Memory: Shortage of supply and demand will continue

        Competition in DRAM industry is fierce, and manufacturers with poor foundation have withdrawn from this field. The production capacity of enterprises still producing DRAM has not increased for nearly four years, but the market demand is increasing, which leads to serious shortage of products.

        Memory is currently divided into two categories: DRAM and flash, mainly NAND. DRAM mainly cooperates with CPU for caching, and NAND mainly stores code and data. Early mobile phones used NOR flash memory to store code. Later, the mobile software system was too large. The price advantage of NAND was obvious. NOR almost disappeared after 2011, and it was still used in a small amount in some of the lowest mobile phones. DRAM can also be simply divided into mobile DRAM for mobile phones, DDR3 DRAM for desktop and laptop computers.

         DRAM is a key component for all digital electronic products. The market scale in 2012 is about 26 billion US dollars. Samsung has 42% market share; Sk-Hynix has 25% market share; Korean manufacturers have 67% market share; Micron has 24% market share, and Taiwan manufacturers have 9% market share. Japan originally had Elpida, but it was bought by Micron. Because of the fierce competition in DRAM industry, manufacturers with poor foundations have withdrawn from this field. The production capacity of enterprises still producing DRAM has not increased for nearly four years, while the market demand is increasing, which leads to serious shortage of products. Since the second half of last year, memory prices have been rising all the time, but in June, when the electronics industry is in the off-season, it’s too difficult to buy it.

        In June, 2Gb DDR3 hit its highest price in a year, reaching $1.81. Although Taiwan's DRAM output accounts for only about 8% of the world's total, the market is hype-loving. There are four major DRAM manufacturers in Taiwan, namely Regin, Winbond, Nanco and   Inotera, which all have fragmentation problems. The supply of DRAM in the coming months may be reduced by 2%-4%.

        The mobile DRAM market is small, about $12 billion, but it is growing fast because smartphones are increasing, requiring about $10 per phone on average. Samsung has 56% market share in this area; Sk-Hynix in Korea has 22% market share; Micron in the United States has 21%, and the market share of the top three manufacturers is about 99%. The three major manufacturers are in a monopoly position. Basically, they are unwilling to expand production, they want to maintain the situation of short supply and demand and raise prices. Many people say that with Apple's strength and supply chain management capabilities, there will be no supply problems for its key components, but in some areas, even Apple, which has hundreds of billions of dollars in cash, can't do anything about it. For example, core chips such as A6X are only provided by Samsung, which is extremely risky. But this problem can't be solved. If Apple invests in hardware manufacturing, it will take at least five years of exploration and 10 years of technology reserve, which is totally unacceptable for Apple, which has always made fast money.

        The NAND flash market is about $22 billion. The 16 GB version of the iPad and the 32 GB version of the iPad are often seen in the market, which is the capacity of NAND. Recently, many mobile phone manufacturers cannot get eMMC-type NAND, affecting shipments. The reason is simple: Samsung has reduced its external supply in order to ensure the supply of its mobile phones. The price of NAND flash memory has been rising since the end of June 2012, up about 30%. The reason is similar to that of DRAM. For many years, it has relied on technology and scale to increase capacity. However, the space of technology is very small, manufacturers are unwilling to expand production, preferring to enjoy the situation of short supply. Samsung's market share is 37%; Toshiba's 29%, Micron's 14%; Hynix's 12%; Intel's 8%; and the top three account for 80% of the market.

        Passive components: Japanese manufacturers have an advantage

        In the MLCC market, the market share of Japanese manufacturers is about 55%. They will also grow well in 2013, and the growth of Japanese manufacturers means that other manufacturers will shrink.

        Passive components are mainly divided into resistors, capacitors, inductors and magnetic cores. In 2013, the annual output value of passive components industry was about 21.7 billion US dollars, of which capacitors accounted for 66%, about 14.4 billion US dollars; resistors about 2 billion US dollars; inductors about 3 billion US dollars; and magnetic components about 1.2 billion US dollars.

        The passive component industry is mainly distributed in Japan, the United States, China and Southeast Asia. Japan has an absolute advantage, accounting for 52% of the market share. With the sharp devaluation of the yen, Japan's share will further increase. Manufacturers in mainland China account for only about 7% of the total, and most of them are low-end products. Because of the highly automated production of chip-type passive components and the small area occupied by factories, most Japanese enterprises produce the vast majority of products in Japan, only a small number of low-end products are produced in China and Southeast Asia, and most products in the United States are produced in China, so there is no possibility of outward transfer.

        Capacitance can be divided into ceramic capacitor, aluminium electrolytic capacitor, solid state capacitor, tantalum capacitor and thin film capacitor. Among them, MLCC has the largest shipment volume and the highest output value, accounting for 62% of the capacitor market. On average, about 300 to 400 MLCC are required for each mobile phone, 400 to 600 MLCC for each laptop and 300 to 400 MLCC for each tablet. With the increasing capacity of MLCC, the market of some aluminium electrolytic capacitors has been seized by MLCC. MLCC is the most growing capacitor, and the aluminium electrolytic capacitor market will gradually shrink in the future. Its market is mainly industrial equipment, white household appliances, television and PC motherboard. Among them, the white household appliances market is basically flat; the TV market is also showing a slight decline; the desktop market is gradually shrinking, and the traditional notebook computer market is also shrinking sharply. The market for the remaining capacitors is relatively stable.

        In the MLCC market, Japanese manufacturers have a market share of about 55%. In 2013, Samsung's SEMCO will surpass TDK to become the second largest MLCC manufacturer in the world. MURATA has the largest production capacity and the best technology, and it has the dominant position of MLCC. If calculated in yen, TDK and TAIYO YUDEN will also grow well in 2013, while the growth of Japanese manufacturers means the shrinkage of other manufacturers. Japanese manufacturers are of excellent quality, while the only disadvantage is their high prices, but the depreciation of the yen lowers their prices. The total market share of YAGEO, WTC and HOLYSTONE is about 9%. Enterprises in mainland China are mainly Fenghua and Yuyang s, with a total market share of less than 2%, about 1.5%.

        Many production bases in mainland China are often just an assembly workshop, and purchasing is controlled by foreign enterprises and joint ventures, while more than 95% of the upstream key components are controlled by foreign enterprises and joint ventures. Among the top 100 exporting high-tech enterprises announced by the Mainland Customs in 2012, there are only 5 enterprises in mainland China, and 46 enterprises in Taiwan, excluding 18 logistics and investment enterprises.