In this battle, the Japanese lost cleanly, occupying nearly 80% of the global DRAM (commonly known as computer memory) share from the peak period, and now it has fallen to zero. This chip war perfectly explains what international political economy is. Adam Smith's theory of free market competition is just a beautiful fairy tale in the PK of big country industry.
The first five years of1980s were a glorious period for Japanese semiconductor chip companies.
Technology startups such as Intel and AMD in Silicon Valley were chased by the Japanese in the field of semiconductor storage, and then they were overtaken and driven off the throne. The field of semiconductor chips (mainly semiconductor storage at that time) became the back garden of Japanese enterprises.
American technology companies failed in this model.
The development model of Silicon Valley is to inject capital into startups through venture capital. After the startup company receives financial support, it will carry out continuous technological innovation to gain the market, improve the company's valuation, and let it go public later. Venture capital will sell its shares and withdraw at a profit. This model is market-oriented and efficient, but it is small in size and difficult to integrate resources between companies. After all, everyone is a competitor who grabs food in the pot.
The Japanese play a completely different game: concentrate on doing big things. 1974, the Japanese government approved the "very large scale integrated circuit (commonly known as semiconductor chip)" plan, and set the goal of catching up with and surpassing American integrated circuit technology. Subsequently, Japan's Ministry of International Trade and Industry organized five companies, including Hitachi, NEC, Fujitsu, Mitsubishi and Toshiba, to demand the integration of semiconductor talent resources in Industry-University-Research, Japan, to break down enterprise barriers, and to let enterprises collaborate to tackle key problems and improve the technical level of Japanese semiconductor chips.
Japan's plan almost died, enterprises were wary of each other and destroyed each other, and the funds promised by the government were not in place. At the critical moment, Yasuo Mitsui, the founder of Japanese semiconductor research, stood up and used his prestige to knead the participants with different minds together.
Yasuo Shaft's statement is simple and clear: Only Qi Xin can make concerted efforts to change the backward situation of Japanese chip basic technology. After the research results come out, each company will conduct product research and development independently. Only in this way can we reverse the dilemma of Japanese enterprises going it alone in international competition.
Four years after the implementation of the plan, Japan has obtained thousands of patents, which has suddenly narrowed the technological gap with the United States. Then, the Japanese government introduced loans, tax incentives and other measures, Hitachi, NEC, Fujitsu and other enterprises once expanded Mazhuang, with sufficient ammunition.
Modern semiconductor memory chip manufacturing plants have mushroomed in Japan. With the production line running day and night, the Japanese launched a saturation attack.
The nightmare of Americans began. 1980, Japan occupied 30% of the semiconductor memory market. Five years later, Japan's share exceeded 50%, and the United States was left behind.
High-tech companies in Silicon Valley can't stand the market share plummeting, and it makes people feel desperate to send people across the Pacific Ocean to Japan for reconnaissance. Andy Grove, then Intel's production director, said gloomily, "People who came back from Japan described the situation as very serious." If Grove visits Japan, he will also feel scared: a Japanese company uses a whole building to develop memory chips, and employees on the first floor develop the capacity of 16KB, employees on the second floor develop the capacity of 64KB, and employees on the third floor develop the capacity of 256KB. The Japanese R&D rhythm is simply the legendary three arrows, which makes Silicon Valley enterprises accustomed to playing with one hand helpless.
What makes Americans feel suffocated is that Japanese memory chips are not only large in quantity, but also of good quality. In the 1980' s, Semiconductor Association of America conducted quality tests on memory chips in the United States and Japan, hoping to find out the weaknesses of its competitors. It turns out that the highest quality memory stick in America is worse than the worst quality memory stick in Japan.
Moreover, the Japanese also clap their chests to assure customers that the quality of Japanese memory chips is guaranteed for 25 years!
Under the aggressive attack of Japan, American chip companies were defeated like a mountain, and financial data were like melting ice cream.
198 1 year, AMD's net profit decreased by two thirds, and the national semiconductor lost 1 1 ten thousand dollars. Last year, it earned $52 million. The following year, Intel was forced to lay off 2000 people. The Japanese continued to expand their achievements, and the Americans continued to cry everywhere. 1985, Intel surrendered and announced its withdrawal from DRAM storage business. This war made it lose $654.38+73 million, the first loss since listing. At the most critical moment of Intel, if IBM didn't come to the rescue and buy its 12% bonds to ensure cash flow, the chip giant would probably go bankrupt or be acquired, and the history of the American information industry might be rewritten.
Robert noyce, the founder of Intel, lamented that the United States had entered the process of "imperial decline". He asserted that if this situation continues, Silicon Valley will be in ruins.
What Americans can't stand is that Fujitsu intends to acquire 80% of Fairchild Semiconductor. Fairchild Semiconductor is a living fossil in Silicon Valley, because the founders of most technology companies in Silicon Valley (including Intel and AMD) used to be employees of Fairchild Semiconductor. In the minds of Silicon Valley people, Fairchild exists as a semiconductor god, but now the Japanese want to buy their "god". Isn't that shameful? An American newspaper wrote in the report: "This transaction tells us through a message that we are already very backward, and what is important is how we should deal with it."
A few years ago, technology companies in Silicon Valley set up the Semiconductor Industry Association (SIA) to deal with Japanese attacks. After several years of lobbying, the results are as follows: the capital income tax rate will be reduced from 49% to 28%, and the pension will be promoted into the field of venture capital. The government is unwilling to help.
By June 1985, SIA finally concocted an idea that upset Washington and turned the situation around in one fell swoop.
SIA's view is that the weakening of the semiconductor industry in the United States will bring great risks to national security.
Isn't Japan an ally of America? The rise of Japanese semiconductors and the decline of American semiconductors look like a game with the left pocket turned upside down. How can it threaten the national security of the United States?
The logical chain of SIA is as follows:
Previously, SIA lobbied for seven years, and the government always responded: the United States is a free market, and government power should not be involved in commercial activities.
This time, when SIA's "national security theory" came out, the US government began to understand, with amazing efficiency:
1in the spring of 986, it was discovered that Japan dumped read-only memory; In September, the US-Japan Semiconductor Agreement was signed, requiring Japan to open the semiconductor market and ensure that foreign companies gain 20% market share within five years; Soon, a punitive tariff of 100% will be imposed on the $300 million chips exported from Japan; Reject Fujitsu's acquisition of Fairchild Semiconductor.
This wave of American actions has created at least two records: the first global attack on the economic interests of allies; For the first time, the trade dispute was changed from economics to political economy on the grounds of national security.
Clyde Prestowitz, America's chief trade representative in Asia, who is in charge of negotiations with Japan, criticized Japan's semiconductor chip industry policy as unreasonable and admired it at the same time. "So I told the US government that we should adopt the same policies and measures as Japan."
Regarding this double standard, Tang Shanglong, who has been researching and developing in Hitachi and Elpida for many years, said angrily in the book: "This man is really deceiving others!"
With the signing of the US-Japan semiconductor agreement, the Japanese semiconductor chip industry, which is at the forefront, turned around and slipped into the abyss.
Japan's semiconductor chip industry has been declining from the highest 40% in 1986 to 20 1 1 year/5%, spitting out more than half of the market share, among which DRAM is the hardest hit area, with the global market share falling from the highest nearly 80% to the lowest10%.
It can be said that in the battle with the Americans, the Japanese have basically lost all their accumulated capital, and the whole country has worked hard for eleven years (from 1975 to 1986) and was defeated overnight before liberation.
However, the meat spit out by the Japanese did not fall into the mouth of the Americans, because more than 70% of the technology companies in Silicon Valley cut off the DRAM business (including Intel and AMD). After 1986, the American market share curve is a dead earthworm lying on the horizon, which has been around 20%.
Then, whose stomach has this huge market of 70% entered?
The answer is Korea.
Around 1986, Japan was beaten out of the water by the United States, and South Korea's DRAM took the opportunity to start, but like a toddler, it had no presence in the global semiconductor chip industry. Moreover, compared with Japan, Korean semiconductor chip companies represented by Samsung are completely 360-degree vegetable chickens with no dead ends: they can't enter the high-end market dominated by the Japanese at all, and they can only make a living at low prices in the low-end market; In terms of market volume, the two are the differences between ants and elephants.
However, knowing that all trade frictions belong to the political and economic category, Samsung took the opportunity to fuck the elephant in Japan.
/kloc-in the 1990s, Samsung faced an anti-dumping lawsuit initiated by the United States, but its head, Lee Kin-hee, skillfully took advantage of the opportunity of Americans to suppress the Japanese semiconductor industry and sent a strong public relations team to lobby the Clinton administration: "If Samsung can't make chips normally, the trend of Japanese companies occupying the market will be more obvious, and the reduction of competitors will further raise the price of chips purchased by American companies, which will be even more unfavorable to American companies."
As a result, Americans only charged Samsung 0.74% anti-dumping duty, while Japan was charged the highest anti-dumping duty of 100%. This method of operation is simply too lazy to pretend.
Samsung's embrace of America's thigh is tantamount to giving Japan a knife from behind and making Japan completely out.
If there is no Samsung mending the knife, Japanese semiconductor chips still have hope to get out of the predicament.
Americans bound the Japanese with the US-Japan semiconductor agreement and beat them up with anti-dumping sticks, but the Japanese semiconductor memory chip industry only suffered a flesh wound, because more than 70% of the enterprises in Silicon Valley have withdrawn from the semiconductor memory chip industry, and the market is still firmly in the hands of the Japanese. Those who survived were a group of Japanese heroes. After all, Japan is still an irreplaceable force in the global semiconductor chip industry chain.
After Samsung joined the war and took the initiative to join the United States, the irreplaceable Japanese suddenly became dispensable, and Koreans became the new favorites. Subsequently, Samsung's DRAM "two-way data gating scheme" was recognized by the American Semiconductor Standardization Committee as a memory matching with the microprocessor, while Japan was excluded. In this way, Samsung successfully caught the microprocessor-driven personal computer era express, ahead of Japanese companies.
From the DRAM share chart above, it can be found that Japan's share has fallen off a cliff, while South Korea's share has shown a steep upward curve. The future of Japanese semiconductor chips is cut off as soon as a pair of huge scissors are formed on the upper and lower lines.
Since then, even if the Japanese government has intensively introduced policies to support the semiconductor industry and invested a lot of money, it is powerless to return to heaven. The fate of Japanese semiconductor chips has been set.
Until today, there are still views that the rise of Korean semiconductor chips and the decline of Japanese semiconductor chips are the result of industrial transfer. This is not accurate, because the industrial transfer is to transfer the production line/factory from a high labor cost area to a low labor cost area. Japanese semiconductor chip companies have not moved their production lines to South Korea, but have been directly replaced. Americans actually joined hands with South Korea to reorganize the global semiconductor industry supply chain, erasing the Japanese from the supply chain and making a seemingly indispensable industrial force in the world disappear completely.
Looking at the chip war between Japan and the United States, whether to master the ability to reorganize the global industrial chain is the key to win the trade war, and market share does not constitute the main strength factor, which is also one of the key reasons why Japan lost the chip war.
Main reference materials:
Lost manufacturing industry: the failure of Japanese manufacturing industry. Author: Shang Shanglong;
The rise and fall of Japanese electronics industry, by Yoshio Nishimura;
"Core", by Joseph Xie;
A Hundred Years of Silicon Valley by Allen Rao and pierrot Garuffi.