Table 10

Program Impact on Participant Income, Treatment Effects Model: Households without Autonomy (n = 161)

Income ComponentImpactImpact with Lag
EstimateρaEstimateρ
Program Impact (δ)
Cropping before subsidy54.54 (252.76)−0.559 (0.621)52.55 (214.78)−0.191 (0.679)
Cropping with subsidy141.22 (141.63)−0.622 (0.299)199.54 (211.59)−0.512 (0.536)
Husbandry176.97* (106.8)−0.098 (0.18)194.43 (136.61)−0.098 (0.18)
Total agricultural before subsidy78.42 (141.63)−0.138 (0.208)220.35 (171.86)−0.136 (0.206)
Total agriculture with subsidy131.3 (146.18)−0.141 (0.216)257.67 (175.82)−0.137 (0.214)
Off-farm−45.86 (189.18)−0.025 (0.236)−46.54 (226.64)−0.025 (0.236)
Noncropping135.61 (219.67)−0.081 (0.219)152.35 (267.29)−0.081 (0.218)
Other−94.18 (251.23)0.118 (0.573)2.53 (215.58)0.093 (0.454)
Total before subsidy−7.29 (303.84)−0.101 (0.269)217.62 (344.28)−0.097 (0.263)
Total with subsidy45.79 (316.07)−0.102 (0.281)254.92 (354.49)−0.097 (0.274)
Program Lagged Impact (δL)
Cropping before subsidy−52.47** (21.71)
Cropping with subsidy−44.31* (23.24)
Husbandry−7.25 (35.37)
Total agricultural before subsidy−59.78 (41.84)
Total agricultural with subsidy−53.51 (42.09)
Off-farm0.28 (51.62)
Noncropping−6.96 (63.31)
Other−35.77 (30.71)
Total before subsidy−95.34 (73.7)
Total with subsidy−89.1 (74.06)
  • Note: Standard errors are in parentheses. Selection and impact equations were estimated simultaneously via maximum likelihood.

  • a Significant levels for the correlation coefficient are for chi-squared tests.

  • * Significant at 10%;

  • ** significant at 5%;

  • *** significant at 1%.