Table 4. The results of mediation effect.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
Variable	RD	RD	DTSI	DTSI	COST	COST	DTSI	DTSI	DTSI	DTSI
SR	0.190	0.718	0.638	4.501**	-0.061*	-0.085	0.483	7.942***	0.596	4.384**
	(0.32)	(0.93)	(0.90)	(2.18)	(0.03)	(0.07)	(0.56)	(1.07)	(0.90)	(2.20)
SRSQ		-1.030		-7.532*		0.044		-13.864***		-7.386*
		(1.77)		(4.20)		(0.12)		(2.38)		(4.24)
RD			0.075*	0.075*					0.080**	0.079**
			(0.04)	(0.04)					(0.04)	(0.04)
COST							0.062	0.062	-0.267**	-0.266**
							(0.10)	(0.10)	(0.13)	(0.13)
Intercept	-2.894**	-2.959**	8.887***	8.410***	-0.388***	-0.385***	9.188***	8.230***	8.641***	8.174***
	(1.23)	(1.21)	(2.18)	(2.26)	(0.10)	(0.10)	(1.19)	(1.15)	(2.17)	(2.26)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry fixed effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Province fixed effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
F statistics	25.7	23.1	101.9	96.3	716.5	646.5	351.3	341.1	92.6	87.7
Adjusted R-squared	0.217	0.217	0.121	0.121	0.969	0.969	0.123	0.123	0.122	0.122
Observation	134108	134108	133384	133384	992412	992412	989052	989052	133384	133384

Note: *, ** and ** is statistically significant at 10%, 5% and 1%, respectively. The models controlled for industry, province and year fixed effects. The robust standard errors of clustering at city level are presented in parentheses. F statistic indicates that the model has joint significance.