Adoption and intensity of agricultural mechanization and their impact on non-farm employment of rural women

This study analyzes the impact of the adoption of agricultural mechanization and its intensity on the non-farm employment of rural women using the 2016 China Labor-force Dynamics Survey data. The study captures mechanization adoption as a dichotomous decision and adoption intensity using three types of farming strategies: non-mechanized, semi-mechanized, and fully-mechanized. Non-farm work is categorized based on work types (self-employment or wage employment) and work locations (local or migrated non-farm work). Both inverse probability weighting with regression adjustment (IPWRA) estimator and multivalued treatment effects (MVTE) model are utilized to address selection bias. The IPWRA estimates reveal that mechanization adoption increases the probability of rural women participating in non-farm work in general and wage employment and local and migrated non-farm work in particular. The impact is greater for unmarried women than for their married counterparts. The MVTE estimates show that relative to non-mechanized farming, the adoption of semi-or fully-mechanized farming increases the probability of rural women participating in non-farm work, wage employment, and local and migrated non-farm work, with fully-mechanized farming playing a larger role. Meanwhile, relative to semi-mechanized farming, adopting fully-mechanized farming does not have a significant impact on any type of non-farm work.