SMALLHOLDER FARMERS’ PARTICIPATION IN SUSTAINABLE LAND MANAGEMENT PRACTICES AND ITS IMPACT ON CROP PRODUCTION AND FARM INCOME IN SOUTHERN ETHIOPIA

Abstract

The heavy dependence of farming communities on agriculture exposes land resources to continuous depletion and ruin. Ethiopia has been implementing sustainable land management (SLM) practices over the last four decades to cope with the problem. Exploring the socioeconomic, institutional, biophysical, and policy aspects contributing to the sustainability and effectiveness of land management practices is of paramount importance. This study analyzed farmers’ participation decisions and intensity of participation, and examined socioeconomic, institutional, biophysical, and policy factors that influenced their perceptions of SLM practices and preferred choices at a household level. It also evaluated the impacts of participation in SLM on the value of crop production and farm income. Cross-sectional data were collected in 2020/21 from 475 households drawn randomly from 6 woredas and 12 kebeles. Data were analyzed using descriptive statistics and econometrics models namely ordered probit, truncated double hurdle, multivariate probit model, propensity score matching technique, and endogenous switching regression model. The ordered probit model result revealed that education, cultivated land, training, land market, biophysical attributes of plot, and policy factors (land certificate, community bylaws, and incentives) influenced farmers' perception of SLM practices. The truncated double hurdle model result revealed that gender, social network, perception, land size, extension service, farm location, fertility status, slope gradient, and soil erosion showed a significant association in influencing the SLM participation decision. At the same time, non-farm income, value of crop production, and land market have reduced the participation decision. The second hurdle result also showed that farm size, value of crop production, training, distance of road, and community bylaws show a significant effect on farmers’ decision to allocate more proportion of farmland (intensity) to implement land management practices. Furthermore, the multivariate probit model result indicated that gender, education, cultivated land size, livestock holding, farm income, crop choice, institutional, and biophysical farm plot attributes affect SLM choices. The analysis further showed that five of the SLM practices combinations, namely fanya juu with soil bund, bench terrace and indigenous measure, and soil bund with bench terrace and the indigenous practices were applied jointly as complementary practices, while bench terrace with indigenous conservation measures has trade-off effect to be applied as a remedy to reduce soil erosion threat. The predicted marginal probability showed that a soil bund with a bench terrace was found to be the highest combination (i.e. 67.6%) and the lowest with indigenous conservation measures (26.9%).The propensity score matching estimator disclosed that farm plots that received SLM practices for continuous five years experienced 40.8% significant increments in the value of crop produced. Furthermore, the endogenous switching regression method disclosed that farmers who used SLM practices but they had not applied the measures to mitigate land degradation and soil erosion decreased the value of crop production and farm income by 27.2% and 73.9%, respectively. The study strengths that development programs and policy initiatives should depend on implementing physical structures, pay attention to the non-monetary aspects of farmers’ perceptions, participation decisions, and SLM choices within the context of their endowed socioeconomic, institutional, biophysical, and policy factors. Based on the impact finding, this paper concluded that it is also very crucial to train and advice farmers to promote and scaling of area-specific SLM practices that maximize social and economic benefits via policy measurement