Abstract: Based on the existing research literature, the paper finds that producer services and manufacturing industry co-agglomeration may have both positive and negative effects on TFP, which makes the causal relationship between the two complicated to certain extent. preliminary tests show that the industry co-agglomeration on national average level significantly inhibits TFP without endogenous grouping. Further endogenous grouping empirical test with the panel data of Chinese 246 prefecture-level and above cities finds that the impacts of industrial co-agglomeration on TFP demonstrates nonlinear differences. With the promotion of specialized agglomeration degree of component industries, the impact of producer services and manufacturing industry co-agglomeration on TFP growth tends to change from negative to positive.

Key words: Industrial Agglomeration, Co-agglomeration, Total Factor Productivity(TFP), GMM, Nonlinear Effect

摘要： 综合既有文献进行理论推演,本文发现生产性服务业与制造业协同集聚可能从正反向同时作用于全要素生产率增长,使两者间的因果关系呈现一定的复杂性。初步检验后发现,在不进行内生分组的情况下,全国均值水平上的产业协同集聚显著抑制了全要素生产率增长。进一步利用246个地级市面板数据进行内生分组检验,发现产业协同集聚对全要素生产率的影响呈现非线性特征:随着构成产业专业化集聚度的升高,生产性服务业与制造业协同集聚对全要素生产率增长的影响倾向于由负转正。

关键词: 产业集聚, 协同集聚, 全要素生产率, 最大矩估计, 非线性影响