| Abstract: | 一國的經濟成長主要來自於技術進步。一般來說普遍用來衡量技術進步的指標便是「總要素生產力(Total Factor Productivity)」。因此本研究以台日製造業為研究對象，首先引用Solow模型來假設Cobb-Douglas型生產函數，利用成長會計式估計出台日製造業(1982-2005年)的TFP成長率。其次，對台日製造業的總要素生產力進行分解，藉由迴歸分析來確立適合的TFP複迴歸模式，並分別探討其結果。
經由實證結果發現以下幾點:
第一、台日製造業的實質GDP成長來源為資本投入及TFP成長，其次為勞動供給。對實質GDP成長的貢獻度分別為，(1)台灣製造業:資本投入(56.22%)、TFP成長(28.80%)、勞動供給(14.98%)，(2)日本製造業:TFP成長(63.38%)、資本投入(20.87%)、勞動供給(15.75%)。
第二、台日製造業在1990年後的實質GDP年平均成長率比80年代小。其原因為，勞動供給不足及TFP成長率低下所導致。
第三、將影響台灣及日本製造業的技術進步(Solow殘差)進一步解析，變數放入迴歸模式後發現，術貿易額收支比率(技術輸出／技術輸入)與台灣製造業的TFP成長有正相關，而與日本製造業的TFP成長有負相關。另外，研發費用占營業額比率與研究人員數與日本製造業的TFP成長有正相關。
第四、台灣製造業乃以技術引進為主，在技術貿易逆差下對TFP呈現正係數,但近年來電子業技術輸出擴大導致技術貿易逆差減少,亦即當國內研發投資對TFP的效率(或係數)不變的前提下,台灣製造業的技術輸出海外比率增加而技術輸入又相對減少時,很可能會使台灣製造業的TFP成長減少。另外，日本製造業若提高研發密集度及培育研究人才將會促進國內製造業TFP成長,相反地當技術輸出大於技術輸入時，有可能使得國內製造業的TFP成長減少。
A country’s economic growth mainly comes from technical progress. Generally speaking, the most common indicator used to measure technical progress is “Total Factor Productivity”. Therefore, the subject of our research is focused on manufacturing industry in Taiwan and Japan.To begin with,we adopt Solow model to assume Cobb-Douglas production function and use Growth Accounting Approach to estimate TFP growth rate of the manufacturing industry in Taiwan and Japan (1982-2005). Next, we analyze the TFP of both countries’ manufacturing industry. By using regression analysis, we can establish the suitable TFP multiple regression model and discuss the results respectively.
According to the empirical finding, the following points can be concluded:
First, the real GDP growth of manufacturing industry in Taiwan and Japan comes from capital input, TFP growth and labor input. The degrees of contribution to real GDP growth are listed individually: (1) Taiwan: capital input(56.22%)、TFP growth (28.8%)、labor input(14.98%)(2) Japan: TFP growth(63.38%)、capital input(20.87%)、labor input(15.75%).
Second, the annual mean growth rate of real GDP of manufacturing industry in Taiwan and Japan after 1990s is smaller than that of 1980s. It is resulted from the insufficient labor input and low TFP growth rate.
Third, we further analyze the technical progress (residual) affecting the manufacturing industry in Taiwan and Japan and apply the regression analysis to all possible variables of TFP. According to the statistic result, we find that there is a positive correlation between the technology trade balance of Taiwan and TFP growth, and a negative correlation between the technology trade balance of Japan and TFP growth. In addition, the ratio of R&D expenditure to turnover ratio and researchers are positively correlated with TFP growth of Japanese manufacturing industry.
Fourth, Taiwanese manufacturing industry is mainly
technology import. The technology trade balance have positive on TFP under technology trade deficit. But in recent years, electronic industries have increased the export of technology to reduce trade deficit. In another word, under the environment which efficiency (or coefficient) of domestic R&D to TFP is constant and if technology export increases and technology import decreases, it will possibly reduce TFP growth of manufacturing industry inTaiwan. Moreover, if R&D intensity and promote the quality of researchers are enhanced, it will raise TFP growth in Japan. On the contrary, if the technology output is more than the technology input, it will possibility reduce TFP growth. |
