Stanford has started another academic year, and all of us here at IPSofacto have recently returned from summers spent making international policy, or at least sitting at desks trying to look like we were making international policy. To start off the year, we are going to run a few posts from talking about our experiences at different organizations in different countries. This is basically the grad school equivalent of the back-to-school how I spent my summer essay, except instead of going to camp some of us went to Kabul. And generally we’re not writing them in crayon.
I spent my summer at Innovations for Poverty Action, a non-profit that conducts evaluations and research on international development projects to determine what works and what does not for reducing poverty and improving welfare (primarily in the developing world but in the U.S. as well). Founded by a development economist at Yale named Dean Karlan, IPA is at the forefront of a movement in development circles to promote the use of rigorous research tools to help figure out if the development assistance is doing what it’s supposed to do. In practice, this usually means using randomized controlled trials to evaluate the effectiveness of specific interventions, for example the use of deworming treatments in schools to improve student attendance and performance (verdict: they work pretty well).
IPA and its sister organization, MIT’s Poverty Action Lab, have been fighting the good fight on randomized evaluations for nearly a decade, and in general they have been very successful in getting donors and other NGOs to adopt a more rigorous approach towards evaluating their work. But that success has not come without controversy. Proponents of RCTs (sometimes called randomistas by people on the internet who like giving things nicknames) have faced an avalanche of criticism, ranging from petty and self-interested to thoughtful and productive. IPA does a good job summarizing some of the more recent critiques here, along with a simple but effective response (for the more econometrically-minded, you might want to check out the paper here by Princeton’s Angus Deaton. It’s not for the faint of heart, but is worth reading if you can handle the funny symbols).
Still, IPA and its partners have added a lot to our knowledge of what works and what doesn’t in development. Their Proven Impact initiative highlights some of the more prominent successes, such as the deworming project mentioned above or the use of chlorine dispensers for safe water. Recently, Dean Karlan published findings that attempted to temper some of the wildest expectations people have about the potential of microlending, while other IPA affiliates (including Pascaline Dupas, now at Stanford) conducted a study showing that microsavings might have untapped potential of its own.
My own work at IPA touched on this last area, though it was not itself an RCT. IPA’s savings research has shown that giving poor entrepreneurs free access to savings accounts can have significant benefits. But even the poorest of the poor, people who generally do not have access to formal savings accounts or other bank services, can save in other ways. My project this summer was a pilot study looking at these informal savings practices in twelve countries. The goal was to design and test a survey on informal savings practices, partly to collect some preliminary data, but primarily to help us learn about the best way to gather this kind of information and to guide future research (it’s not always effective to just approach a random farmer in Rwanda and ask them where they keep all their money).
We don’t have final results yet, but the qualitative data we collected showed an incredible diversity of savings practices around the world. Rotating savings and credit associations, or ROSCAs, are one common example in which people contribute a fixed amount each week or month over a period of time. Each week (or month) a different person will receive all the money saved by the group to spend or invest or use for some predetermined goal. In this way, over the course of a savings cycle everyone will receive the pot once, allowing them to effectively save a large chunk of money without having to keep it in a bank account or under their mattress.
IPA is expanding its microsavings research to four larger studies in an attempt to replicate the initial findings and hopefully provide some external validity for a study that was initially only conducted in one country. And it currently has dozens of other projects going in dozens of countries, covering everything from official corruption to helping people quit smoking. It is a young, dynamic organization, and hopefully it will keep contributing to this kind of cutting-edge research for a long time to come.