以資料包絡法與隨機邊界分析法評估亞洲地區二十大港埠之貨櫃經營效率

Abstract

近年來，根據「國際貨櫃化雜誌」（Containerisation International）的統計，可以發現中國在經濟快速發展以及其經濟地位日漸重要之際，嚴重的壓縮與排擠台灣及鄰近亞洲國家港埠的成長空間，使得台灣與東南亞國家的經濟發展與港埠貨櫃量出現變化，值得注意的是中國大陸港埠的成長幅度，中國港埠已有六個港埠擠進亞太地區前十大，其五個港口的成長率達到10%以上，香港除外。因此，港埠營運績效與港埠的競爭力息息相關，為了進一步衡量、分析港埠經營績效，須選擇適當的效率衡量方法，但港埠事業係屬於營利事業或非營利事業，說法很多，亦可以公營、公民營、民營等不同形式存在。SFA分析以經濟計量方法估算投入與產出間之效率值，常應用於營利組織，而DEA為一種績效評估的方法，透過線性規劃來找出包住所有受評估單位的效率包絡線，可衡量多項投入與產出之相對效率值，常應用於非營利組織，根據港埠事業具有多元投入及產出組合的特性，故本研究在此同時運用資料包絡分析法(DEA)與隨機邊界分析法(SFA)二種不同的方法分析2004-2006年間亞洲地區20大國際貨櫃港的經營績效以及其競爭力的變化。 研究發現：在DEA分析法中採用CCR與BCC模式，2006年之大陸地區的寧波、天津、廈門、上海、香港在亞洲地區二十大港埠中較具有競爭力。SFA法中採用Cobb-Douglas及Translog生產函數兩種模式進行分析，其中兩模式效率最高的港口排名中前三名皆為相同的港埠分別為深圳、上海、香港。模式間相比，可發現平均總體效率值為BCC模式最高，順序依序為順序依序為TEBCC (0.759)＞TECCR (0.684)＞TETR (0.650)＞TEC-D (0.542)。運用Spearman等級相關係數分析使用四種衡量模式中各效率排序間之相關性，顯示效率值排序與SFA兩模式較一致，僅DEABCC與三模式之相關係數相對較低，加入2004-2006年平均貨櫃吞吐量排序與四模式比較，其不存在較明顯的趨勢關係。

According to the statistics of Containerisation International, with the recent fast growing of economics, China greatly competes and challenges Taiwan and other Asia countries’ development of ports. Thus the economic developments of containers ports of Southeast Asian countries have been changed. As a result of the recent growing of China economies, now there are six China ports being in the top ten Asian, with a growing rates of more than 10 percents, excluding Hong Kong. As the port operating performance links closely with port competitiveness, to further measure and analyze the operating efficiency of ports, we have to select a suitable method to evaluate the efficiency. On the methodology, SFA is an econometric approach, and is suited to measure efficiencies of stochastic industry for input/output information. SFA has been applied to measure performance of profit organizations. In contrast, DEA is a non-parametric approach and is suited to measure efficiencies of deterministic industry for multiple inputs/outputs information. DEA has been applied to assess performance of non-profit organizations. Since seaport can be public-owned, private-owned and both public and private-owned, this paper aimed at measure the operating efficiencies and the change of competition of Top Twenty Container Ports in Asia from 2004 to 2006 by applying both SFA and DEA models. From the results of the analysis, with CCR model and BCC model in DEA, Ningbo, Tianjin, Xiamen, Shanghai and Hong Kong have the best performance in 2006. For SFA, Cobb-Douglas function and Translog production function were used to analyze the data, and the result showed that Shenzhen, Shanghai and Hong Kong had the best performances in the two models. Comparision of the four models showed that the average operating efficiency value of BCC model TEBCC was 0.759 which is the highest value, followed by TECCR (0.684), TETR (0.650) and TEC-D (0.542). Adopting Spearman ranking correlation coefficient analysis to measure the efficiency ranking of the four models, we found that the efficiency values were consistent for the two models in SFA. The correlation coefficient of DEABCC was the lowest as compared to the result of the other models. Furthermore, the efficiency ranking of each of the four models does not depend on the average container volumes ranking for the data studied in 2004-2006.