地铁对服务业集聚的影响——以广州市为例

以广州市为例，基于POI数据，运用SAR模型识别城市地铁开通对服务业集聚的影响。研究发现：服务业集聚具有显著的空间依赖性特征，本地服务业集聚会受到周边地区服务业集聚的影响；开通地铁对广州市服务业空间集聚产生显著的正向影响，可提高城区的服务业集聚水平，但这种集聚效应因行业不同而有所差异；地铁开通对批发与零售业、住宿与餐饮业等生活性服务业的影响要高于其他服务业行业，但对科学研究和技术服务业集聚的影响并不显著；开通地铁对服务业集聚的影响还受到所在区域经济发展水平的影响，经济发展水平较高的区域，地铁对服务业集聚的提升作用较明显；而地铁换乘站对服务业集聚的影响要高于非换乘站。最后，建议完善城市地铁网络，加强地铁站点的基础设施配套，提高地铁网络密度和地铁站点间的空间距离以进一步发挥地铁对服务业集聚作用。

The influence of Subways on Service Industry Agglomeration: Taking Guangzhou as an Example

Agglomeration is an important feature of the spatial distribution of an urban internal service industry. Most of the previous studies on the influencing factors of urban internal service industry agglomeration have ignored traffic factors, especially the influence of subways. The improvement of traffic convenience can often affect the location of a service industry by gathering a flow of people. With the rapid growth in the number of China's metro cities, it is necessary to evaluate the impact of subways on the service industry activities within the city. In addition, most of the previous literature has not considered the spatial dependence of service activities, and there is little discussion on the heterogeneity of the service industry. In view of this, based on POI(Point of Information) data in Guangzhou as an example, this study uses a spatial autoregressive (SAR) model to identify the impact of subways on service industry agglomeration and analyzes the heterogeneity of different types of service industry. The results are summarized as follows. 1) Using Moran's I to measure the spatial correlation of Guangzhou's service industry agglomeration, the results indicate strong spatial correlation characteristics. In addition, according to the regression results of the SAR model, the spatial lag coefficient is significantly positive, which indicates that the service industry agglomeration has a considerable spatial dependence. Specifically, the degree of local service industry agglomeration will increase with an improvement in the surrounding areas. 2) The opening of subways has a significant positive impact on the spatial agglomeration of Guangzhou's service industry, which could increase the agglomeration levels in urban areas. One possible reason for this is that subways bring a floating population and reduce transaction costs. 3) There are different industries within the service industry, each of which has varying characteristics; thus, the impact of subways is heterogeneous, specifically as follows: the impact of a subway opening is higher on the wholesale and retail industry, accommodation and catering industry, and other life services than on other service industries but not significant on scientific research and technical services. 4) The impact of the metro on service industry agglomeration is also affected by the level of regional economic development. In areas with a higher level of economic development, the promotion effect of the metro on service industry agglomeration is more obvious. 5) Finally, a metro transfer station has a higher impact than a non-transfer station. One possible reason for this is that a metro transfer station is the intersection of multiple metro lines, which can often attract a greater flow of people and promote the flow of production factors, making service industry agglomeration more likely. The main contributions of this paper are as follows: first, taking Guangzhou, which has a well-developed metro network, as an example, this study evaluates the impact of the metro on service industry agglomeration and analyzes industry heterogeneity. Second, on the basis of constructing a spatial distance matrix of each economic unit, when considering the spatial dependence of service activities, the method of spatial economics was used to investigate the impact of subways on service industry agglomeration; third, taking big data represented by POIs as the source data, and taking streets and towns as the units of analysis, it more objectively and accurately reflects the spatial distribution characteristics of service industry agglomeration in cities.