In light of the recent publication of “Poor Economics”, the new book by Abhijit V. Banerjee and Esther Duflo, MMU caught up with Dr. Duflo to discuss some relevant themes to help our readers understand the broader impact of mobile financial services in the economics of the poor. Today, we publish the second and final part of our interview focusing on the topics of mobile money, insurance and credit.
MMU: What advice do you have for mobile operators and others who are trying to develop financial services for the poor?
ED: Keep your eyes open, evaluate the impact of what you are doing, be willing to adapt and change, don’t over promise, and measure what you deliver.
MMU: You touch upon the notion of risk and how agricultural profitability could be 30% higher if there were no weather risks. In a time of climate change, what role do you see for agricultural mobile insurance products such as Kilimo Salama in Kenya?)
ED: The problem many current weather insurance product is that the demand for them tend to be quite low. The reason is that the products that the market can sustain are not considered to be very attractive by the farmers: these products are generally very simple (for example a weather insurance product will pay you if the rainfall level falls below a specified levels), which make them easy to administer, but also implies that they don’t cover many of the relevant risk: for example it may be very dry in your field but the rainfall at the weather station is above the threshold. There is a need to continue to develop better products and products that can be administered as cheaply as possible (so that they are worthwhile for farmers). But in the book, we argue that there may also be a need for governments or international organizations to subsidize insurance, at least for a while, while the market gets established.
MMU: The development community believed microfinance could increase the poor’s earnings partly through business growth, partly though new business creation and partly through greater investment in human capital. Yet, in practice microcredit was used not just for business investment but also for consumption, repaying other loans, etc. Was the non-business use of microcredit its downfall?
ED: No, I don’t think so. First I think that the problems that microcredit is experiencing today have nothing to do with its intrinsic value; they are mostly of a political nature. Second, it is a great use of microcredit to repay an old debt: if you have a debt on which you pay 70% interest or more (actually up to 5% per DAY for the fruit vendor), and a microcredit loan comes along, and you only have to pay 25% or 30% in interest, then it really help freeing up money. The rate of returns of doing that is likely greater than what you could do in your business. In a sense I would say that was surprising in our evaluation of a microcredit program in Hyderabad is that few people used their loan to repay an old loans. This was perhaps because they want to maintain a relationship the money lender in case they need money quickly.
I think the problem with microcredit was less what it is doing, which I think is quite respectable and important, than what it promised to do, which was a bit inflated. If the hopes had been reasonable, people would not be so disappointed today. I hope that microcredit will be able to continue and not be a victim of a fad in reverse.
MMU: Increasingly the development sector is turning to RCTs as the new standard for measuring impact? Every methodology has its drawbacks – how can development practitioners strengthen their understanding of impact? Are there methodologies complementary to RCTs than need to be maintained?
ED: Obviously I think RCT is a great methodology, and to be honest, I don’t see many drawbacks to it, but then of course I am biased. But there are other useful methods as well. For example, Tavneet Suri is evaluating the impact of M-PESA in Kenya using the progressing introduction of the program across Kenya: essentially she compares places over two periods, and some of them got the M-PESA program in the meantime, while other did not. One needs to be careful in exploiting the data, but this can be done and is very helpful. That said, a program like mobile money lends (or programs that can be layered on it) lend themselves wonderfully to RCT, so there should be a lot of exciting opportunities there.
MMU: Thank you very much
Esther Duflo is Abdul Latif Jameel Professor of Poverty Alleviation and Development Economics in the Department of Economics at MIT.. She has received numerous honors and prizes including a John Bates Clark Medal for the best American economist under 40 in 2010, a MacArthur “genius” Fellowship in 2009. She was recognized as one of the best eight young economists by the Economist Magazine, one of the 100 most influential thinkers by Foreign Policy since the list exists, and one of the “forty under forty” most influential business leaders under forty by Fortune magazine in 2010. Together with Abhijit Banerjee and Sendhil Mullainathan of Harvard University, she founded the Abdul Latif Jameel Poverty Action Lab in 2003