Unbundling the impacts of economic empowerment programmes: evidence from Malawi

Unbundling the impacts of economic empowerment  programmes: evidence from Malawi

Download PDF 1.1 MB

Burchi, Francesco / Christoph Strupat
Discussion Paper 32/2018

Bonn: German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

ISBN: 978-3-96021-089-4
DOI: 10.23661/dp32.2018
Preis: 6 €

Social protection schemes, such as cash transfers are effective in improving the ability of beneficiaries to meet their basic needs: however, the evidence available shows that they are unlikely to move beneficiaries sustainably out of extreme poverty. While alternative anti-poverty programmes, such as the BRAC graduation scheme in Bangladesh, seem to contribute to this outcome, they are costly and it is unclear which component of these multi-sectorial interventions really makes the difference.

The present study assesses the impacts of an innovative pilot project called the Tingathe Economic Empowerment Project (EEP) in Malawi, designed by GIZ. The project targets ultra-poor and labour-constrained households. Designed as a cluster-randomised-control-trial, it provides to different village clusters a) a lump-sum transfer; b) financial and business training; or, c) both a lump-sum transfer and training. Furthermore, the project incorporated the feature that beneficiaries were allowed to appoint someone to carry out project-related activities on their behalf (the proxy), which could be of particular relevance to labour-constrained households. The main goal of the project was to position beneficiaries on a “graduation pathway”, that is, to provide them with the necessary resources with which to improve their well-being significantly and to lay the foundation for escaping poverty and their dependence on social assistance.

Our study has three main objectives: 1) to investigate the impacts of the overall project and of each of the three project components on several outcomes; 2) to verify whether project impacts differ significantly between households that satisfy all the criteria to be defined as “labour-constrained” and those that do not; 3) to assess the impacts of the proxy, and whether labour-constrained households benefited from this option in particular. In order to reach these objectives, we relied on the cluster-randomised-control-trial design and collected longitudinal information for 786 households (nearly 85 in each project component and 530 in the control group) before and after the implementation of the EEP.

Our results show that the project had substantial positive impacts, in particular on financial literacy, savings, loans, livestock wealth, agricultural production, and drought resilience. Impacts on the first three outcomes are entirely driven by the financial and business training. The lump-sum played a fundamental role in increasing livestock wealth and drought resilience, while the training in combination with the lump-sum were responsible for improving agricultural production. A key finding was that the training systematically increased the productive use of the transfer. However, longer-term impact assessments are needed to verify whether these improvements will translate into the beneficiaries’ graduation out of poverty.

Regarding the second objective, we find that results do not differ between labour-constrained and non-labour-constrained households. This finding provides initial evidence that poor and labour-constrained households can indeed benefit from an economic empowerment project and casts some doubt on the common view that labour-constrained households will always need to depend on social assistance.

Regarding the third objective, our findings point to the crucial role that active proxies play in enabling most of the project impacts. In particular, we find that an active proxy is a powerful enabler for productive activities. The larger benefits accrue to labour-constrained households. This highlights that proxies can be an important part of more inclusive economic empowerment programmes that also can include labour-constrained households.

Über die Autoren

Burchi, Francesco



Weitere Expertinnen/Experten zu diesem Thema

Brüntrup, Michael


Herrmann, Raoul


Loewe, Markus


Negre, Mario


Rippin, Nicole


Malerba, Daniele