專利對企業生產效率的貢獻 ─ 以中國產業為例

Abstract

本文回顧文獻對於企業知識資本累積影響生產效率的見解，探討知識資本存量衡量，與企業生產效率評價。並以 Olley and Pakes (1996) 的三階段半參數估計法修正過去對於知識資本影響生產效率評估時未考慮生產力估計會面臨的聯立性和選擇偏誤問題，並結合知識資本生產函數的假設，估計生產力與知識資本累積的影響效果。 利用2001年至2007年中國工業企業數據庫交通運輸設備製造業、電器機械製造業與資訊與電信設備製造業的資料結合中國國家知識產權局專利註冊紀錄，得出專利註冊權的持有一定程度反映企業的知識資本累積成果，且對於全要素生產力有正向且統計顯著的影響，但其影響隨著時間經過而衰退，且衰退的速度隨產業而不同。 專利註冊權也不僅僅代表企業知識資本累積的成果，知識產權作為生產要素投入的一種，也同時是企業排除競爭對手和創造市場障礙的手段，這個結果與我們對於產業內專利註冊權累積的估計結果相吻合。 根據我們的實證結果，未來政府若要從事產業創新活動的獎勵應避免單純增加市場障礙，且無益於企業知識資本累積的政策，改善市場環境和增加競爭程度仍是增進企業生產效率最重要的因素。

This study aims to explore the relationship between the patent activity and the production efficiency of the manufacturing firms in China industries. After reviewing the main concepts of measuring production efficiency and knowledge capital, I use a 3-stage semiparametric estimator developed by Olley and Pakes (1996) to solve selection bias and endogeneity problems in order to derive the actual productivity at firm level. In addition, combining the data of patent applications documented by SIPO with the financial information of the enterprises in three different industries documented by NBS, I form a seven year long panel to regress each firm’s productivity on its patent stock and several control variables. The estimation result shows that the number of patents has positive but moderate impact on the firm’s total factor productivity. This positive effect becomes weaker as time goes by, which implies the patents attained earlier become outdated; however, the decaying process of the effect differs across the industries. Furthermore, the patent property, as an indicator of innovation activities, not only reflects firm’s level of knowledge capital but also creates legal barriers obstructing opponents to use patented technologies. As a result, a high level of patent granted at industry level leads to a lower productivity. In conclusion, the study suggests that the government set up the patent law carefully and ensure adequate competition to boost productivity at industrial levels.