臺灣博碩士論文加值系統

本研究試圖提出一個整合性的架構，從交易成本、社會關係、企業網絡與統理結構等觀念出發，藉以探討企業經營者如何透過產業網絡降低交易成本，與網絡組成份子彼此間維持著競爭又合作的關係，進而追求共同的利潤，本研究所選擇的個案公司為世界第一大晶圓代工廠－台灣積體電路公司。 十年前台積電初成立時，以「專業代工，不和客戶競爭」的策略，此外配合客戶需求，強調彈性，不斷提升台積電的製程技術、人力素質、良率，使台積電開工兩年，就達到20億台幣的營業額。台積電成立十年，成長為近13億美元的企業，不但破台灣企業紀錄，更超越美國高科技公司英代爾、康柏的紀錄，甚至開始影響美國半導體業興榮。目前台積電已是全世界最大的晶圓代工廠，之所以會有今日的榮景，主要是因為當初專業晶圓代工策略定位正確之故。 所謂「專業晶圓代工策略」，即是台積電不設計或生產自有品牌的產品，而提供所有的產能為客戶代工生產。在台積電的經營理念裡，確立了台積電只做代工，不與客戶競爭的永續原則。 本研究試著將Coase與Williamson的交易成本理論與網絡統理理論相結合，提出了CO－WANG整合性架構（即Coase、Williamson, And Network Governance等字的縮寫），解釋了台積電採用專業晶圓代工策略之所以奏效的原因，在於台積電與其客戶之間的網絡關係，因彼此互信而大幅地降低了彼此之間的交易成本，再加上台積電本身因進入專注生產的領域，產生了規模經濟與範疇經濟。 本研究由CO－WANG架構引申出6個命題，本研究所提出的台積電個案分析則是說明了命題1的情境與結果：命題1：當資產具有中度專質性、交易頻率為重覆發生，且網絡效果很強的時候，廠商不會自己進行垂直整合，而會與網絡式組織中的廠商進行雙邊統理的契約購買方式。 這個命題就直接解釋了台積電採行專業晶圓代工策略之所以會成功的原因，因為其主要的客戶－IC設計公司基於彼此之間因交易與長期信賴關係所形成的網絡，降低了交易成本，創造了更大的規模經濟效果。客戶愈信任台積電，網絡效果就愈強，減少的成本就愈多，這又導致顧客更信賴台積電。 在台積電的個案中，我們可以看出其與客戶之間很明顯的存在著雙邊統理的現象，主要就是因為網絡效果的關係，IC設計公司不進行策略聯盟或垂直整合，而將產品只交由專業代工的台積電生產，形成買賣雙方的雙邊統理機制，這也是台積電創造出台灣半導體產業奇蹟的關鍵因素。

The perspectives of economic sociology of observing and analyzing transaction characteristics and governance structures. The micro perspectives of the traditional management theory and the macro perspectives of sociology are integrated in this research. My objective in this research is to provide an integrated structure that explains variations in three critical features of transactions: degree of asset specificity, frequency, and network effect, lead to the emergence of alternate institutions for governing exchange. This research integrated Coase''s and Williamson''s transaction cost theories and social network theories to propose an integrated structure-"CO-WANG". The research used the CO-WANG integrated structure to analyze the pure integrated circuit (IC) foundry strategy of Taiwan Semiconductor Manufacturing Company (TSMC). TSMC was the first pure integrated circuit (IC) foundry company in the world. Since its founding, the company has been dedicated to providing manufacturing services for advanced ICs. The company''s charter prevents TSMC from designing or making its own brandname IC end products. For this reason, TSMC is a partner, and not a competitor, for other semiconductor companies. TSMC''s success in the foundry business has served as a model for the many new entrants in the market. With TSMC as the engine of change, what was once just a concept of a pure foundry is today a multi-billion dollar industry. In the CO-WANG integrated structure, when the transaction frequency is recurrent, the asset specificity is between mixed and idiosyncratic and the network effect is strong, the buyer will select the bilateral governance mechanism. In the case of TSMC, the transaction characteristics between TSMC and her customer are repeat-purchase, medium asset specificity, strong network effect, and the governance structure is bilateral governance. TSMC use the pure IC foundry strategy to get thebuyer''s trust. Trust lowed down the buyer''s transaction cost , so they selected the bilateral governance mechanism with TSMC. For the reason, TSMC will have more orders. When TSMC increased it''s output and produced two or more different products than just one in the long run, economies of scale and economies of scope reduced long run average costs. The low cost will led the trust of TSMC''s customer increased more and more. We can use the CO-WANG integrated structure to explain the pure IC foundry strategy, the other propositions from CO-WANG integrated structure can explain some industrial phenomenon, maybe somebody will research more deeply in the future.