本文將薪資報酬分為固定與非固定報酬二部份，以探討電子製造業與傳統製造業之薪資報酬結構差異。由Heckman(1979)兩階段估計之實證結果顯示，全體製造業在固定上，除了文獻曾提及的性別、教育程度與工作年資等人力資本變數，以及工時、工作地點等工作特徵因素皆具有影響力之外，退休福利制度所反映的利潤分享傾向亦具顯著影響效果，電子製造業與傳統製造業受僱者在固定薪資結構上並未呈現明顯差異。在非固定薪資報酬結構上，電子製造業受僱者之報酬結構和傳統製造業則有明顯差異，顯著影響傳統製造業受僱者非固定薪資報酬的因素包括性別、婚姻狀況、自評健康狀況與工作年資等人力資本變數，職等、廠商規模、工作地點與同事人際關係等工作特徵變數，以及退休福利制度之利潤分享變數，而這些變數中除了工作年資變數外，其餘變數對電子製造業受僱者皆未呈現顯著影響結果。最後，由Oaxaca(1973)薪資差異分解結果發現，兩產業薪資估計值在每月固定薪資上並沒有顯著的差異出現，但在非固定薪資報酬上其差異較為明顯。其中價格差異的影響幅度較特徵差異為大。

The high-tech manufacture industry in Taiwan has expanded rapidly during the past few decades. As a result, more and more workers are attracted into the high-tech industry due to the promising future of this industry. Do high-tech industry employees receive better compensation? This study examines empirically the difference in the compensation structure for employees of high-tech manufacture industry and traditional manufacture industry by decomposing the total compensation into the fixed salary and non-fixed compensation. The influence of three different sets of variables, namely, human capital variables, job characteristic variables, and profit sharing variables, is analyzed in this study. The empirical results based on the Heckman's two stage procedure indicate that human capital variables (gender, education and tenure), job characteristic variables (work hour and work place), as well as profit sharing variable (pension system) are significant determinants of the fixed salary structure for employees in both the high-tech and the traditional manufacture industries. However, as to the non-fixed compensation structure, significant determinants for the employees of high-tech manufacture industry are different from those for the employees of the traditional manufacture industry. Human capital variables (gender, marital status, self-reported health status and tenure), job characteristic variables (firm size, work place and work peer relationship), as well as profit sharing variable (pension system) are significant determinants of the non-fixed compensation for employees in traditional manufacture industry while only tenure variable shows significant influence on the non-fixed compensation for employees in high-tech manufacture industry. Moreover, job satisfaction variable is significant in determining the non-fixed compensation for employees in the high-tech manufacture industry. But it shows no significant influence on the non-fixed compensation for employees in the traditional manufacture industry. The Oaxaca decomposition results further imply that there is no significant difference in the fixed salary structure between the high-tech and traditional manufacture industry. But more apparent difference is shown in the non-fixed compensation structure. The difference can be attributed more to the price differentiation than to the characteristic differentiation.