Table 5.

The results of the dynamic spatial Durbin model.

		Pool	Subsample
		China	East	Middle	West
		(1)	(2)	(3)	(4)
$L.ga{d}_{t-1}$		1.157 ***	0.904 ***	0.957 ***	1.352 ***
		(0.015)	(0.038)	(0.022)	(0.032)
short-term	Direct effect	0.189 ***	0.556 ***	0.635 ***	0.284 ***
		(0.036)	(0.140)	(0.092)	(0.069)
	Indirect effect	1.040 ***	1.527 ***	1.587 *	0.886 *
		(0.124)	(0.268)	(0.207)	(0.403)
	Total effect	1.229 ***	2.083 ***	2.222 ***	1.070 ***
		(0.125)	(0.328)	(0.235)	(0.457)
long-term	Direct effect	0.117	0.067 ***	0.014 **	0.051
		(0.641)	(0.067)	(0.017)	(0.057)
	Indirect effect	0.783	0.312	0.048	0.062
		(0.083)	(0.068)	(0.048)	(0.062)
	Total effect	0.056	0.146 **	0.249	1.079
		(0.031)	(0.063)	(0.027)	(0.196)
Control variables		Yes
$\rho$		0.277 ***	0.192 ***	0.058 ***	0.543 ***
$R^2$		0.987	0.995	0.998	0.932
${\sigma }^2$		0.001 ***	0.001 ***	0.001 ***	0.001 ***
L-L		249.700	233.500	378.200	272.400

Notes: $ga{d}_{j,t-1}$ represents the one-period lag of agricultural green development. t statistics in parentheses. $\rho$ represents the spatial autoregressive coefficient. L-L represents the log likelihood estimator. *, ** and *** denote statistical significance levels at 10%, 5%, and 1%, respectively.