How to GiveWell Using Impact Evaluation and Transparency to Foster Accountability, Innovation and Efficiency in NGOs

Abstract

Despite trillions of dollars spent on development assistance over the past sixty years, global poverty is still a seemingly insurmountable problem. Thus there is a pressing need for improved development outcomes. This thesis investigates how individuals and governments can use impact evaluation and accountability mechanisms to improve outcomes for development NGOs by synthesizing the literature on foreign aid, NGOs, accountability, impact evaluation, and randomized controlled trials and supplementing this with interviews. My analysis demonstrates the benefits of an upward accountability mechanism for NGOs to report to meta-charities. I conduct a detailed case study of the meta-charity GiveWell which reveals the obstacles of limited capacity, limited scope, weak marketing and complete transparency that need to be addressed to create an improved ‘GiveWell 2.0.’ I recommend that individuals (1) create and fund GiveWell 2.0 and (2) apply its research to their donation decisions. I recommend that the government foster a culture of evaluation by (1) including questions about evaluation on the non-profit tax form (IRS F- 990) and (2) embedding an evaluation institution within government. These steps would create a powerful feedback loop that would ultimately improve development outcomes. As research reveals better and more impactful giving opportunities, more people will be inspired to give effectively. As more people are inspired to give effectively, more NGOs will be encouraged to become accountable and transparent in their operations to capture these funds. As NGOs become accountable and transparent in their operations, meta-charities will be able to evaluate more NGOs, further increasing the confidence of donors in their giving opportunities, and thus creating donors who give more and who give more effectively.