It remains a challenge for agricultural research-for-development (AR4D) institutions to demonstrate to donors which technologies contribute significantly to poverty reduction due to a multitude of impact pathways. We attempt to overcome this challenge by utilizing the potential outcomes framework and quantile treatment effects analytical approaches applied on panel household data collected from Central Africa. Our findings show that adoption of AR4D technologies reduced the probability of being poor by 13 percentage points. A large share of this poverty reduction is causally attributable to adoption of improved crop varieties (32%) followed by adoption of post-harvest technologies (28%) and crop and natural resource management (26%), with the rest 14% attributable to unidentified and/or unmeasured intermediate outcomes or factors. The findings further indicate that relatively poor farm households benefit from adopting improved crop varieties more than the relatively better-off households. Correspondingly, the relatively better off households benefit from adopting post-harvest technologies enhancing crop commercialization much more than the relatively poor households. The findings reveal interesting policy implications for successful targeting of agricultural interventions aimed at reducing rural poverty.

http://dx.doi.org/10.1016/j.worlddev.2018.03.013

Multi standard citation

APAHarvardMLAVancouverChicagoIEEE

Tahirou Abdoulayehttps://orcid.org/0000-0002-8072-1363

Shiferaw Felekehttps://orcid.org/0000-0002-0759-4070

Arega Alenehttps://orcid.org/0000-0002-2491-4603

DONTSOP NGUEZET Paul Martinhttps://orcid.org/0000-0001-5098-1853

Victor Manyonghttps://orcid.org/0000-0003-2477-7132

http://dx.doi.org/10.1016/j.worlddev.2018.03.013

SOCIAL SCIENCE & AGRIBUSINESS

Agribusiness; Impact Assessment

Poverty; Evaluation Techniques; Farmers; Households; Agriculture Research-For- Development; Technology

Africa; Central Africa

Burundi; Congo, Dr; Rwanda

World Development