My last post highlighted seven randomized evaluations of microfinace programs from around the world. I’ll admit, you have to be a little wonky to read through even one of the papers completely. Luckily, Innovations for Poverty Action (IPA) has created an easy to understand policy bulletin. [Read the entire brief here] Here are some highlights:
1. Demand for many of the microcredit products was modest.
2. Expanded credit access did lead some entrepreneurs to invest more in their businesses.
3. Microcredit access did not lead to substantial increases in income.
4. Expanded access to credit did afford households more freedom in optimizing how they earned and spent money.
5. There is little evidence that microcredit access had substantial effects on women’s empowerment or investment in children’s schooling, but it did not have widespread harmful effects either.
Take-up is an important indicator of the success of any customer focused enterprise. In settings where access to the services of the MFI were provided to anyone who passed as “eligible” (poorly defined) 13-30% actually took advantage. That’s 3 out of every ten people. At best! (Even in places where access to the MFI was selected out of a group of people who expressed interest for microcredit take-up was roughly around half of the population.) This fact alone should be a huge reality check for those who advocate for microfinance as the vehicle which paves a road out of poverty for the masses.
Microcredit did lead to increased business ownership in some locations. But a quick statistics lesson and a caveat seem important.
First, for a lesson in statistical significance. Only two out of the seven studies reported results of statistically significant difference between the treatment group and the control group. What this means is that the other five studies did not find much variation in increased business ownership of those who had access to MFI’s compared to those who had no access to MFI’s. This is important as the impact (or average treatment effect) of any program represents the effect of those who received treatment minus the effect of those same people who did not receive treatment. It is actually impossible to measure this, as we can’t go back in time and see how a household would fair in the absence of an MFI. When we randomize assignment (access to the MFI, in this case) we are mimicking this experimental ideal by comparing individuals who are statistically the same.
Second, increased business ownership may not be something we want to see from an MFI. Spend any time in any developing country and you will notice that there is no shortage of small businesses. They line the street and side alleys. Some mistake this popularity in business ownership as a propensity for entrepreneurship among the global poor. It may rather be due to a lack of other viable alternatives for economic activity which drives this popularity in business ownership than anything else. Starting a business is often easy, it’s sustaining it and expanding it which is the hard but important part.
In summary microcredit fails to impact the things that matter most, consumption (economists favorite observable variable for economic well-being) and social well-being. In fact some studies find decreases in these outcomes!
What does this mean for microcredit moving forward? I don’t think these studies should spell the end for microcredit as a micro-development strategy around the world. I simply think our collective enthusiasm for this medium of assistance needs to be a bit more muted and our expectations need to be a bit more realistic. Also, perhaps microcredit misses the target of what the global poor actually need. Many use microcredit as a way to smooth consumption when income (particularly for farmers) is lumpy. Perhaps we need to think more about products that assist in helping people have more freedom with how they spend their money. Clearly microcredit is not perfect, much can be done to tweak and improve this method. Of course, the best way to do this is through iteration. Trying something new, testing it, gathering feedback, and improving.