Table 4.
Parameter estimates of cotton land model for different land sizes
Variables | (1) | (2) | (3) | (4) | (5) |
---|---|---|---|---|---|
Land Size (ha) | |||||
0–3 | 3–10 | 10–15 | 15–20 | 20 + | |
Cotton land (t-1) | 0.2920*** | 0.3373*** | 0.1556*** | 0.2125*** | 0.2371*** |
(0.0310) | (0.0264) | (0.0325) | (0.0430) | (0.0464) | |
Other land (t-1) | 0.0992* | 0.2824*** | 0.0443 | 0.0980** | 0.0658** |
(0.0516) | (0.0570) | (0.0442) | (0.0418) | (0.0303) | |
Cotton support (t) | 0.0002 | 0.0011*** | 0.0041*** | 0.0027 | 0.0046** |
(0.0001) | (0.0004) | (0.0012) | (0.0018) | (0.0020) | |
Plot number (t) | − 0.0067*** | − 0.1111*** | − 1.0229*** | − 1.3543*** | − 1.4474 |
(0.0021) | (0.0302) | (0.2328) | (0.4550) | (1.5349) | |
Cotton price (t-1) | 0.1424*** | 0.8744*** | 1.2441*** | 1.2584*** | 2.3891*** |
(0.0343) | (0.0940) | (0.2773) | (0.4025) | (0.5431) | |
Fuel oil price (t-1) | − 0.2241*** | − 1.2809*** | − 1.2883*** | − 1.1718 | − 3.9182*** |
(0.0613) | (0.1325) | (0.4129) | (0.7196) | (1.1471) | |
Tenant (t) | 1.0935*** | 2.7685*** | 4.2878*** | 5.9217*** | 9.2951*** |
(0.0631) | (0.1193) | (0.4077) | (0.8001) | (1.3957) | |
Farmer’s age (t) | 0.0070 | 0.1561*** | 0.0536 | 0.0896 | 0.2143 |
(0.0087) | (0.0269) | (0.0733) | (0.1082) | (0.1851) | |
Year 2011 | 0.1876*** | 0.5898*** | 1.2966*** | 1.5236*** | 2.4194*** |
(0.0127) | (0.0347) | (0.1199) | (0.2204) | (0.2834) | |
Constant | 0.7507* | − 4.8107*** | − 0.7481 | 2.3291 | 7.4789 |
(0.4379) | (1.6177) | (4.2593) | (6.2848) | (7.4780) | |
Arellano/Bond Test | |||||
m1 | − 16.4140 [0.0000] | − 23.9870 [0.0000] | − 15.7940 [0.0000] | − 11.2710 [0.0000] | − 6.9782 [0.0000] |
m2 | 0.5437 [0.5866] | 2.6629 [0.0077] | 3.9482 [0.0001] | 2.3595 [0.0183] | 1.7509 [0.0800] |
Sargan/Hansen Test | |||||
χ2 | 83.5233 [0. 0000] | 162.5112 [0. 0000] | 92.3768 [0. 0000] | 63.4521 [0. 0000] | 80.0747 [0.0000] |
Number of instruments | 30 | 30 | 30 | 30 | 30 |
Observations | 23,971 | 40,880 | 9766 | 4838 | 6960 |
We used a dynamic panel model, a two-step generalized moments approach, and considered land and support variables endogenous. Robust standard errors were given in the parenthesis. Stars represent the significance levels: *** p < 0.01, ** p < 0.05, * p < 0.1. Arellano/Bond is the autocorrelation test, m1 is first order, m2 is second order, Sargan/Hansen test is the overidentification restriction, χ2 Chi-square. In the test section of the table, the numbers in the brackets show the p values