廠商規模、市場結構對研發活動之影響—熊彼得假說再檢驗

Abstract

過去針對熊彼得假說(亦即廠商規模與產業集中度對研發活動具有正面影響)進行實證研究之文獻，並未獲得一致的結果。本文使用中國2000∼2007 年製造業廠商資料，探討影響廠商研發活動的因素，特別針對熊彼得假說重新進行檢驗，以了解該假說是否適用於中國製造商。 由於廠商之研發意願可能影響其研發密度的選擇，因此不同於過去文獻之分別探討影響廠商研發意願與研發密度的做法，本文將廠商研發活動分成兩個階段，第一階段選擇是否投入研發(亦即研發意願)，在決定投入研發後，第二階段選擇研發度，並採取Heckman selection model 做為實證模型，藉以解決由於第一階段決策影響第二階段決策所產生的樣本選擇偏誤問題。本文實證結果顯示廠商研發意願的部份支持熊彼得假說，即廠商規模、產業集中度對研發意願有正向影響；但廠商研發密度的部份則發現廠商規模、產業集中度對研發密度的影響存在非線性關係，顯示熊彼得假說並不成立。此外，本文亦將樣本分為3 種不同規模進行分析，仍發現無論何種規模之下，廠商研發意願的部份皆支持熊彼得假說，但廠商研發密度的部份則不支持熊彼得假說。至於其他影響研發活動之因素也會隨3 種不同規模的樣本而對研發活動產生不同的影響。

The Schumpeterian hypothesis that large firm size and market power stimulate innovation has been tested by numerous studies. However, there is little evidence in support of the hypothesis, and the consensus regarding the relationship between firm size (market power) and R&D has also not been reached yet. By using the large Chinese data set from the survey on Industrial Enterprise Statistics for the period from 2000 to 2007, this paper reexamines the Schumpeterian hypothesis.This paper treats a firm’s R&D activity as a two-stage decision, where whether or not to engage in R&D (R&D propensity) is determined in the first stage and R&D intensity is then determined in the second stage by those who decide to undertake R&D.As a firm’s R&D intensity may be significantly affected by its R&D propensity, the Heckman selection model is used to correct for the self-selection problem that may arise from ignoring the relationship between the two stage decisions.Our empirical results show that the Schumpeter hypothesis is valid for a firm’s decision on whether or not to undertake R&D, but it is not for the R&D intensity. Specifically, the paper shows that firm size and market power (measured by CR4) have positive effects on R&D propensity, but their effects on R&D intensity are no longer linear and display U shape for firm size and inverted-U shape for market power .