從台灣DRAM產業競爭優勢觀點探討西進大陸之可行性

Abstract

在全球半導體產業逐漸轉移至亞洲的大趨勢下，中國已經成為全球最大的新興半導體市場和低成本的生產基地，國際半導體大廠向中國投資以轉移生產製造基地的趨勢更愈難以阻擋。目前全球前十大半導體大廠，幾乎均已在中國進行投資佈局，群聚效應顯現。全球半導體大廠對中國大陸投資所採取的普遍策略是利用中國廉價的土地及勞工，以獨資或合資方式達到降低生產成本的目的。再加上最近幾年，大陸政府政策性大力扶植半導體產業，全力營造出適合發展半導體產業之大環境。

由於台灣半導體因為受政治的影響，昔日IC製造業只有晶圓代工業「名正言順」在大陸設廠;另一方面，於2006年底DRAM業者茂德、力晶終於亦被核准。雖考量全球資訊電子產業發展生態和企業全球化佈局，企業赴大陸投資不但是拓展其經營市場，亦是國家經濟實力之延伸。但台灣DRAM業者是否必定要到大陸地區進行投資呢? 實在有需要進一步深入研究之必要。而本研究之動機即希望透過分析台灣DRAM產業的競爭優勢，來對台灣DRAM業者作為投資大陸的策略方向與建議，以繼續維持台灣DRAM產業的國際競爭力與確保其競爭優勢。

本研究主要是藉由資料的蒐集與整理以瞭解產業特性、現況及發展趨勢，並以採用五力分析、SWOT 分析及鑽石模型分析為產業分析的架構，歸納出台灣DRAM 製造廠商所擁有的競爭優勢分別為:「優異的生產管理能力」、「領先的製程技術」、「優異的製造成本控制能力」、「完整的支援產業」、「籌措擴廠資金不虞匱乏」、「充沛而質優的人力資源」、「政府政策性的支持」、「上下游垂直整合」及「產品多角化經營」。

整體來說，台灣的DRAM產業不論成本控制、製程技術及生產管理都較中國優良，在人力的專業素養上也較中國優秀許多。故以台灣DRAM廠商所擁有之競爭優勢，台灣DRAM業者是可以到大陸地區進行投資，但在政府從過去的「戒急用忍」政策調整為現在的「積極管理、有效開放」下，仍然只能有條件、低度的開放，為的是台灣整體經濟利益的考量。

另一方面，並針對台灣DRAM產業對大陸投資作分析，以「DRAM 成本結構」、「接近市場、先行卡位」、「先進製程技術」、「12吋晶圓廠與8吋晶圓廠成本效益」及「8吋晶圓廠之出路」五項因素來考量。發現雖然國內DRAM廠商雖有赴大陸設廠的需求存在，但西移大陸未必能大幅提升自身競爭優勢，因此本研究不認為現階段台灣DRAM 製造廠商有到大陸地區進行投資8吋晶圓製造之急迫性和必要性。另提出對台灣DRAM業者投資大陸的策略方向與建議如下:

1. 如果要到大陸設廠，一定要用最尖端的製程技術，假設現在要考慮在大陸破土蓋廠，二年後進入量產，要考慮的製程是45奈米左右的製程，而且必然是十二吋廠，因為以12吋廠生產DRAM才有經濟效益。如此DRAM廠才有赴大陸設廠的動機及利益存在。如果僅開放8吋晶圓廠、0.18微米以上製程登陸，對台灣DRAM製造廠商而言已不具競爭力。

2. 如果「西進是讓8吋晶圓舊產能找到出路」。那麼適度的大陸佈局，在大陸尋得一席之位，可增加伺機而動之彈性。如果能夠將8吋晶圓廠遷移至大陸，讓這一些已經不適合生產DRAM的舊產能，能夠找到出路，並且將現址8吋晶圓廠，就地提升成為更先進的12吋晶圓廠，以達到產業升級的目的。

Under the major tendency that global semiconductor industry transfers to Asia gradually, China has already become the biggest global semiconductor market as well as the lowest cost of production base. It’s hard to prevent major international semiconductor factories transferring the manufacturing base by investing on China. Currently, by taking the advantage of low cost in land and labor, most of global top ten semiconductor factories have already carried out their investment in China for achieving low cost production by way of individual proprietorship or joint venture. For recent years, government in Mainland China makes every effort to strongly support semiconductor industry with development of macro environment.

Due to semiconductor industry in Taiwan is subject to development of political impact, within the IC manufacturing industry, the foundry industry is the only business which is allowed to be legally built in Mainland China in the past. DRAM manufacturing industry is not approved until late of 2006. Considering of developmental ecology in information technology and enterprise globalization, it not only implies enterprises’ intention to promote business in Mainland China, but also represents enterprises’ expansion of economic strength in the market. Is it necessary for DRAM Manufacturers in Taiwan to invest in Mainland China? We need to thoroughly study such intention in details. The purpose of this research is hoping to analyze competitive advantage of DRAM industry in Taiwan so as to sustain the international competitive ability of Taiwan’s DRAM industry and secure its competitive advantage.

This research is organized by data collection for understanding of industry characteristics, current situation and trend of development. By using of five- strength analysis, SWOT analysis and the diamond model analysis as the industry analytical structure, this research induces the competitive advantage that the DRAM manufacturing in Taiwan has adapted for: "the excellent production management ability", "the leading manufacturing process technique ", "the outstanding manufacturing cost control ability", "the integrity of support industry", "the easy collection of a factory expansion funds", "the abundant and superior human resources", "the government policy-type supporting", "the upstream and downstream industry with vertical integration conformity" and "the product diversification operations".

Overall, in spite of the manufacturing cost control, the process technique and production management of DRAM industry in Taiwan is better than China. With the competitive advantage of Taiwan’s DRAM manufacturer, they may continue to invest in Mainland China. Government policy has been already adjusted slightly, but it is still insufficient if we want to integrate benefit for entire economic development in Taiwan.

Face to the fact of investing in Mainland China, analysis of such trend is made according to five perspectives of “cost structure of DRAM”, "time to market”, “advanced technique of manufacturing process", "cost effectiveness of 8-inch and 12-inch wafer factories", and "future of 8-inch wafer factory". Although the domestic manufacturers of DRAM need to invest in Mainland China for whatever reasons, moving toward Mainland China won’t significantly promote competitive advantage as desired. This research won’t support that DRAM manufacturers of Taiwan should urgently invest on 8-inch wafer manufacturing in Mainland China. Instead, it is proposed that DRAM manufacturers of Taiwan should set their direction of strategy as followings:

1. If DRAM manufacturers of Taiwan intend to continue the investment in Mainland China, they must utilize the most state-of-art manufacturing process technique. For example, if two years of project planning is required for entry of Mainland China, both of manufacturing process for 45 Nanometer and establishing a 12-inch factory should be considered simultaneously for reason of scale of economy. This is to enhance benefit and keep momentum so that factories are attracted to invest in Mainland China. If the government can only allow 8-inch wafer factories with 0.18 micrometer manufacturing process to be established in Mainland China, it will minimize competitive advantage of DRAM manufacturers of Taiwan.

2. If investing 8-inch wafer factories in Mainland China would help the DRAM industry find the way out, manufacturers should also plan for a feasible location in Mainland China for installing such 8-inch facility properly. When 8-inch wafer factory moves out, the 12-inch wafer facility with more advanced technique would move in for upgrading in the industry.