Industrial agglomeration, geographic innovation and total factor productivity: The case of Taiwan

Abstract

The paper analyses the impact of geographic innovation on total factor productivity (TFP) in Taiwan in 2001 using 242 four-digit standard industrial classification (SIC) industries. We compute TFP by estimating Translog production functions with K, L, E and M inputs, and measure the geographic innovative activity using both Krugman's Gini coefficients and the location Herfindahl index. We also consider the geographic innovation variable as an endogenous variable and use two stage least squares (2SLS) to obtain a consistent, albeit inefficient, estimator. The empirical results show a significantly positive effect of geographic innovation, as well as R&D expenditure, on TFP. These results are robust for the Gini coefficients and location Herfindahl index, when industrial characteristics and heteroskedasticity are controlled. Moreover the Hausman test shows that the geographic innovation variable should be treated as endogenous which supports the modern theory of industrial clustering regarding innovation spillovers within clusters. © 2008 IMACS.