創新與研發、知識外溢之因果分析--以G20國家集團為例

Abstract

本研究以Griliches（1979）所發展之知識生產函數為基礎，並採用G20國家集團1997－2008年追蹤資料為樣本，分析知識產出與知識外溢的關聯性。並應用晚近Krugman（1991）所發展之新經濟地理學架構，以空間誤差模型估計知識產出與研發投入、人力品質、政府政策、產業聚集之因果關係。實證結果顯示，當G20各國增加研發經費、每人平均收入、研發人員數量與產業聚落之際，各國知識產出將產生顯著正面之直接影響。本文另一重點為知識外溢效果，當鄰近國家增加研發經費、高等教育入學率、高等教育經費支出，對G20集團國內知識產出產生正面影響。值得注意的是知識外溢所產生的負面效果，當鄰近國家大量延攬研發人才與他國產業聚落興起之際，將導致G20國家知識產出之下降。

Based on the Knowledge Production Function developed by Griliches (1979), this research analyzes the relationships between knowledge production and knowledge spillovers for open economies. The empirical study applies G20 countries panel data for the period of 1997 through 2008. Also incorporated in the theoretical framework is Krugman's New Economic Geography theory (1991), which employs a spatial error model to estimate the causal relationships between R&D investment, human qualities, government policies, industrial agglomeration, and knowledge production. The results of this study show that R&D funding, per capita income, the number of researchers, and industrial agglomerations have positive direct effects on domestic knowledge production for G20 countries. Another important finding is the spillover effect from the neighboring countries: R&D funding, education enrollment, and education expenditures from neighboring countries have a positive impact on domestic knowledge production for G20 countries. However, when a large number of researchers are recruited by neighboring countries and industrial agglomeration in neighboring countries rises, a significant negative impact on knowledge production is observed.