In this dissertation, I estimate the impacts of three pro-poor interventions implemented in Southern and Eastern Africa. I assess the heterogeneity of these policies, their intended and unintended effects. In so doing, I compile evidence on their strengths and limitations. In Chapter 1, I assess the role of a subsidy programme that aims to improve access to formal housing in South Africa. Against the background of inequality being associated with higher crime rates, I show that housing subsidies have not only reduced inequality, but they also mitigated violent crimes. The impact on property crimes, however, is limited. Moreover, I evaluate the impact of school feeding programmes on education in Malawi in Chapter 2. I show that significant results are limited to food-insecure areas. In addition, I find that the effects of school meals are also limited to attracting children into school for the first time, while the impacts on dropout rates are less robust. Finally, in Chapter 3, I estimate the impacts of the elimination of primary school fees in Tanzania. I compile evidence that the policy has increased enrolment rates, but it has had a limited impact on student retention. I show that the elimination of school fees has narrowed the gender gap in education, and it has especially benefited girls from districts with historically higher investments in school infrastructure. Overall, I argue that policy-makers must be aware of reform heterogeneities so as to exploit the strengths of educational policies and address their limitations by adopting complementary measures. Finally, I bring evidence that the impacts of pro-poor policies, such as housing subsidies, are not limited to individual gains, but they can also improve outcomes that are of importance to society as a whole, such as crime.