Ryan C. Bakow, Political Science

PROJECT ABSTRACT

Billions of dollars each year are spent on foreign aid projects, much of which ends up in the hands of non-governmental organizations (NGOs) all around the world. NGOs are involved in many phases of development and have often accomplished great things; however, despite a long history of NGO work, there have been very few studies which consider the actual effectiveness of this process. In this study we hypothesize that because NGOs work outside of the traditional market, and because of a lack of accountability, there may be significant inefficiencies at the project level. To test this hypothesis, we perform a randomized controlled trial in Uganda wherein we measure the level of NGO efficiency by eliciting project cost estimates and comparing them to other cost level measures. This experiment is done primarily through e-mail. Our research demonstrates a surprising amount of inefficiency among NGOs; additional treatments indicate that this may be the result of opportunism caused by information asymmetry. We also demonstrate that correcting these inefficiencies via the introduction of simple accountability mechanisms is untenable. Anticipating such an outcome, we implement a second, similar, experiment using businesses as the subjects. This experiment will test whether an alternate, free-market distribution system corrects these inefficiencies. However, extremely low response rates indicate that there are additional issues which may prohibit businesses from effectively participating in development to the same degree as do NGOs.

SIGNIFICANCE

This project was unique in development literature for two reasons: First, there is very little research on NGOs themselves. Most development literature is focused on the work being done by these organizations (i.e. the most effective way to distribute mosquito nets, or the effectiveness of micro loans, etc…). However, understanding the delivery mechanism for foreign aid gets at the heart of the critical question, is foreign aid effective? Second, this project is unique because it effectively uses a relatively unused political science research methodology, namely the field experiment. The properly implemented field experiment, or randomized controlled trial, is difficult to create outside of the lab (hence its scarcity). However, in the right circumstances, a field experiment can identify cause and effect, and in this case it also allows us to better understand the motivations driving both.

SUCCESSES AND FAILURES

This project was a very large undertaking for one student. The experiment took almost a year of planning and preliminary work. The analysis and thesis took an additional six months to complete. This could not have been accomplished without the support and help of my faculty mentor and my colleagues in the political science department, all of which put in many hours of work.

The project turned out to be more successful than any of the team could have hoped for. We were able to successfully test multiple factors with statistically significant samples. Even better, we achieved many of the results that we predicted, something which does not often happen in the social sciences. The major challenges we faced included: narrowing down the scope of the project, developing effective treatments, creating the proper sampling pool, and coding and analyzing the data. But with a lot of help and mentoring we were able to successfully navigate these challenges and publish a very fulfilling project.

WHERE IS THE RESEARCH NOW?

Our research immediately spawned several similar projects at universities around the country and received some attention from other professors in the field. More importantly, Professor Daniel Nielson, my original faculty mentor, was able to get funding to do a second round of research. This research is currently ongoing. They have a dedicated team of research assistants under the management of one of my colleagues Bryan Reed, and they are currently testing thousands of NGOs with a finely tuned version of our original experiment. There is a lot of excitement surrounding this project and we anticipate that the results will be a force for good throughout the development community.

Thank you again for your generous donation toward this important research. It is my hope that this research will help my friends in Uganda, Senegal, and all those who need a helping hand in any country in the world. I know it has already impacted me.