment is 3.8 billion US dollars, the 61.4% share of the single-chip cleaning equipment is 4.1 billion US dollars, and the 90.7% share of the batch cleaning equipment equipment costs 900 million US dollars , 68.7% share of CD-SEM costs $800 million. There is no Japanese monopoly in the field of equipment, and the market size exceeds 5 billion US dollars.

Figure 3: Front-end equipment market share by company, market share in Japan, US and Europe (2021) Source: Author based on data from Nomura Securities

On the other hand, in the field of equipment with a market size of more than 10 billion US dollars, European and American equipment manufacturers occupy the market share. Specifically, ASML accounted for 95% of the $16.4 billion exposure equipment, and Lam Research (Lam) and Applied Materials (AMAT) combined accounted for 64.8% of the $18.9 billion dry etch equipment, or about $10 billion. KLA and AMAT account for 73% of the $10.4 billion worth of visual inspection equipment.

In other segments with a market size of less than $10 billion, AMAT accounted for 86% of the $4.4 billion in sputtering equipment, 68% of the $2.8 billion in CMP equipment, and $3.5 billion in pattern inspection. KLA and AMAT together monopolize 87.5% of the equipment.

ASML, AMAT, Lam and KLA are the four Western companies that dominate the market for various types of equipment, four of which are worth more than $10 billion. In short, the four equipment manufacturers in Europe and the United States firmly occupy the equipment share in the field of large market size. This shows the strategic nature and concentration that is unique to Western companies.

Equipment Manufacturer Sales Ranking

From the current analysis, 4 European and American equipment manufacturers have monopolized the market share in the equipment field, and the market scale is relatively large. When did these four companies start taking market share?

Figure 4 shows the top 10 sales rankings of device manufacturers from 2000 to 2021. As can be seen from this picture, since 2007, the top five rankings have been fixed as AMAT, ASML, Lam, Tokyo Electron (TEL), and KLA.

Figure 4: Top 10 Sales Rankings of Equipment Manufacturers (2000-2021) Source: Produced by the author based on the electronic journal “Semiconductor Manufacturing Equipment Data Book” and data from Nomura Securities.

However, the top five rankings change from time to time. For example, AMAT topped the list, while ASML only topped the list in 2011. In addition, TEL, which ranked second from 2000 to 2008, has gradually declined since 2015, and has been ranked fourth since 2015.

So, from 2017 to 2021, the order has shifted to 1st AMAT, 2nd ASML, 3rd Lam, 4th TEL and 5th KLA.

Starting in 2007, the top 5 has been determined, but the top 6 and below fluctuate wildly every year. Among them, since 2013, SCREEN, which has the world’s largest market share of cleaning equipment, has occupied the sixth place. However, South Korea’s SEMES, which entered the top 10 for the first time in 2016, has jumped to No. 6, replacing SCREEN in 2021.

This is a bit off topic, but I’ll explain the history of SEMES and how it developed.

Background of SEMES

In 1993, Samsung Electronics (Samsung) became the world’s No. 1 company in terms of DRAM sales. It is said that Samsung had entrusted Japanese equipment manufacturers such as Nikon, TEL, and Dainippon Screen Manufacturing (referred to as DNS, later SCREEN) to manufacture equipment for Samsung in South Korea.

However, many Japanese manufacturers refuse to do so, fearing technology leaks. However, at Samsung’s request, SCREEN established K-DNS as a joint venture with Samsung in January 1993. K-DNS probably means “Korea’s DNS”. A large number of SCREEN engineers are dispatched to K-DNS, where Samsung’s cleaning equipment is produced (Fig. 5).

Figure 5: SCREEN (former DNS) and SEMES (former K-DNS) equipment sales Source: author based on electronic journal Semiconductor Manufacturing Equipment Data Book and Nomura Securities data

In fact, the first locally produced cleaning equipment was delivered to Samsung’s Qixing factory in May 1994. Since then, K-DNS has continued to provide Samsung with cleaning equipment. K-DNS was renamed SEMES in January 2005. “SE” in SEMES stands for Samsung Electronics.

In 2010, SCREEN sold its stake in SEMES (or rather, it was acquired by Samsung), making SEMES a wholly owned subsidiary of Samsung. Taking this as an opportunity, all of Samsung’s cleaning equipment is manufactured by SEMES. It is reported that SCREEN can no longer supply any equipment to Samsung. In short, SCREEN rented out the eaves and the main building was taken over.

In 2021, SEMES will eventually overtake SCREEN in equipment sales, albeit by a small margin.

Now, back to the topic. Next, let’s take a look at the sales trends of major equipment manufacturers.

Sales Trends of Major Equipment Manufacturers

Figure 6 shows the sales trends of major equipment manufacturers in Europe, the United States, Japan, and South Korea. Many device makers had a big peak during the IT bubble in 2000, a decline after the 2008 Lehman shock, a peak during the 2018 memory bubble, a memory recession in 2019, and a Covid-19 storm in 2020. Seems to be increasing sales regardless.

Figure 6: Sales Trend of Major Equipment Manufacturers (2000-2021) Source: Author based on electronic journal Semiconductor Manufacturing Equipment Data Book and Nomura Securities data

Since 2013, the gap between the top four (AMAT, ASML, Lam, TEL) and the 5th KLA has widened, and the gap between the 5th and the 6th and below has also widened. If you compare it to a marathon, the leading groups are AMAT, ASML, Lam and TEL, while the KLA is a little further afield at 5th. I’m guessing this is the location.

Therefore, below we will analyze the growth potential of equipment manufacturers ranked 5 and above and equipment manufacturers ranked 6 and below.

Growth Potential of Top 5 Equipment Manufacturers

Every equipment manufacturer had a peak during the 2000 IT bubble. So, I created a graph that normalized sales per company to “1” in 2000 (Figure 7). In short, I want to clarify how much the top five companies have grown in the 21 years since 2000.

Figure 7: Top 5 Standardized Equipment Manufacturers by Sales, 2000 Source: Author based on electronic journal Semiconductor Manufacturing Equipment Data Book and Nomura Securities data

As a result, Lam is the fastest growing, with sales increasing 8.9 times in 21 years. Below, in descending order of growth rate, ASML is 5.45 times, KLA is 3.11 times, TEL is 2.76 times, and AMAT is 1.75 times.

Next, I created a graph normalizing each company’s 2011 sales to “1” (Figure 8). This result is more important than Figure 4. This is because this has the potential to lead to a rapid decline in the share of front-end equipment in Japan since 2013.

Figure 8: Top 5 standardized equipment manufacturers by sales in 2011 Source: Author based on electronic journal Semiconductor Manufacturing Equipment Data Book and Nomura Securities data

Looking at Figure 8, Lam again showed the highest growth, with a 4.62-fold increase over the past ten years. The second-placed group is in a tight contest. The second place is 2.49 times of AMAT, the third place is 2.29 times of TEL, the fourth place is 2.06 times of ASML, and the fifth place is 1.90 times of KLA.

Arguably, the only Japanese company in the top five, TEL, is doing well, but nowhere near Lam’s growth potential. So why did Lam get so big?

What is the source of Lam’s growth?

First, in 2008, Lam acquired SEZ, an Austrian manufacturer of single-wafer cleaning equipment. Around 2008, the paradigm of cleaning equipment shifted from batch to monolithic, allowing SEZ to contribute to Lam’s growth.

In 2011, Lam acquired Novellus, a maker of CVD equipment. Since 2016, NAND-type flash memory has become 3D, and the demand for CVD equipment has increased sharply, so Novellus has made a huge contribution to the growth of Lam.

In addition, Lam originally had the largest market share in dry etching equipment for gate and metal conductive films. Here, Lam began to eat into the market share of Cu/Low-k dielectric etch equipment that TEL put into use after 2000. Based on this technology, they almost monopolize the world share of HARC (High Aspect Ratio Contact) etching of 3D NAND memory cells.

In summary, it is believed that the acquisition of SEZ and Novellus and the expansion of sales of insulating film dry etching equipment such as HARC for 3D NAND flash memory are the sources of high growth.

Lam and TEL compete in dry etching equipment, CVD equipment, and single-wafer cleaning equipment, but TEL lags Lam in overall growth. This is believed to be one of the reasons for the decline in the share of front-end process equipment in Japan.

Growth Potential of Top 6 Under Equipment Manufacturers

Now let’s analyze the growth potential of the top six and below equipment makers. Figure 9 is a graph normalizing the sales of the top 6 and below major equipment makers with sales of IT foam at “1” in 2000.

Figure 9: Source of sales for the top 6 following equipment manufacturers in normalization in 2000: Authors based on electronic journal Semiconductor Manufacturing Equipment Data Book and Nomura Securities data

Note that normalized sales on the vertical axis are logarithmic. Surprisingly, South Korea’s SEMES increased 31.6 times. The growth of device makers other than SEMES is almost incomprehensible if the graph is plotted on the linear axis, so I had to plot it on the logarithmic axis. Even so, the growth potential for SEMES is enormous.

Outside SEMES, the growth rates from high to low are Kokusai electric 2.91 times, European ASMI 2.02 times, SCREEN 1.98 times, Hitachi High-tech 1.31 times, Ebara Corporation 0.81 times, Canon 0.67 times, and Nikon 0.15 times. The negative growth of Ebara Corporation, Canon and Nikon was less than 1 times.

Next, I created a graph normalizing each company’s 2011 sales to “1” (Figure 10). Again, SEMES had the highest growth rate of 4.33 times. Followed by ASMI 3.46 times, Kokusai electric 2.47 times, Ebara Corporation 1.81 times, Canon 1.47 times, SCREEN 1.04 times, Hitachi High-tech 0.97 times, Nikon 0.16 times.

Figure 10: 2011 Standardized Sales of Equipment Manufacturers Below Top 6 Source: Author based on electronic journal “Semiconductor Manufacturing Equipment Data Book” and data from Nomura Securities

Here, Kokusai Electric is the only Japanese equipment maker to more than double, and overall, the growth rate of Japanese companies is not high. SCREEN and Hitachi High-Tech barely grew, and Nikon’s growth rate was noticeably negative.

There are many Japanese equipment manufacturers in the top 10 in equipment sales. However, if we analyze its growth potential, it can be said to be dwarfed by SEMES in Korea and ASMI in Europe.

Summary and future outlook

Figure 11 shows the growth potential and equipment mix of major equipment manufacturers in Europe, the US, Japan and South Korea. We analyzed the growth rates in 2000 and 2011 by dividing equipment manufacturers into the top 5 or above and the top 6 or below. In Figure 11, blue shows high growth, green shows moderate growth, yellow shows moderate growth, and pink shows low growth.

Figure 11: Growth rate and equipment mix of major front-end equipment manufacturers Source: Author based on electronic journal Semiconductor Manufacturing Equipment Data Book and Nomura Securities data

TEL is the only Japanese company to make the top five. TEL’s growth rates from 2000 and 2011 were “a bit high”. Therefore, it can be said that TEL is performing well, but it does not match the high growth potential of Lam, which competes in the fields of dry etching equipment, CVD equipment, cleaning equipment, etc.

Below the top six, both SEMES and ASMI have shown high growth rates since 2011. Among them, Kokusai Electric is the only Japanese company rated as “slightly higher” growth rate from 2000 to 2011, which can be said to have performed well. However, no other Japanese equipment makers showed high growth potential. Conversely, low-growth companies are everywhere.

In addition, Western companies involved in the equipment field of more than 10 billion US dollars generally showed “high” or “slightly high” growth potential. On the other hand, unfortunately, it has to be said that the growth potential of Japanese equipment manufacturers is generally not as good as that of European, American and Korean companies.

We analyze the reasons for the rapid decline in the share of front-end equipment in Japan since 2013. The conclusion is that the four companies in Europe and the United States strategically focus on the equipment field exceeding 10 billion US dollars, occupying a stable market share, while Japan is not as good as Europe, America and South Korea in terms of sales growth. It is said to be the reason for the decline in market share. So what should Japanese equipment makers do?

How can Japanese equipment makers continue?

In the article mentioned at the beginning of the article, I explained the competitiveness of Japanese equipment manufacturers as follows (Figure 12).

“The equipment with a high market share in Japan often handles liquids, fluids, and powders, and the initial shape is not fixed. Therefore, the parameters that need to be optimized are many and very complex. In this case, the Japanese have found the best solution through experience and intuition. Optimal solution. This process has a lot of tacit knowledge and know-how that cannot be recorded, so it can be like a skill or craft.

In these worlds, continuous improvement and refinement on the shop floor is important, with diligent and patient Japanese optimizing parts down to the smallest detail. Therefore, equipment, materials and parts are created from the bottom up. It is believed that these Japanese characteristics are the source of high market share.

Figure 12: Differences in equipment development between Japan and Europe [click to enlarge] Source: Takashi Yunoue / EE Times Japan

However, this approach to device development in Japan may be reaching a turning point as devices of all kinds become extremely complex and highly precise. Then, in the future development of equipment, it may be necessary to make full use of Japanese characteristics and at the same time strive to absorb the advantages of Europe and the United States. The above article discusses the equipment development methods in Europe and the United States as follows.

“First, we captured the needs through marketing. The science was at the beginning of developing various devices. Based on those needs and the science, the whole device was architected by a strong top-down leader. At the time, it was usually modular.

In addition, we take full advantage of simulation in every process of device development. In addition, technology and know-how are translated into software and integrated into devices. Then, by putting these together, we’ll create a world-standard device… (omitted)

Stopping the decline in market share and increasing it again is no easy task. However, if Japan wants to maintain a high level of competitiveness in the field of front-end equipment, it is an unavoidable problem. This is a pivotal moment for Japanese equipment makers. I hope this will not repeat the decline of the Japanese semiconductor industry.

Source: Semiconductor Watch