Developing Jen

Two things I read recently have sparked my thinking on microfinance: this Economist article about whether microfinance reduces poverty, and this blog posting by Rachel Strohm, who I follow on twitter and works in microfinance.

Something about microfinance never sat quite right with me.  Sure, it’s impossible to argue that group lending models that allow credit to overcome adverse selection and moral hazard problems aren’t an important innovation in improving the lives of the poor.  Still, heralding it as the panacea to poverty always seemed a bit optimistic.  I’ve always wondered if somehow fostering micropreneurship (if that isn’t a word it should be so I’m sticking with it) through microfinance diverted funds to less efficient activities. 

Certainly it does.  Operating at the micro level inhibits specialization and doesn’t achieve scale, so by definition, there are more efficient ways to invest capital.  Often microfinance goes into helping the poor start businesses that don’t even require a single person’s full time effort.  So yes, inefficient.  Definitely.

But that’s not to say that microfinance doesn’t improve the lives of the poor, and reading recent articles has made me realize that we might not be looking at this quite right.  The development community is guilty of two misconceptions that I think are confusing the debate about microfinance: 1) that microfinance should be used for investment vs. consumption smoothing and 2) that by alleviating poverty we mean bringing people above the poverty line.

Let’s be clear on what investment vs. consumption smoothing mean.  Investing is obviously straight forward, loans go towards an activity that provides a return on capital, such as education.  Defining consumption smoothing is perhaps less obvious.  Theoretical economics draws the conclusion that optimizing intertemporal utility will lead to smoothing consumption over time.  If you know you’ll have income in the future, you’ll take out a loan to finance consumption now so that your consumption levels will be consistent over time.  People don’t actually do this to the extent the theory concludes, but that’s beside the point.   The point is that we take out loans to consume things that do not provide a return that we will pay for in the future.

Too often the debate focuses on whether the microfinance loans are used for consumption vs. investment.  The argument sounds like this: “consumption is bad, investment is good, because investment leads in income growth.”  But I would argue that if a microfinance loan is used for consumption smoothing, this still has an important impact on the lives of the poor: both in terms of alleviating poverty and preventing disinvestment during an income shock.  Let’s say I take out a microfinance loan to smooth consumption because of a bad crop year, which enables me to not go hungry, not take my kid out of school, or to send my sister to the doctor.  My future income may not grow but I'm able to borrow against it to improve my well-being through tough times and very importantly, prevent a future income decline through disinvesting to smooth consumption (i.e taking your child out of school).

Which brings me to the next problem with the microfinance debate. We’re focusing on income growth and poverty measurements instead of whether the lives of the poor are improving.  We’re confusing the end goal with the thing we’re measuring as a proxy for the end goal.  By focusing on the proxy, we’re missing the evidence of impact to the thing we’re trying to get at in the first place. 

If you haven’t already, read your Sen.   Economic growth is only a means to an end.  Hopefully my case for the value of consumption smoothing above makes it clear that microfinance can alleviate poverty even without investment that leads to growth.   

The other problem is that we’re overly focused on our poverty measurements, which are inherently flawed.  We shouldn’t forget about these flaws.  The recent JPAL and IPA studies using randomized evaluation and consumption surveys highlight this point.  We measure how poverty by conducting consumption surveys and defining poor and extremely poor in terms of consumption levels relative to some bar we set.  The percentage of the population in poverty is measured by how many fall above or below these levels.  The studies concluded that these poverty measures didn’t change due to microfinance.  But they did find evidence that consumption patterns shifted. 

That final finding should have been a wakeup call on the measurement issues.  If I’m only looking at aggregate consumption and not the consumption mix, I don’t differentiate between a household that spends 70% of its consumption on alcohol and one that spends 70% on food, shelter, and health care.  This is the flaw in our measurement that we mustn’t forget.  If studies show that microfinance leads to consumption shifts that improve the lives of the poor but does not raise overall consumption levels, poverty – the thing we are trying to get at with these consumption bars – has been alleviated.

I get the need for scientific evidence.  I studied physics in undergrad, after all.  I’d love to see a study showing that the percentage of people below the poverty line decreased due to microfinance.  But I think the current evidence and debate serves as a reminder to look behind the numbers.