attractive enough for farmers, and since poor farmers tend to have short planning
horizons and as a result discount future benefits quite severely (cf. Sect. 5.3.4),
long-term positive effects on soil fertility are valued less highly than much more
immediate monetary gains, those to be made at the end of each cropping season. As
well as the (perceived) economic unattractiveness, institutional deficiencies also
constrain the adoption of specific techniques, such as inadequate access to planting
materials and training. In the above-mentioned stakeholder workshop in 2011,
farmers emphasized the need for field trials to be held at the local level to test the
performance of different soil conservation techniques and adapt them to farmers’
needs.
The level of knowledge on SCTs in general did not differ significantly between
wealth groups, though there were two exceptions to this rule: agroforestry and
terracing. Agroforestry was known by 45 % of the non-poor and 32 % of the poor
households (chi-square test significant at p < 0.10) and terracing by 24 % of the
non-poor and 10 % of the poor households (chi-square test significant at p < 0.05).
Further association tests carried out on the relationship between the incidence of
adoption, the scale of adoption and farmers’ wealth levels did not show significant
correlations for any of the SCTs listed above. Hence, with regard to the use of soil
conservation techniques in northern Vietnam, we did not find that poor farmers
invested less in long-term natural resource maintenance activities than those who
were wealthier; however, the adoption constraints identified and cited by
respondents indicated that their willingness and/or ability to invest in these
technologies were strongly determined by their access to natural, physical, financial
and human capital. The lack of correlation between wealth and conservation
investment indicates that these capital constraints may be binding upon most of
the farmers in the area, not only the poorest.
Econometric analyses on the adoption determinants of agroforestry presented in
Saint-Macary et al. (2010) and discussed in more detail in Chap. 7, provide further
evidence on this issue. The results from a multivariate household-level adoption
model showed that when controlling for households’ endowments of different types
of capital (natural, human, financial and social), the wealth level, as measured by
per-capita expenditures, appeared to be a significant determinant of both farmers’
level of knowledge on agroforestry techniques and their adoption decisions. In
addition, the results of the study also suggested that education was a significant
determinant. While access to formal credit did not appear to be a significant
determinant, the financial support and advice that farmers received when
implementing a given technology acted as a strong and positive influencing factor.
This confirms our hypothesis, that most farmers’ capital constraints are binding, and
also explains the low adoption rates we observed.
The linkage between poverty and investment with respect to long-term natural
resource maintenance was further investigated in a study by Ahlheim et al. (2009)
into the economic importance of landslides in Yen Chau district. Together with
floods, landslides constitute a major environmental risk in the area, as they cause
substantial damage to public infrastructure every year and destroy farmers’ fields
and houses (Schad et al. 2012). Forest removal and soil erosion are two direct
206
C. Saint-Macary et al.
horizons and as a result discount future benefits quite severely (cf. Sect. 5.3.4),
long-term positive effects on soil fertility are valued less highly than much more
immediate monetary gains, those to be made at the end of each cropping season. As
well as the (perceived) economic unattractiveness, institutional deficiencies also
constrain the adoption of specific techniques, such as inadequate access to planting
materials and training. In the above-mentioned stakeholder workshop in 2011,
farmers emphasized the need for field trials to be held at the local level to test the
performance of different soil conservation techniques and adapt them to farmers’
needs.
The level of knowledge on SCTs in general did not differ significantly between
wealth groups, though there were two exceptions to this rule: agroforestry and
terracing. Agroforestry was known by 45 % of the non-poor and 32 % of the poor
households (chi-square test significant at p < 0.10) and terracing by 24 % of the
non-poor and 10 % of the poor households (chi-square test significant at p < 0.05).
Further association tests carried out on the relationship between the incidence of
adoption, the scale of adoption and farmers’ wealth levels did not show significant
correlations for any of the SCTs listed above. Hence, with regard to the use of soil
conservation techniques in northern Vietnam, we did not find that poor farmers
invested less in long-term natural resource maintenance activities than those who
were wealthier; however, the adoption constraints identified and cited by
respondents indicated that their willingness and/or ability to invest in these
technologies were strongly determined by their access to natural, physical, financial
and human capital. The lack of correlation between wealth and conservation
investment indicates that these capital constraints may be binding upon most of
the farmers in the area, not only the poorest.
Econometric analyses on the adoption determinants of agroforestry presented in
Saint-Macary et al. (2010) and discussed in more detail in Chap. 7, provide further
evidence on this issue. The results from a multivariate household-level adoption
model showed that when controlling for households’ endowments of different types
of capital (natural, human, financial and social), the wealth level, as measured by
per-capita expenditures, appeared to be a significant determinant of both farmers’
level of knowledge on agroforestry techniques and their adoption decisions. In
addition, the results of the study also suggested that education was a significant
determinant. While access to formal credit did not appear to be a significant
determinant, the financial support and advice that farmers received when
implementing a given technology acted as a strong and positive influencing factor.
This confirms our hypothesis, that most farmers’ capital constraints are binding, and
also explains the low adoption rates we observed.
The linkage between poverty and investment with respect to long-term natural
resource maintenance was further investigated in a study by Ahlheim et al. (2009)
into the economic importance of landslides in Yen Chau district. Together with
floods, landslides constitute a major environmental risk in the area, as they cause
substantial damage to public infrastructure every year and destroy farmers’ fields
and houses (Schad et al. 2012). Forest removal and soil erosion are two direct
206
C. Saint-Macary et al.