On Why Microfinance Doesn't Reduce Poverty

Julia Galef talking about this very interesting podcast interview she did with Gates Foundation senior advisor David Roodman, the transcript of which you can read here.

In the discussion of the history of microfinance, there is one key event which seemed to trigger the collapse of the microfinance movement:

Then, I think, perhaps the most difficult development was in India where the leading microlender, SKS, went public, turning its founders into millionaires. And that really began to rub a lot of people the wrong way.

At the same time, there began to be stories of women committing suicide because of inability to repay loans from the same lenders. Politicians sensed this, and also, the news media picked on this.

There came a day in October 2010 where in the state of Andhra Pradesh in India, which is the leading state for microcredit, the state government essentially shut down the industry overnight, citing these suicides. And it just generated a huge amount of negative publicity. I think that the whole movement have not been the same since. ...

... I do take the reports of suicide in India five years ago with a grain of salt, because I know that they were promoted by politicians with their own agenda.

Roodman says, of the RCT studies done on the effectiveness of microfinance in reducing poverty:

If the studies had found that microcredit was reliably increasing income, or increasing household spending, that would have been fantastic. It was actually worth checking for both. Because reducing poverty is, of course, important in itself, but also because many claims on behalf of microcredit were grounded in this idea that it was reducing poverty. So, absolutely worth checking.

Later Roodman talks about how the poor do need financial services, for some things

In that context, savings, credit, insurance, even money transfers all help them solve what is a fundamental problem. Which is the need to gather lumps of money for important spending purposes, whether it be to buy a cow, or pay school fees, or cover a health bill when the husband gets sick.

But none of these things will increase their income or their household spending, which is what "reducing poverty" necessarily entails, according to his criteria. What would "reduce poverty" in this peculiar definition, would be charging more interest on these loans and services. So for example, if FaceBook, say, started making crypto-currency loans to poor people in India, and allowing them to send the money to each other over FaceBook, ....

What many people don't realise is that in a rural economy, a family with a cow has surplus milk that they can trade for other things like vegetables, so a small loan like this can be the basis for improving the local village economy without registering on the measure of poverty by income and household expenditure. In fact, it would soon reduce local household expenditure when people spent less on powdered milk bought from outside, which is something that they would need to earn money to pay for. See Julia Galef on the Intuition Behind Bayes' Theorem...

See Buffett, Gates and Munger on Eonomics, and Étienne de la Boétie on Slavery and Revolution.