Japan Sees Recovery in 2000
Hideki Motai, Manger, research Department, Press Journal Inc. -- Semiconductor International, 7/1/1999
In 1998, the Japanese semiconductor market
faced a strong adverse wind and experienced substantial negative growth. As
shown in Fig. 1, the market shrunk by 11.4% from the previous year to 3.655
trillion yen ($27.92 billion, assuming 130.9 yen = $1). Although high growth had
suddenly swung to negative growth in 1996, some positive growth was recovered in
1997. However, this did not lead to full-fledged recovery, and 1998 again saw
negative growth.
In dollars, the negative growth became even worse, falling by 18.1% from the previous year to $27.92 billion. The falling trend of the last several years could not be stemmed, and Japan's share of the worldwide market (in dollars) has also fallen from 22.6% in the previous year to 20.1%.
The biggest causes for this downturn were the decline in domestic production of electronic equipment due to the significant economic slowdown in Japan and Asia, and the fall of average unit prices of devices.
A vicious cycle exists in Japan, where worsening corporate performance has led to wage cuts and uncertainty about employment. In turn, this has chilled consumption, which further worsens corporate performance. As a result, the demand for electronic equipment has shrunk considerably, seriously dampening the semiconductor market.
 |
| Fig. 1 After a severe downturn in the late 1990's, Japan's semiconductor market appears to be back on track. |
On the other hand, MOS logic, which had been maintaining growth until the previous year, turned negative in 1998 because of falling unit prices. MPUs fell sharply by 29.5% from the previous year due to falling prices and anemic growth in the domestic PC market. MCUs, standard cells and ASSPs also saw a negative growth because of falling unit prices. As a result, MOS logic as a whole fell by 15.0% from the previous year to 1.4505 trillion yen ($11.08 billion).
Stagnation to continue in 1999
What is the outlook for 1999? Health of the domestic economy is the key. While some have declared that 'the bottom has been reached,' others maintain the pace of recovery will be very slow, leaving plenty of uncertainty.
Our projection is that the economy will bottom out around the middle of the year, and production will gradually begin recovering as uncertainty about the economy is gradually removed through stimulus measures such as the infusion of public funds into banks and tax reductions for housing loans.
However, since a favorable cycle -- in which income growth leads to increased consumption -- cannot be expected soon, significant recovery in consumption cannot be anticipated in 1999. Even if the economy begins a climb out of the bottom, income growth is unlikely due to corporate restructuring, which is in progress, and therefore expansion in personal consumption will be extremely slow.
 |
| Fig. 2 Japan's semiconductor equipment market is expected to see upward growth over the next several years. |
Positive factors may include growth in the demand for home and personal-use PCs, and market expansion for portable equipment and digital home appliances. However, here again, no rapid growth can be expected since income is not increasing. Measures for the Y2K problem may increase the demand for information equipment such as computers but will not be sufficient to really lift the domestic semiconductor market.
As a result, the 1999 semiconductor market, while falling less, will not achieve a positive growth. It is projected to reach only 3.5473 trillion yen ($27.1 billion), a fall of 2.9% from the previous year.
Recovery in 2000 and beyond
We project that full-fledged recovery in growth will begin in 2000 and beyond. This projection is based on assumptions that corporate capital investment will increase and income will gradually increase. This will lead to gradual expansion in consumption; continued solid growth of the PC market, mainly for home and personal-use models, and gradual emergence of the market for digital home appliances in Japan.
We expect the semiconductor market will grow to 3.78 trillion yen ($28.8 billion) in 2000 (6.5% increase from the previous year); 4.11 trillion yen ($31.4 billion) in 2001 (8.8% increase from the previous year); 4.53 trillion yen ($34.6 billion) in 2002 (10.2% increase from the previous year); and 5.6 trillion yen ($42.78 billion) in 2003 (11.8% increase from the previous year).
However, these growth rates are lower than the worldwide rates, and Japan's share of the worldwide market will continue to decline. This is largely because the electronic equipment market in Japan is saturated, unlike those in other regions. Japan's growth rate will be low because market penetration of electronic equipment is already high, unlike in Asia where many markets have growth potential. Japan's share is projected to fall to 16.2% in 2003.
Semiconductor manufacturing equipment market
The Japanese semiconductor manufacturing equipment market also experienced substantial negative growth in 1998, falling by 22.4% from the previous year and reaching only 600.4 billion yen. In dollars, the market fell by 30.2% from the previous year, to $4.4606 billion. Because this rate of decline is more severe than that for the worldwide market, Japan's share of the worldwide market continued to fall, from 27.7% in the previous year to 24.4%.
The biggest cause of this negative growth was the severe cutback in capital investments made by semiconductor makers because of stagnation in the semiconductor market. As a matter of fact, the total 1998 capital investment by 30 Japanese semiconductor makers fell sharply, by 37.1% from the previous year, and reached only 752.2 billion yen ($5.75 billion), as shown in Fig. 3.
Nearly all types of equipment have declined from the previous year. Especially hit hard was pre-processing equipment. Heat treatment systems and ion implanters fell by 37.4% from the previous year to 37.5 billion yen ($286 million); film deposition equipment (total of epitaxial growth systems, CVD systems, sputtering sys tems, etc.) fell by 27.2% from the previous year to 103.2 billion yen ($788 million); lithography systems fell by 13.6% from the previous year to 141.6 billion yen ($1.08 billion); and etching systems fell by 40.9% from the previous year to 36.5 billion yen ($279 million). The only exception was CMP systems, which managed to grow (by 5.6% to 19 billion yen or $145 million), but their growth rate slowed substantially from the previous year's 55.2%.
 |
| Fig. 3 In 1998, the total capital investment by 30 Japanese semiconductor makers fell sharply by 37.1% from the previous year. Gradual recovery is expected through 2003. |
Assembly and testing systems also experienced negative growth. Assembly systems fell by 16.9% from the previous year to 50.4 billion yen ($385 million) and testing systems by 22.3% to 132.2 billion yen ($1 billion).
Positive but slow growth in 1999
What is the outlook for 1999? Booking conditions began improving at the beginning of the year, a sign of market recovery. According to the Semiconductor Equipment Association of Japan (SEAJ), the B/B ratio has exceeded the threshold of 1.00 since January 1999 for equipment made in Japan (all equipment made by Japanese companies for both Japan and overseas) and since December 1998 for the Japanese market (all equipment made by both Japanese and overseas companies for Japan). February 1999 data show B/B ratios of 1.11 for equipment made in Japan and 1.29 for the Japanese market.
These improvements appear to be driven by investment for 0.18 µm production lines, investment for the introduction of Cu wiring processes and investment toward mass production of Direct RambusDRAM (e.g., memory tester purchases).
We project that this condition will continue in the future and, for 1999 as a whole, the manufacturing equipment market will wipe out negative growth. However, this will not likely result in a full-fledged high-growth recovery for the following reason: Since it takes several months before booking recovery is reflected on sales and many makers are in severe financial difficulties because of the semiconductor recession, there is little possibility that capital investment will grow from the 1998 level. Our estimate of the total 1999 capital investment by Japanese semiconductor makers shows only slight growth, rising by 2.0% from the previous year. We estimate the manufacturing equipment market will reach only 613 billion yen ($4.68 billion), growing only 2.1% from the previous year.
300 mm to be the driving force in 2000 and beyond
What is the outlook beyond 2000? The answer to this question depends on capital investment for building manufacturing lines for 300 mm wafers. If this capital investment occurs full-scale, the manufacturing equipment market definitely will revive. This is because investment as large as 200 billion yen will be required for building 300 mm lines.
Press Journal's simulation projects the capital investment required for building 300 mm lines as follows: 211.1 billion yen for a line that will mass produce 30,000 wafers/month for 256 MDRAM; 233.7 billion yen ($1.785 billion) for a line (with a production capacity of 20,000 wafers/ month) that will process multilayer interconnection logic products in which Cu is used for wiring; and 252.7 billion yen ($1.93 billion) for a line (with a production capacity of 20,000 wafers/month) that will process hybrid system LSIs (Cu for wiring) containing large-capacity DRAM. (In all cases, the figures include manufacturing equipment, gas and pure water supply facilities, buildings, etc.) Manufacturing equipment alone will account for 80% to 90% of these figures. Therefore, if the building of 300 mm lines goes full-scale, the manufacturing equipment market will be revitalized.
The question is timing. When will semiconductor makers begin building mass-production lines that can handle 300 mm wafers? For now, Japanese semiconductor makers are giving higher priority to the renewal of lines toward circuit dimension reductions than to the introduction of 300 mm lines. They appear less enthusiastic about capital investment for 300 mm lines than makers in the U.S., Europe and Korea.
Nonetheless, many Japanese makers believe that in order to strengthen their cost competitiveness and win in the market in the future, circuit dimension reduction alone will not be enough, and the use of larger wafers will be essential. There is no doubt Japanese semiconductor makers will begin building 300 mm lines sooner or later. According to our projection, building of 300 mm lines will pick up speed beginning in 2001 and the number of lines that will start running will increase rapidly in 2002 and beyond. Our simulation projects the timing at which the existing lines that handle 200 mm wafers will be in short supply, assuming a case involving mass production of DRAM. When the worldwide PC shipment is assumed to grow at an average growth rate of 13.4% from 1998 to 2003, and the average amount of memory in each PC is assumed to increase at an average growth rate of 39.8%, the simulation shows 200 mm line DRAM production capacity will become insufficient by 2002. Consequently, new lines will be needed in 2002, which will most likely be built as 300 mm lines, not 200 mm lines.
Given these developments, we project the manufacturing equipment market also will begin to recover, driven by systems that handle 300 mm wafers. We make projections about the worldwide market (on a booking basis) for pre-process manufacturing equipment that can handle 300 mm wafers. According to these projections, this market will begin growing full-scale in 2001, reaching $13.2307 billion ($101 million) in 2002 and $16.5926 billion ($126.76 million) in 2003. These 300 mm compatible systems will become a powerful driving force, again vitalizing the Japanese market for manufacturing equipment in 2000 and beyond, returning the market to a full-scale growth path. We project this market will grow by 16.8% from the previous year to 716.2 billion yen ($5.47 billion) in 2000; by another 23.1%, to 881.9 billion yen, ($6.737 billion) in 2001; and by 26.4%, to reach 1.1147 trillion yen ($8.515), in 2002.
However, this pace of growth is slower than the markets in other regions. As
a result Japan's share of the worldwide market will continue falling year after
year, down to 22.8% in 2002.
