From Ghana, Suvojit Chattopadhyay writes:
So my friendly neighbourhood fruit-seller never started a savings scheme with me. No issues there, since it turns out that she has been saving with a susu-agent. Same deal - 1 GHC a day, saving up towards a certain targeted goal. The susu agent issues a little card to keep record and charges 1 GHC a month, irrespective of the amount saved during the month. Nothing surprising there - paying a charge on savings is a well-documented subject in the microfinance world and can be found in most parts of the world. Not surprising therefore that when I asked about why she paid a charge, she promptly replied that the susu agent not only kept the money safe, but also came to her everyday to remind her to save.
Recently though, she had been going slow on her savings, often not managing enough savings per month. As a result, she had not been saving at all for the last couple of months and I assume that's because she must have figured out that it isn't worth paying a fixed charge if she saved only a few days of the month.
About a month back, I had loaned her 50 GHC. I was slightly unsure if we had established that it was a loan and not a gift. Today, she said she had been saving up with her susu agent to pay me back part of the loan next week. She had apparently asked him to inform her when she touched a certain amount, so she could take out 20 GHC for me. Essentially, my interest-free loan to her wasn't interest-free anymore, since she had to pay to save in order to repay me. But she also seemed happy to have started saving once again...
For a brief moment, I contemplated if I should offer and attempt once again to explain to her the option of switching from the susu agent and saving with me, without any fee on her savings. But then it seemed like a bad idea and one that might disturb a thriving and sustainable arrangement on the ground, at least in the absence of a formal bank linkage...
Now what's that I hear, you are ready for some theory and quantitative evidence? Well a recent NBER working paper by IPA research affiliates Dean Karlan, Margaret McConnell, Sendhil Mullainathan, and Jonathan Zinman builds a theoretical model explaining why reminders to save work, and then demonstrates that they do with data from 3 field experiments in Bolivia, Peru, and the Philippines.
IPA Ghana is also currently running evaluations of