| In this paper, we generated a map of industrial clusters and established the Locational Quotients as well as Locational Gini Coefficients for the manufacturing sector in Taiwan. From the quantitative results, we found evidence to support our hypotheses on the effects of industrial clustering to the overall industrial performances. The results conclude that highly clustered industries in average perform better than less clustered industries in term of productivity; they also show tendency to participate in higher level of export activities. We found slightly increasing return to scale from both aggregate and disaggregate industrial production data (at both SIC-3 & SIC 4-digit levels). Furthermore, industrial productivity growths are evident during the process of agglomerating, that is, cluster building in part explains the reason for productivity growth. Contrary to common intuition, we demonstrated a negative relationship between firms' profitability and the level of industrial clustering. Our hypothesis that the profit margin "thinning" due to intensive competition among firms in highly clustered industries is supported by the quantitative analysis. Which implies the existence of strategy convergence and product homogenization among the industry. Lastly, we tested the effect of clustering on the growth of firm number in an industry and found evidence to supports our hypothesis that cluster stimulates entrepreneurship. |
