Tim Büthe
The Politics of Private Foreign Aid
This project analyzes humanitarian and development aid that is "private" in a dual sense:  The funds are raised from private sources, mostly through donations from private individuals, and allocated by non-governmental organizations (NGOs).  For this project, I have conducted expert surveys to identify the most important NGOs in four development issue areas (health, educations, water/sanitation/sewerage, and emergency relief), subject to the conditions that they be not-for-profit organizations and conduct/carry out on-the-ground development projects.  In a multi-year effort with two of my former teaching assistants and a number of research assistants, we then gathered disaggregated financial information from these NGOs.  The resulting dataset allows us to identify separately the resources that come from private sources and analyze the amount of such funds allocated by the NGOs to the 100+ developing countries identified by the World Bank as "low" or "lower-middle" income countries.
"The Politics of Private Foreign Aid: Humanitarian Principles, Economic Development Objectives, and Organizational Interests in the Allocation of Private Aid by NGOs"  (co-authored with Solomon Major and André de Mello e Souza).  Forthcoming in International Organization.
Private money—raised and allocated by transnational NGOs—increasingly enables services and investments in health, education, and infrastructure in poor countries.  Yet, we know very little about NGO-allocated private aid.  We do not even know the answer to the basic empirical question: How is private development aid distributed across recipient countries?  This paper presents an original dataset, based on detailed financial records from most of the major US-based development NGOs, which allows us to map for the first time the allocation of private development aid.
Our initial mapping of private development aid then allows us to ask the analytical question:  What explains the allocation of private aid?  In rigorous statistical analyses, we find no support for the common claim that aid NGOs systematically prioritize their organizational self-interest and fundraising opportunities when they allocate private aid, and we find only limited support for the hypothesis that expected aid effectiveness drives aid allocation.  By contrast, we find strong support for the argument that the deeply rooted humanitarian discourse within and among aid NGOs drives their aid allocation to the countries with the highest level of "objective" humanitarian need, consistent with a view of aid NGOs as principled actors and constructivist theories of international relations, which may be particularly applicable to these NGOs.
The strong finding in support of the humanitarian hypothesis are robust to the use of seven different measures of aid recipients' objective need, including per capita GDP, the Human Development Index, the Physical Quality of Life Index, and the share of the population living on less than $1/day or below the local poverty line.  By contrast, measures of the likely efficacy of aid—including multiple measures of recipient country corruption, regime type, and political violence—yield only limited support for the development hypothesis.  Finally, the lack of support for the systematic strategic use of private aid allocation for fundraising purposes in the organizational self-interest of the NGOs is robust to the use of several different measures, including print media and TV news coverage, as well as the inclusion of control variables for numerous other possible explanations of NGO aid allocation.
In future work, I plan to examine in greater depths the causal mechanisms hypothesized in the paper on private aid allocation and to analyze more closely several of the factors which the above paper treats simply as control variables in the statistical analysis.  Most importantly, there is little doubt that most humanitarian aid NGOs experience competitive pressures for funding.  A wealth of research in behavioral economics suggests that such material concerns should crowd out other, more normative motivations, but we find no evidence of such crowding out in our data.  In the conclusion of the paper, we speculate as to possible reasons, which await systematic analysis in future work.








< page last updated September 2011 >