The verdict is in on microfinance
And it's not pretty. The results from the first large-scale randomized trial of access to microfinance indicate that it comes up short in many areas of human development. 52 of 104 slums in Hyderabad were randomly selected to receive new branches of a microfinance outfit called Spandana. Abhijit Banerjee and the other randomistas from the Poverty Action Lab describe the results in The Miracle of Microfinance? Evidence from a Randomized Evaluation:
...microcredit does have important effects on business outcomes and the composition of household expenditure. Moreover, these effects differ for different households, in a way consistent with the fact that a household wishing to start a new business must pay a fixed cost to do so. Existing business owners appear to use microcredit to expand their businesses: durables spending (i.e. investment) and business profits increase...
...While microcredit succeeds in affecting household expenditure and creating and expanding businesses, it appears to have no discernible effect on education, health, or womens' empowerment. Of course, after a longer time, when the investment impacts (may) have translated into higher total expenditure for more households, it is possible that impacts on education, health, or womens' empowerment would emerge. However, at least in the short-term (within 15-18 months), microcredit does not appear to be a recipe for changing education, health, or womens' decision-making.
I have no doubt these findings will provoke strong reactions in the microfinance community - I expect a deluge of rebuttals. There's one thing that's worth noting right now, though. Unfortunately, this particular trial may not go that far in ending the debate around microfinance.
As the authors note, the impact of microfinance in education, health, etc. may not appear for quite some time. In normal times, we would just wait another year or so and take another survey. However, the financial crisis means these are not normal times. If, in a year or two, a new set of results appears to indicate no impact on human development, the randomistas could claim this as a failure of microfinance. In turn, the microfinance proponents will claim that the financial crisis invalidates the findings, arguing that the ROI for small-scale entrepreneurs was abnormally low due to the crisis. Unfortunately, we'll simply have no way to know for sure who's right.
Update: I wanted to respond to the criticism I've received regarding the the title I used for this post. In fact, I contradict the notion that the "verdict is in" in my own discussion of the findings: "Unfortunately, this particular trial may not go that far in ending the debate around microfinance." I was gently poking fun at the randomistas who authored this study - although they do include provisos, there is a bit of triumphalism in places: "Randomized trials, by deliberately effacing all ex-ante differences between the treated and non-treated group, make it possible to compare like with like and extract the true impact of microcredit." Surely, it will take more than one trial to figure out "the true impact of microcredit" - for the reasons I gave, and the many reasons offered by the smart readers of the PSD blog.
Update II: The Economist has just come out with its own review of The Miracle of Microfinance? Overall, the magazine takes away a positive view from the microfinance study. Money quote:
Tiny loans are unlikely to be enough to allow these businesses to grow to an efficient scale, of course. But the role of microcredit in allowing people to signal their creditworthiness is valuable, especially if their success makes banks more willing to lend them larger sums and leads to even more economic activity. By being willing to take a risk on entrepreneurial sorts who lack any other way to start a business, microcredit may help reduce poverty in the long run, even if its short-run effects are negligible.
Comments (8)
Delicious
E-mail
Facebook
Follow us on Twitter
I've long believed that the biggest problem with the "debates" over microfinance is that too many proponents, as well as most skeptics, are expecting too much from microfinance. Not only is it not a silver bullet, it can't be expected to stimulate entrepreneurship, revolutionize health and education outcomes and upturn cultural patterns of women's roles in a society without a whole lot of other complementary stuff happening. Much of it is outside the control of those working on microfinance implementation.
But the study does find "quality of life" effects that expand individuals' options for investment and household expenditure. Surely "it doesn't suck as much to be poor" has to be an important positive outcome of widening access to financial services!
Posted by: nadezhda | May 23, 2009 3:17:55 PM
I have not read the study yet. But the results reported points out immediately to the problems inherent in this kind of study:their findings cannot be generalized. The fact that microfinance did not appear to show sizable improvements in human development in slums of Hydrabad does not mean that it is not causing improvements in other slums. More importantly, the major use of microfinance is happening in the rural area of South Asia. Slums are full of people who live in an environment of non-transparent property rights and low social capital. These factors might explain why families are not investing in human capital rather than the fact that microfinance is an unsuccessful intervention.
Posted by: Asif Dowla | May 23, 2009 6:15:18 PM
Higher household income doesn't necessarily mean higher spending on health, education, etc. It's pretty intuitive to assume that a spike in spending power will result in more expenditures on previously unobtainable luxury goods (that middle-class folks take for granted but are especially sought after by those who can't afford them). It's a rare breed of individual (socioeconomic conditions regardless) that reacts to a concrete spike in disposable income with a long-term outlook.
Microfinance addresses one half of the equation - the ability to obtain goods/services that improve quality of life. We need societal systems in place to tackle the other half - overarching values and goals - and incentivize individuals to have a longer-term mindset more favorable to achieving that ideal of high QOL.
Posted by: Harish Venkatesan | May 24, 2009 12:19:31 AM
Very interesting study and good to see the data, but the main problem I see is that not all microfinance is created equal. There are some microfinance organizations that are almost as bad as the predatory lenders, and others put a strong emphasis on improving health, education, and other aspects of borrowers' lives.
One study never answers all the questions.
Posted by: Malorye Allison | May 24, 2009 1:51:07 PM
I agree with "nadezhda" that it is important to define, and quite possibly limit, our expectations of microcredit (which is what the Banerjee et. al study was studying, not microfinance more broadly). Microcredit will not in itself improve education or access to healthcare, particularly in a 3-year span.
The fact that households are spending their money on more durable goods, and if they own a business, on their small business, is a positive finding in itself. If microcredit compels poorer households to spend more responsibly, these improved spending habits could be beneficial for their household over the years to come.
Posted by: Michael Chasnow | May 26, 2009 9:11:50 AM
I just blogged my own take on the new study: http://blogs.cgdev.org/open_book/2009/05/first-randomized-trial-of-microcredit.php. I think the PSD take is a bit too sweeping, though it is nice as a headline and opening sentence. Can you conclude from a one-year follow-up that microfinance does not work? Not that I feel defensive on behalf of microfinance practitioners. I am a skeptic by nature.
Posted by: David Roodman | May 29, 2009 3:02:35 PM
Interesting results, but I’m surprised that nothing is said about the inputs provided by Spandana other than its ‘microfinance’. Would we do a study on the impacts of ‘medicine’ or on ‘education’? No, we would look at the impact of a primary health clinic, a specific drug, or of maths classes for primary kids.
It amazes me how little understanding there is that there is no such thing as ‘microfinance’. Programmes differ in their design, implementation and management and therefore the outcomes and impacts differ.
Let me give one example from a recent visit to an MFI in Malawi that illustrates my point. I visited a number of group meetings at a group-based MFI. I attended one meeting run by a poorly performing field officer. Clients were late, there were repayment problems and the group told me that one person was absent who had not paid, and after the meeting they were going to go to her house to "make sure she pays".
Another group was performing well, with good attendance and smoothly running meetings. Again, one person was absent who hadn’t paid. The other clients informed me that she was ill, and again they said they were going to go to her house after the meeting, but this time to "find out how she is and see if there is anything we can do to help".
After the bad meeting I talked to four clients who had dropped out of the programme. One of them was resting due to the timing of the agriculture season. The other three had left because of financial problems and failed businesses. All three had sold major productive assets to repay their loans (a sewing machine, an ox cart and a goat) and were worse off because of their participation in the microfinance.
When I asked the good group about whether clients left and sold assets to repay loans, they thought hard and said that yes people left, but they did not know of a case where they had to sell assets.
The difference between groups? Solidarity and a supportive response to client problems. Perhaps also reflecting the input and way of working of the field officer and the MFIs training, operations policies and management supervision.
So we have one organisation, with a single methodology, but with different implementation and different levels of success in terms of building client solidarity. The impacts in these two groups will be very different, let alone compared to another MFI that has a different methodology, different products and services, different management systems.
MFIs do influence the impacts that they have. They do this through the design of their products, services and systems, and the day to day management of issues such as avoiding over-indebtedness, incentivising good client service, incentivising outreach to poorer or excluded clients, monitoring client-staff relationships and a host of other day to day activities. Some MFIs do this better than others.
Please let's have some rigorous impact assessments that pay as much attention to the inputs as well as the results. Then we will be able to start making meaningful verdicts on meaningful questions about microfinance.
Anton Simanowitz
Director, Imp-Act Consortium
www.Imp-Act.org
Join the SPM network www.spmconsortium.ning.com
Posted by: Anton Simanowitz | Jun 1, 2009 10:48:28 AM
Microfinance has two completely different markets and they should be analyzed separately.
If it is to help small entrepreneurs to get financing, at as good a rate as possible, in order for them to enter profitable ventures, then I am all for it.
But, if it is only to accelerate the consumption of the poor, by loans at rates higher than the risk free rate, I am against it since that will only make the poor poorer, and that is not what development is about.
Posted by: Per Kurowski | Jul 20, 2009 12:24:18 PM