产业地理集中、产业集聚与产业集群:测量与辨识

Geographical Concentration and Agglomeration of Industries:Measurement and Identification

本文综述了产业地理集中、产业地理集聚以及产业集群的测量与辨识方法。传统的测量产业地理集中方法包括集中系数、变差系数、赫芬代尔系数、赫希曼-赫芬代尔系数、信息熵、锡尔系数以及基尼系数等,这些系数测量产业总体地理集中程度,没有考虑企业规模分布对产业地理集中的影响。基于企业区位选择模型,经济学家发展了测量产业地理集聚的指数,控制产业内企业规模分布对产业地理集中的影响。无论地理集中指数还是地理集聚系数都以行政单元为基础,仅描述单一空间尺度上的产业区位模式。Ripley的K函数通过计算某个企业一定距离内的邻居企业个数来测量产业的地理集聚程度,可同时反映产业在不同空间尺度的集聚程度。相互联系的一群企业在地理空间上的集聚构成了产业集群,产业集群的辨识不仅要测量产业间联系,也要考虑产业地理临近性。区位商和标准化区位商法、空间相关与产业联系法、因子分析和聚类分析等多元统计方法以及基于投入产出关系的图谱分析方法等可以用来作为辨识区域产业集群手段。

This paper presents a literature review on the measurement tration and agglomeration of industries. Indices of concentration should meet the following criteria： （1） being comparable across industries and spatial scales; （2） taking a unique （or known） value under the null hypothesis that there is no systematic component to the location of the industry;（3） the significance of the results should be reported where appropriate; and （4） measures should be unbiased with respect to arbitrary changes to the spatial classification and industrial classification. There are many concentration indices, including coefficients of variation, Herfind ahl index, Hoover coefficient, Entropy index, Theil index and Gini coefficient. Those aggregate measures of geographical concentration, however, ignore the impact of plant distributions on ge- ographical concentration of industries. Based on the model of plant locational choices, economists propose agglomeration indices designated to measure the excess of raw geographic concentration on productive concentration. However, both concentration and agglomeration measures only describe the location of industries on a single scale based on administrative regions. Distance-based methods, such as Ripley＇s K function, serve to describe the spatial structure of industries at different scales at the same time. Existing literature also attempts to propose quan- titative methods to identify industrial clusters. I^cation quotients and standardized locational quotients, spatial and industrial linkages, factor analysis and graphic methods based on input output have been applied. The identification of industrial clusters not only requires the consideration of industrial linkage, but also geographical proximity.