Skip to main content

Aid in the SDG-era: Understanding aid effectiveness using a disaggregated approach

Research project
Active research
Project size
495 000
Project period
2018 - ongoing
Project owner
Department of Economics

Swedish National Science Foundation

Short description

The question of if and how foreign aid can help promote economic development in poor countries is a subject of controversy for academics and policymakers. This project merges new aid data identifying where and when aid programs took place along with geographically located outcome data to investigate the effectiveness of aid. We use the data to study the effectiveness of aid projects, with a focus on aid promoting health, education and institution building. As such, this section identifies what aid programs work, where and when, generating policy recommendations for future aid projects. We also evaluate the unintended impacts of aid programs on behavior using both location specific aid data and detailed field studies. Aid program may have unintended consequences that are often not measured resulting in positive or negative spillovers.

Although the last half-century has seen significant advances in many dimensions of development, differences in the standard of living between citizens in the rich and poor countries are still enormous, with about 3/4 of a billion people still living below the World Bank’s extreme poverty line.

SDGs raise demand for research on aid effectivenes

To improve global living standards, the international community has agreed on a set of ambitious sustainable development goals (SDGs). The 17 goals, broken down into 169 targets, cover a broad range of economic, social, and environmental issues, the most fundamental objective arguably being to ‘end poverty in all its forms everywhere’ (United Nations, 2015). The need to design aid policies that assist developing countries reach the SDG targets creates demand for highly specific, yet generalizable, research on aid effectiveness.

Evaluating both direct and indirect effects of aid

The question of if and how foreign aid can help promote economic development in poor countries is, however, a subject of controversy for academics and policy makers alike. Unfortunately, the mixed  empirical evidence on aid effectiveness exacerbates rather than resolves the controversy (Deaton, 2013). In particular, there has traditionally been a sharp divide between the macro and micro literatures evaluating the impact of aid. Indeed, it is common to speak of a micro-macro paradox, highlighting the tendency of scholars to be able to identify positive impacts of individual aid projects but a difficulty to establish corresponding positive effects at the macro level. Further to this, most empirical studies only measure average impacts and ignore important variation and indirect effects of aid, such as cross-sectoral spillovers and behavioural side effects arising due to the donor presence. Considering the increased demand for detailed knowledge about aid effectiveness, and the difficulty reaching a consensus in the macro and micro literatures, this project will address the ‘missing middle’ in the aid effectiveness literature, evaluating both direct and indirect effects of aid.

A novel disaggregated approach

The aim of the proposed project is to investigate aid effectiveness using a novel disaggregated approach to shed light on the effectiveness of sector-, donor- and within-country location-specific aid and how it can be improved. Our research will thus focus on what works where, when and why, addressing the following, broad, questions:

  1. To what extent does aid from different donors to different sectors and different types of localities meet their intended objectives?
  2. Do these disaggregated aid flows differ in terms of spillover effects?
  3. What factors explain differences in impact across aid donors, sectors, and locations?

To answer these questions we will use Geospacial Impact Evaluation (GIE), a novel method utilizing subnational geocoded aid, outcome, and covariate data to evaluate the correlates and effects of aid programmes. Matching geo-referenced project level aid data, i.e. information on the content and sub-national location of specific aid projects, with geocoded outcome and covariate data from individual/household level surveys makes it possible to evaluate the sub-national distribution and local effects of aid projects systematically and on a wide scale across multiple recipient countries. That is, rather than estimating country-wide impacts of total aid, which is notoriously difficult, or analyzing the impact of single projects, GIE enables us to systematically estimate, for instance, whether health aid projects have direct effects on relevant health outcomes in the targeted areas, as well as their potential indirect effects on other relevant outcomes (e.g. poverty and school attendance). The GIE analysis will be complemented by qualitative and quantitative field studies of individual aid programmes. This additional information will be used to further understand why programmes may or may not be effective and to evaluate direct effects and indirect spillovers generated by aid programmes.

Main contribution of the research

As such, our main contribution is threefold. To meet the increasing demand for evaluating target specific aid impacts, we use a novel method to examine is aggregated aid flows by donors, sectors and sub-national localities. This analysis is difficult without sub-national, sector specific aid data, the kind we use in this project. And to capture the full extent of aid effects, we evaluate direct as well as the often-overlooked indirect effects of aid projects originating in externalities. Finally, we carry out in-depth analyses on selected projects to, qualitatively, disentangle mechanisms.

Ann-Sofi Isaksson

Joseph Vecci

Arne Bigsten

Dick Durevall