Adoption of conservation agriculture-based tillage practices in the rice-maize systems in Bangladesh

• Published in: World development perspectives Vol. 21; p. 100297
• Main Authors: Akter, Shaheen, Gathala, Mahesh K., Timsina, Jagadish, Islam, Saiful, Rahman, Mahbubur, Hassan, Mustafa Kamrul, Ghosh, Anup Kumar
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2021
• Subjects: Conservation agriculture-based tillage; Double-hurdle model; Econometric analysis; Farm household survey; Propensity score matching; O13; Econometric analysis; Conservation agriculture-based tillage; Double-hurdle model; Q15; C34; Farm household survey; Q16; Propensity score matching
• ISSN: 2452-2929; 2452-2929
• DOI: 10.1016/j.wdp.2021.100297

•A double hurdle model was used to examine factors affecting the adoption of conservation agriculture-based tillage (CA-T) practices.•Overall adoption of CA-T in R-M system was about 7% of annual cropped land.•Factors explaining adoption of CA-T are age, cropping system, climate, soil, machinery and project services.•Cropped area and farm size are factors that affect only the intensity of adoption.•CA-T produced same or higher yields than farmers’ conventional tillage practices. In this study, conservation agriculture-based tillage (CA-T) practices (zero tillage, minimum tillage and strip tillage) in rice-maize production systems were implemented through on-farm participatory research trials for five years in Bangladesh to evaluate them under farmers’ field conditions, build awareness among farmers about such practices, and diffuse them in the farming communities. This study examines the adoption of those practices using farm level data from participating and non-participating farmers from and surrounding the field trial areas. For this, a farm household survey was conducted with 606 farmers, who earlier either participated or didn’t participate in the trials. A double hurdle model was employed to determine the factors influencing CA-T adoption and intensity of its adoption. During five years of the trial period, participatory and non-participatory farmers respectively adopted CA-Ts in 12% and 3% of annual cropped areas with overall adoption of 6.6%. Several factors contributed to the probability and intensity of adoption. Variations in adoption existed between locations, cropping systems, and seasons. Contrary to conventional wisdom, the propensity score matching analysis showed that yields either remained unchanged or were higher in the CA-T practices compared to farmers’ practices. Access to critical inputs, knowledge dissemination, capacity building and cropping season influenced the diffusion of CA-T practices.