Can the big push approach end rural poverty in Africa?: Insights from Sauri millennium village in Kenya

This study sought to provide the first independent rigorous evaluation of the Millennium Villages Project, using Sauri millennium village (Kenya) as a case study. Sauri has been coined as a success story of the MVP across Africa, which makes it an ideal case study to assess the impact of the MVP interventions. The study is an ex-post evaluation and therefore applies non-experimental evaluation methods. The first empirical chapter of the thesis seeks to estimate the magnitude of the changes induced by the MVP interventions in agricultural productivity, production margins, selfconsumption and different sources of household income. We find significant MVP effects on agricultural productivity and total household income (including selfconsumption) but insignificant cash income effects. The chapter concludes that there is a need to test assumptions on which the theories of change are based in order to ensure effectiveness of the interventions. The second empirical chapter assesses whether the MVP effects could be explained by the level of diversification of sources of income. We find that: (i) there was an insignificant difference in the level of diversification between the millennium villages and the control villages; (ii) the main push factors into diversification were the small land and large household sizes; (iii) variations in income and poverty were also explained by land and household sizes, among other factors, with an insignificant MVP dummy and (iv) households that diversified more had lower MVP effects, a likely indication of MVP’s focus on agricultural interventions as opposed to the nonfarm economy. The third empirical chapter examined whether the performance of market institutions (input, output and credit markets) promoted or undermined the achievement of the MVP goals. We find that (i) increased access to inputs enhanced agricultural productivity in the first year, but the gains were not sustained after the phasing out of the input subsidy; (ii) collective marketing of produce through cereal banks was not successful and (iii) interventions in credit markets were not successful, given the low repayment rates of microfinance loans and low uptake of credit financing from commercial banks. Constraints to access to credit (especially lack of collateral and cost of borrowing) were still prevalent in Sauri after the MVP. The concluding chapter provides a summary of the findings, identifies key emerging issues and provides areas for further research. The key emerging issues were: the importance of designing the project as an experiment to enable more accurate impact evaluation, the failure to test the assumptions of the theory of change and sustainability of the gains beyond the project period