92
A. Pratiwi et al.
however, is found to be negatively correlated with the partnership. Thus, it can be
inferred that relatively wealthy farmers with larger farm size, with motorized vehicles
possession, are more influential and connected, hence the higher likelihood to join
the company partnership.
Several possible mechanisms arise to explain the results. In our surveyed region
which may reflect the Indonesian rural situation in general, farmers value face-to-face
meeting for information sharing and tend to seek it primarily using motorbike due to
the distance and contour of the rural road (Matous et al. 2015). These farmers with
high capacity for mobility are more likely to be exposed with important informal
channels for information sharing about resource-conserving practices (Bodin and
Crona 2009); thus higher likelihood to get connected with extension officials or
company officers who then refer them to join company partnership. While mobile
phone is important for farmers to contact their peers, they mostly used the text
message to ensure that the person they would like to meet is available at that time
(Matous et al. 2015). Mobile phone serves as a catalyst of information seeking, not
the primary tools.
However, only knowing such opportunity is not enough to warrant the membership, as they have to maintain the quality of their production in addition to using
non-chemical pesticide. Even if only one farmer in the group did not meet the production standard, all members will be given a warning to adhere to the requirements.
This is when the strong network intensity of the group helps to push the members to
meet certain production qualities as set by the MNCs because the group partnership
status relies on individual compliance. These peers networks may lead them to keep
the constant quality of their produce, thus maintaining their membership over the
time. This corroborates the results of Pratiwi and Suzuki (2017, 2020), who found
that farmers who have more active networks inside their locality tend to be more
innovative and resourceful in their farming practices.
Effects of Company–Community Partnership on Adaptation
and Mitigation Practices
We then employ logit to match the partnership group with the non-partnership group
using kernel-based matching method and estimate the propensity score of whether a
farmer will join the partnership. Figures 5.2 and 5.3 show an overlap of treated and
untreated cases distribution and matching situation. Through the matching, average
bias decreases from 10.4 to 5.9%, and 291% untreated and 319% treated cases are
located in the region of common support. The balancing test result shows that four
variables were significant before the matching, but no more significant after matching,
implying desirable matching quality.
Using the matching, we obtained the average treatment effect on the treated (ATT)
of the company partnership in Table 5.4. We show the ATT of the company partnership upon the adoption of soil and conservation techniques, grafting methods,
A. Pratiwi et al.
however, is found to be negatively correlated with the partnership. Thus, it can be
inferred that relatively wealthy farmers with larger farm size, with motorized vehicles
possession, are more influential and connected, hence the higher likelihood to join
the company partnership.
Several possible mechanisms arise to explain the results. In our surveyed region
which may reflect the Indonesian rural situation in general, farmers value face-to-face
meeting for information sharing and tend to seek it primarily using motorbike due to
the distance and contour of the rural road (Matous et al. 2015). These farmers with
high capacity for mobility are more likely to be exposed with important informal
channels for information sharing about resource-conserving practices (Bodin and
Crona 2009); thus higher likelihood to get connected with extension officials or
company officers who then refer them to join company partnership. While mobile
phone is important for farmers to contact their peers, they mostly used the text
message to ensure that the person they would like to meet is available at that time
(Matous et al. 2015). Mobile phone serves as a catalyst of information seeking, not
the primary tools.
However, only knowing such opportunity is not enough to warrant the membership, as they have to maintain the quality of their production in addition to using
non-chemical pesticide. Even if only one farmer in the group did not meet the production standard, all members will be given a warning to adhere to the requirements.
This is when the strong network intensity of the group helps to push the members to
meet certain production qualities as set by the MNCs because the group partnership
status relies on individual compliance. These peers networks may lead them to keep
the constant quality of their produce, thus maintaining their membership over the
time. This corroborates the results of Pratiwi and Suzuki (2017, 2020), who found
that farmers who have more active networks inside their locality tend to be more
innovative and resourceful in their farming practices.
Effects of Company–Community Partnership on Adaptation
and Mitigation Practices
We then employ logit to match the partnership group with the non-partnership group
using kernel-based matching method and estimate the propensity score of whether a
farmer will join the partnership. Figures 5.2 and 5.3 show an overlap of treated and
untreated cases distribution and matching situation. Through the matching, average
bias decreases from 10.4 to 5.9%, and 291% untreated and 319% treated cases are
located in the region of common support. The balancing test result shows that four
variables were significant before the matching, but no more significant after matching,
implying desirable matching quality.
Using the matching, we obtained the average treatment effect on the treated (ATT)
of the company partnership in Table 5.4. We show the ATT of the company partnership upon the adoption of soil and conservation techniques, grafting methods,