At the invitation of Timothy Ogden and Sona Partners, I am covering the Microfinance Innovation and Impact Conference an event co-hosted by Innovations for Poverty Action, the Financial Access Initiative, Moody’s, Deutsche Bank and CGAP. These posts reflect my personal observations and thoughts. Given the semi-live nature of these blog posts, please forgive any typos, or unintentional errors or omissions. Please feel free to add comments if you have any questions/thoughts.
There were two interesting findings that came out of this morning’s sessions. Duflo and co. found that offering microfinance to rural areas in Morocco which never had previous access to it lead to men leaving their wage positions and turn to personal farming. In addition, they found that money spent increased on the farms. In other words, the additional money did not cause people to spend money on new ventures, rather they invested in what they were already doing. Laura Starita covered of much of what was discussed in the opening panel in more detail (featured Esther Duflo, Dean Karlan, and Abhijit Banerjee).
It is key to note that the findings from the Moroccan study are still quite raw. Though she said it quickly, Duflo said that it is only three weeks old. This is very exciting that the information was presented from a two year study, but means that there are still questions that have to be teased out (and more research that will come from the findings).
The big question to answer will be: why did the farmers leave their wage jobs? A few possibilities include; having the capital to invest in their own work, dissatisfaction with wage labor, lost incentive to work, and perception that working on own farm will better take care of family. It will be improper to make any assumptions or conclusions based on this finding, but it does teach microfinance a few things. First, traditional and straightforward microfinance is not going to work the same way in every circumstance. Some of you will say, ”of course,” but studies that provide different forms of microfinance are only just getting started. What has worked in cities will not work everywhere else.
The solution, as prescribed by all the panelists thus far, has been randomized controls. However one might feel on the topic, there has been a common idea so far: “We need more research.” Yea, it does sound like a bunch of academics looking for more work, but their results have, thus far, yielded very little in terms of significant findings. This may largely be due to the fact that more time is needed to make a more complete evaluation, but two years is a somewhat significant amount of time. So, it is worth following to see if the need for more research is due to an actual need verses a problem in methods (I do not want to suggest randomized controls do or do not work as that is ‘above my pay grade’).
The purchase of livestock was present in both the first panel and the second panel with Christopher Udry (Yale), James Vickery (Federal Reserve Bank of New York) and Michael McCord (MicroInsurance Centre). It was found that people in both studies did not increase their consumption levels, but did put more money in to the purchase of farm goods, such as livestock. It is notable that this occurred, but again, still too early to draw any significant conclusions. Does the holding of more assets provide for future income gains? How does the purchase of livestock coupled with the move away from wage labor affect the local economy?
There are still many questions, but some conclusions have been drawn thus far: 1) What people want is savings, not credit. Mere access to savings is not enough. 2) Evidence of heterogeneity – (households w/ prior activities have significant reduction in consumption, without had insignificant increase) 3) When given loans people who will start business reduce consumption, have a business see small increase in consumption and will not starting business have a large increase in consumption.
One interesting footnote about the second panel on micro-insurance. When providing capital or insurance alone, there were often modest changes in behaviors. The combination of the two saw significant benefits such as a large reduction in skipped meals.