June 29, 2009 — 07:15 am
Rural women driving their own change. Used under Creative Commons license.
Yes, Microfinance has Positive Effects on the Poor
Microfinance in its simplest form involves extending loans to a group of borrowers (usually called self help groups) who agree to help each other by means of group savings and informal support. The typical self help group consists of 10 to 20 people who meet regularly to discuss social issues and activities and deposit their savings in a joint bank account. Once enough savings have been accumulated, group members can apply to internal loans within the group or apply for loans through a commercial bank.
Even though microfinance is estimated to have directly reached 100 million customers in 2008 (for more details see my previous post) there is still plenty of debate about whether it has a significant impact on the lives of the poor or not. There are many good reasons why it may benefit or harm the people it tries to help. In theory, microfinance self help groups are better at allocating resources because it takes advantage of more comprehensive local information regarding local needs and lower monitoring costs, thus mitigating challenges like moral hazard and adverse selection. Therefore transferring resources to self help groups should result in local empowerment and efficiency. On the other hand, those resources devoted to the self help group may be appropriated by local elites. Additionally, if these resources are channeled through institutions parallel to local governments they may undermine rather than strengthen local capacity.
The problem with testing which side of the story is true is that researchers need very high quality data along a period of time. Most of the studies so far have depended on either incomplete or anecdotal information. As a result the debate has gone on for several years now and opponents of the microfinance approach could at least claim that no study has conclusively proved its positive effects on the poor. That was until now.
World Bank researchers, Klaus Deininger and Yanyan Lu have recently published 3 articles in which they take advantage of very detailed household data to find out what exactly are the effects of microfinance self help groups on the poor (see article 1, article 2 and article 3). Using a randomized experiment methodology (which I have advocated for BoP research in a previous post as well) they use survey data for years 2004 and 2006 comprising around 2,400 households and more than 3,000 loans in the Indian state of Andhra Pradesh.
The authors find that there are significant economic gains from program participation in the form of better nutrition, increased asset accumulation, higher levels of consumption and consumption smoothing. Apart from the economic impact, self help groups also have important social effects as reflected in the authors’ measures of female empowerment. Moreover, the authors find that for most of these cases, benefits often accrue to non-participants as well and thus microfinance self-help groups have important positive spillover effects.
Surprisingly, no significant effects where found with regards to the incomes of the poor. This may mean that either microfinance does not have any impact on income or, as the authors believe, that the hypothesized positive impact has not been able to materialize in such a short period of time due to the severe drought occurring in the area at that time which resulted in crop failures.
In spite of the authors’ best efforts these studies are still subject to criticism. The most important weakness is the fact that data is available only for two periods. This weakness is palliated by using a randomized experiment methodology. Nonetheless, the effect of self-help groups is not fully rooted after one period and so one may argue that negative effects have yet to kick in or that the study capitalizes on chance. Moreover, since the survey data is quite localized it may be the case that such positive findings are dependent on some specific local factor in Andhra Pradesh or in the microfinance program in which the survey was conducted.
Although all these criticisms are valid, considering how widespread and standardized microfinance endeavors currently are, it is hard to believe that one particular detail which can not be extrapolated to other microfinance programs may be driving the results. Basically and in my opinion, thanks to these studies the debate about the impact of microfinance on the poor is laid to rest.
So, if you find yourself engaged in a debate about the usefulness of microfinance, from now onwards let the speaker know that yes microfinance has been proven to have a significantly positive economic (consumption, nutrition and asset accumulation) and social (female empowerment) impact on the of the poor and that this positive effect is very likely to be felt by at least 100 million people. (And if they still have doubts direct them to this post!).
7479 Views
Comments
lakshmi
Jun 29, 2009
finance for the poor
Dear Manuel it's good to know about the microfinance for the poor and mentioning about this from Andhra Pradesh state from your article
I know about this scheme and in Andhra Pradesh it's named as "Dwakra" mahila sangham or a group formed by the ladies in and around the village and as per I know this was a good scheme and many of them found it's benefits also
And the other thing from your article was not having the proper data or the records not available ?
Manuel I think that creating the data base might be a problem previously but now we can do that in a better way with more flexbility
The other thing rearding microfinance for the poor this is meant for the people who live less than $2 per day ?as per BoP
Microfinance may be used as a mode for the upliftment of the people in the rural area ..
— Manuel Bueno replied over one year ago
Dear Lakshmi, The "problem" (in my opinion) with the data is that the authors use household surveys for years 2004 and 2006. I was wishing that they would also have surveys for several other years to track in more detail the impact of microfinance. At any rate, coming across this type of very detailed data is quite hard, as you will know, so it is a weakness which is hard to fix. Best, Manuel
Rob Katz
Jun 29, 2009
Microfinance reduces poverty?
Manuel,
This post is timely because a colleague of mine literally just sent out a note to the team with a link to this blog post by David Roodman:
http://blogs.cgdev.org/open_book/2009/06/roodman-morduch-2009.php
In it, Roodman argues that there is actually no proven link between microfinance and poverty reduction! So interesting to have both sides of the story represented.
Can you take a look at Roodman's arguments and tell us - as readers - what he is saying that these articles you cite above are not? And vice versa?
Thanks for bringing up an important topic here.
Manuel Bueno
Jun 29, 2009
Re: Microfinance reduces poverty?
Dear Rob,
Thanks a lot for the link. It is indeed a nice coincidence. Let me briefly start with what both studies have in common.
One of Roodman and Morduch’s statements alludes to the possible relationship between microcredits and income. This does not contradict with the studies I quoted above. In fact, Deininger and Lu openly admit that no relationship was found between income and microfinance and argue that it may be because this relationship could arise in a longer time-span.
The apparent contradiction lies in the causal relationship between microcredit and consumption. Deininger and Lu argue that this relationship is positive. Roodman and Morduch replicate previous studies and find two things:
1) These studies have not found a causal relationship between microcredit and consumption. Some may have claimed to have done so, but Roodman and Morduch when trying to replicate such results have found no statistical evidence for their claims.
2) Such a causal relationship has not been found because the methodology employed was not able to rule out reverse causation. Reverse causation (or endogeneity) is the econometric problem one encounters when it is not clear what causes what. In this case it may be debatable whether increased consumption allows for more microcredits or more microcredits stimulate greater consumption, since both possibilities are plausible. This is important because in the first case, such a causal relationship would amount to elite appropriation, which is not the point of microfinance efforts.
The problem of reverse causation is a very serious one and, as the authors state, it throws into question the results found. However, Deininger and Lu use a randomized experiment to unravel reverse causation (for more details about why this methodology solves this problem go to my previous post entitled "Moving Beyond Case Studies: Randomized Experiments"). By conducting such an “experiment” they need not concern themselves with this problem and so the results found are more robust.
Having said this, I keenly await Roodman and Morduch’s comments and criticisms about Deininger and Lu’s studies, as well as possible weaknesses that I may have missed out. Moreover, I would love to hear additional input from other readers as well.
Best,
Manuel
Meredith
Jun 30, 2009
There will always be a critic
No matter what anyone does, or their intentions, there will always be a critic. I think the great thing about micro-finance is that it has brought the issue of poverty into the world of people who can afford to help - investors. Before micro-finance, the best way to help was through volunteering or donating funds to a charity. These methods might not have resonated with everyone. Now there's a new way to help - give a loan to person and need, and maybe get a return on the loan. Whether or not micro-finance is THE answer to alleviating poverty, it's an answer. And it's working to a certain extent. It's a step in the right direction, regardless of what the critics say.
demianus nawipa
Jun 30, 2009
sejarah terbentuknya pulau papua
http://geopapuastudent.wordpress.com/
Rj
Jun 30, 2009
...are these positives sustainable?
Thanks for bringing up a very relevant topic up here.
I would agree to what you have to say on improved consumption but question mark comes if we have to think of long term positive impacts of Micro loans. Most of these loans goes into consumption and not towards productive activity. Hence, although better way of life, it is not contributing to increase in sustainable income.
— Manuel Bueno replied over one year ago
You make an important point - sustainability is something we should keep an eye on. The problem here is that to measure whether the impact of these initiatives are sustainable we need data for a long period of time. This could be inordinately hard since it is still not clear whether in the short run the effect is significantly positive.
Saurabh
Jun 30, 2009
Positive, perhaps unintended effects
Hi Manuel,
An excellent post! I have not yet read all 3 articles, though I have heard a lot about this debate, which promises to be a very interesting one, as further research is developed. What I find interesting is that the research suggests microfinance has a positive impact on smoothing consumption patterns and essentially helping the poor through difficult times, rather than necessarily having a positive impact on income. While some economists may argue that the ultimate goal of microfinance is to raise people out of poverty, I would think that the issue of poverty and income levels is simply too complex, with too many variables, to be effectively influenced by one approach like microfinance, and any study will suffer from a number of issues like time periods and omitted variables.
This doesn't mean that microfinance doesn't have important positive effects, and in fact, helping the poor through difficult times by smoothing their consumption is an important factor in poverty alleviation. I think it would be interesting to do further research using using variables like consumption patterns and nutrition that are components of poverty, and that can be more directly related to a microfinance approach, rather than the ultimate outcome of alleviating poverty.
A similar debate occurs with the impact of SMEs on economic growth, with several studies that find a correlation between the size of a country's SME sector and its GDP, but cannot find a causal relationship. Again, I think the factors that can affect a variable as complex as GDP are too many to effectively control for, and any study will always suffer from omitted variables. It may be better for researchers to actually study effects that are more direct at first.
David Roodman
Jun 30, 2009
Actually, the studies are not randomized
Manuel, unfortunately these studies are *not* actually randomized. I think a couple of sentences in one of the studies are confusing on this point, making them seem randomized. I explain more on my blog:
http://blogs.cgdev.org/open_book/2009/06/new-impact-studies-of-indian-self-help-groups.php.
Manuel Bueno
Jun 30, 2009
Re: Randomization of studies and other comments
Dear David,
Thank you for your comment and your views on these articles. As you say, the study the authors conduct is not fully randomized (those of our readers who want to get into the detailed reasons why it is not should go to David's post). In this line, I must admit that I bypassed a core weakness of the article.
Nonetheless, this does not mean that we should outright reject the conclusions it finds. Although we may argue that the study is not fully randomized, there is randomization to a certain extent (which would in turn help to fix the reverse causation problem you point). Moreover, the methodology the authors use try to overcome the non-randomized should palliate the remaining problem (whether they effectively do so is a matter of debate).
Therefore, I agree with you that, given the weakness that you point out, the debate about the effects of microfinance on poverty is still open, though maybe a step closer to resolution after these studies.
Best,
Manuel
— Manuel Bueno replied over one year ago
Sorry, there is a typo in the last sentence of the second paragraph. It should read: "Moreover, the methodology the authors use to try to overcome the remaining non-randomization should palliate this problem (whether they effectively do so is open to debate)."
Mike Gechter
Jul 1, 2009
Consumption smoothing
Manuel,
Great post highlighting microfinance's proven ability to provide a consumption smoothing to the poor in the developing world. With a credit market outside of moneylending, the poor in the developing world can now enjoy the benefits of stable consumption brought to people in the developed world through options such as credit cards.
However, the claim that microfinance alone leads to human development (women's empowerment, education, consumption, etc) doesn't follow from the studies you cite. The self-help groups that Deininger and Lu are investigating, in their own words, do "not exclusively focus on credit or savings but also [include] emphasis on social empowerment, outreach, and capacity building." So, there are at least three interventions included in the SHGs: a microfinance intervention, a financial literacy one, and a community organizing one. Any of the interventions could be responsible for the positive outcomes measured by Deininger and Lu. The idea that one of the other interventions is causing the positive outcomes in human development could also explain the fact that these studies clash with the results of others (by MIT's Poverty Action Lab, for example).
Until we have a study showing a pure microfinance program’s leading to human development, I don’t think we can attribute positive outcomes to microcredit beyond consumption smoothing (which is certainly very important to highlight!).
Mike
Amit
Jul 2, 2009
Other factors at play?
Thanks Manuel for this post and the interesting debate that has ensued. It was particularly helpful to get David Roodman's insights and read about his research as well.
Reading on his blog, I find myself concurring with his conclusion that:
"I am not saying that microcredit, much less microfinance as a whole, is bad for poor people. For me, the take-home lesson is that social scientists and promoters of social programs respond to incentives to overestimate and exaggerate the power of mathematics to enlighten us about causality in social systems. Math does not substitute for wiser reflections on the nature of development and how financial services can contribute to it."
I think this is really a great lesson with respect to many of these studies. Also, I may be off base here (please tell me i am if so!), but it appears to me that there are many complex factors and variables that I don't see particularly being taken into account. For instance, the Pitt and Khandker study in Bangladesh, as well as the Deininger and Lu study in Andhra Pradesh, both seem highly localized in geography and in the specific program circumstances that they examined.
Manuel mentions this:
"Moreover, since the survey data is quite localized it may be the case that such positive findings are dependent on some specific local factor in Andhra Pradesh or in the microfinance program in which the survey was conducted."
But I think this criticism can be applied even more broadly regarding other localized issues and factors, such as the self help program vs. other mf or mc programs around the world, the type of mf or mc institution involved (for profit, nonprofit, etc.), where the institutions resources come from and what relationships are created in their distribution?
I think these and other issues are important to parse out if one is to make any type of generic statement about the overall effects of microfinance. Even more important perhaps, is that rather than simply saying whether microfinance works or not, we may get a better understanding of what may be working and what not, and be able to discover specific conditions under which it can work in the future, discovering lessons for innovation in this critically important field.
Thanks to all for the discussion!
Manuel Bueno
Jul 2, 2009
Re: consumption smoothing
Dear Mike,
I think you make a very important point in your comment - and one I was pondering about when writing the post. In my opinion the issue here is how to define microfinance.
Microfinance is about offering financial services to the BoP. However, within the broad microfinance business model there are certain activities that, in my opinion, make this approach work and become scalable. Some of these activities, as you mention, are related with social empowerment, outreach, and capacity building. These activities are not directly involved with the provision of financial services and hence, strictly speaking, we may not consider them as belonging to microfinance.
The problem here is that these activities are the ones that make the whole microfinance business model work. That is, one may argue that they are inextricable components of what makes microfinance profitable. Thus, to try to separate them and see their effects individually may lead us to the wrong conclusion. In this line, microfinance can also be seen as the sum of several activities only one of which is the provision of financial services. At the very core you would have the financial services alone. In the periphery you may attach essential activities without which the core activities can not be profitable.
At the end of the day we may not find any microfinance operations without the adjunct activities mentioned above (social empowerment, outreach, and capacity building), basically because they are not profitable. Thus regardless of where we draw the boundaries defining microfinance, we would not be able to analyze the impact of such a narrow definition of microfinance, because such firms do not exist. On the other hand, the risk here is that the definition of microfinance could become increasingly diffused and blurred.
Finally, you also mention that any of the interventions (a microfinance intervention, a financial literacy one, and a community organizing one) could be responsible for the positive outcomes measured by Deininger and Lu. I wholeheartedly agree – nonetheless, I suspect that the cause of the outcomes does not lie within one intervention alone, but within the interaction effect between these three interventions together.
— Mike Gechter replied over one year ago
Dear Manuel, my concern is trying to extricate which interventions are yielding which results. We've seen a number of studies yield different results. The difference in the results may come from different populations, but they may also come from different services provided. In that sense, I think we need to find out which services provide the interaction effects that we want to see, as you say, and endorse those as best practice.
David Lehr
Jul 8, 2009
Necessity Entrepreneurs
Thank you for an informative post and for keeping the conversation going through the comments!
I think the debate on effectiveness is not yet over, and that at least part of the reason is that microfinance, or more accurately microcredit, has as a fundamental premise that there are entrepreneurs in low income communities that could unleash their entrepreneurial abilities if only they had access to capital.
While some people do actually want to design and implement new businesses, I believe that most hardworking and intelligent self-employed poor are actually reluctant or “necessity” entrepreneurs who would rather work in an existing business than create one of their own. Given the limited choices and barriers to success, most new businesses started by micro-entrepreneurs have the odds stacked against them.
This does not negate all the other positive effects of microfinance on savings, cash smoothing, and asset creation, but it does call out for other solutions to the employment quandary (did I hear someone say microfranchising?).
A.JohnPeter
Jul 19, 2009
Impact of micro-finance.
Micro-credit does create positive effects on the poor
if this programe is implemented 'as the poor centered".There has been no serious effort to study the impact.Greater importance is given to study to ensure 100% recovery.The same level of importance is not given to study the various positive impacts.
A.JohnPeter.
Micro-finance practisioner .
email:ceoanisha@gmail.com
nasir hussain alvi
Jul 27, 2009
we agree with article
we also want to start microfinancing in our area can you guide us
You are commenting on a reply.
To add a new comment, submit or cancel your comment reply.
Comment on article: Yes, Microfinance has Positive Effects on the Poor