Table 8. Robustness tests.

Variables	Panel A				Panel B
	Insurer Dummy i,j,t		Holding Ratio i,j,t		Insurer Dummy i,j,t	Holding Ratio i,j,t
	(1)	(2)	(3)	(4)	(5)	(6)
Local province i,j,t	0.2915***		0.2935***
	(4.35)		(4.11)
Ln (distance) i,j,t		-0.0599***		-0.0609***
		(-4.27)		(-4.06)
Local i,j,t					0.4756***	0.4928***
					(3.78)	(3.78)
Controls	√	√	√	√	√	√
region-, industry-, and year- FE	√	√	√	√	√	√
Observations	336,825	336,825	336,825	336,825	336,825	336,825

Column 1 and column 2 (Column 3 and column 4) of Table 8 present the results of regression for Eq (1) using Insurer dummy (Holding Ratio), respectively, as two forms of proxies for insurer, which is described and defined in Section 3.2. Columns 1 and 3 report regression results for Eq (1) in which the coefficients estimate of Local province were 0.2915 (t = 4.35) and 0.2935 (t = 4.11), respectively, both statistically significant and positive at the level of 1%. Columns 2 and 4 report regression results for Eq (1) in which the regression coefficients of Ln (distance) of -0.0599 (t = -4.27) and -0.0609 (t = -4.06), respectively, both statistically significant and negative at the level of 1%. Column 5 and column 6 of Table 8 present the results of regression for sub-sample that exclude giant insurance companies. The coefficients estimate of Local were 0.4746(t = 3.78) and 0.4928(t-3.78), respectively. The numbers in parentheses are t -statistics and standard errors clustered at the firm and year level. * p<0.1; ** p<0.05; *** p<0.01.