Abstract：
Based on the estimation of the threshold value with reference to the general threshold model, this paper constructs piecewise function to solve model parameters by using virtual variable generated from the threshold value. Then the paper carries out an empirical test using the panel data of 246 cities of prefecture-level and above in China during 2003-2016. It is found that the effect of industrial co-agglomeration with producer services on TFP(total factor productivity), PEC(pure efficiency change) and production scale change turns from negative to positive after the threshold value of 1.0299, 2.1741 and 2.0651 are passed respectively in the specialized agglomeration index of manufacturing, while this effect on pure technological progress turns from positive to negative after reaching the threshold value of 2.1741. With the specialized agglomeration of producer services being the threshold variable, the threshold effect of industrial co-agglomeration on TFP and its decomposition indicators is not significant, or even null.
WU Xian-fu.Research on Threshold Effect of Industrial Co-agglomeration’s Impact on Total Factor Productivity ——Based on the Empirical Test of 246 Cities in China[J] Economic Survey, 2019,V36(2): 72-78
2019 
