3.3. Do marginalized farmers benefi

3.3. Do marginalized farmers benefit from participating in BVIS?
PSM in combination with ESR was also used to assess whether marginalized farmers benefit from
participating in BVIS in terms of net annual agricultural income per household.We expected the impact
on agricultural income to be fairly distributed across the marginalized groups using the three matching
estimators. The PSM analysis was run on sub-samples in three categories, namely, femaleheaded,
low-income farmers, and the youths, respectively (Table 6). The ESR conditional expectations
in equation (8) were also computed according to the sub groups.
The ATT for youths estimated using the kernel method (ESR) is MK92,149 (US$549) and is significant
at the 1% level across all the matching estimators. In other words, the youths benefited from irrigation
and earned about 315% more annual net agricultural income than what they could have earned
if they did not participate in the irrigation scheme. The ATT estimates for ESR and PSM are consistent
and having similar values. This is further supported by the sensitivity analysis results whose value of
Γ = 2 5 . implies that participants with the same vector would have to differ in their odd of participation
in irrigation by a factor of 2.5 (150%) in order to render the ATT results invalid (Table 6).
The ATT for female-headed farmers estimated using the kernel method (ESR) is MK75,367 (US$449),
representing a 312% increase in annual net agricultural income than what they could have earned if
they did not participate in BVIS. The ATT estimates for ESR and PSM are consistent and having similar
values. The value of Γ = 2 8 . implies that participants with the same vector would have to differ in their
odd of participation in irrigation by a factor of 2.8 (180%) in order to render the ATT from PSM results
insignificant. This finding suggests that irrigation has a positive impact on the net annual agricultural
income per households among female-headed farmers. Under the BVIS, 13 boreholeswere constructed
within the 14 villages that significantly reduced the time women spend to fetch water. This motivated
women to participate in the scheme thereby realizing multiple benefits from reduced drudgery
of water collection and income from the sales of produce from the irrigation scheme.
The ATT for poor farmers estimated using ESR is five times higher than that estimated using PSM.
This is also consistent with sensitivity results which showed a value of Γ =1 5 . implying that participants