Value chain interventions are increasingly popular amongst donors aiming to promote market-oriented growth and poverty reduction. Based on the reflections of the community of practice itself and extensive desk research, this review critically examines the causal models underlying value chain interventions and asks how and to what extent their poverty alleviation impacts have been systematically investigated.
Concentrating on a selection of 30 donor-led value chain interventions, the review finds two main patterns of engagement: (a) one which funnels assistance by partnering with lead firms in the value chain – lead firm projects; and (b) one which works with chains without a lead firm – value chain linkage projects.
Targeting of the poor seems more effective in value chain linkage projects and in those lead firm projects where beneficiaries are identified in both the chain’s suppliers and distributors. Controversially, despite a wealth of positive anecdotal evidence, the vast majority of projects did not carry out an impact assessment of their poverty alleviation objectives and it is therefore unclear whether the value chain intervention: (a) is responsible for the improvements observed; (b) benefits the poor disproportionately; and (c) is more cost effective than other alternative approaches. Assessing the poverty alleviation effects of individual interventions in a rigorous way is costly and challenging but necessary to ensure long term effectiveness of the interventions as well as optimising the use of public funds.
There is a need to carry out systematic impact assessment at the programme level to develop a strong evidence base. Finally, this review provides some guidelines for designing and managing value chain interventions, particularly regarding the identification of situations in which the value chain approach is most appropriate and those where other private sector-oriented approaches (such as Business Development Services and Making Markets Work for the Poor) may be more suitable or complementary.
Authors
