Up to our knees in Climate Change in Bangladesh

August 27, 2009

Ugandan comic books; cash transfers in New York; praise for Jacob Zuma; Reaganite timewarp on healthcare reform and wonderful Magnum pics: links I liked

August 27, 2009

The great Microfinance debate: Comments on the Comments, some loose ends and some new info

August 27, 2009
empty image
empty image

Back from Bangladesh and still processing both the real life and blog discussions on microfinance institutions (MFIs), following last week’s post and the good debate in the comments. A few final (probably…) observations:

Microcredit v Microfinance: point taken. A lot of the doubts and criticisms apply to microcredit (loans), not to the wider range of financial services (insurance, savings etc) that MFIs sometimes provide. Non-loan financial services can avoid people getting into a debt trap, and be tailored more to their individual needs. The problem here (ref Nicholas Colloff) is that the business model skews efforts towards microcredit because it is more profitable. Interesting thoughts from Suyash on how the Hyderabad experiment could have been transformed if non-credit products had been part of the package, and what looks like a fascinating microsavings project from Oxfam America, which should generate some useful research in due course (when’s that due, Jeff?). I was a bit worried about 49% rates of return in Mali though – where does that money come from? If all the capital is generated internally and members lend to each other, it must come from net borrowers in any group, right? If so, what is the equity impact – do we risk creating a new class of community loan sharks? But I don’t fully agree with Milford Bateman’s critique of Saving for Change – doesn’t what it is doing resemble the credit coops that Milton praises in China? And the planned RCT can check whether the impact is greater than merely the injection of $x into the community via the scheme.

Interest rates and equity: microcredit providers may be better than loan sharks, but they still charge a lot of interest. I’m a bit baffled by the numbers on this, to be honest: the nominal rate charged by large MFIs is 10-15%, but according to the book by QK Khan mentioned in a subsequent post, the effective rate is about 2.7 times greater, mainly because the interest repayment applies over the whole year to the full value of the loan, whereas because repayments commence immediately, the average value of the loan over the year is half the initial amount (on a one year payback schedule). If this is true, MF borrowers are paying effective interest rates of 27-41% a year, far more than commercial bank rates. This would make credit a bit like clean water – poor people in shanty towns often pay several times more for it than rich people, driving up inequality, as Mukta points out.

Springboard or Safety Net? Sure, if MF provides a woman with a way to reduce her vulnerability eg a loan to buy a goat, that is a good thing, but wouldn’t it be better if she didn’t have to repay the loan? If MF is actually acting as a safety net rather than a means to finance would-be entrepreneurs, it starts to look more like poor quality social protection, as Kate Carroll argues.

Gender impact: MF advocates have simply seen money flowing to women as a ‘good thing’ (see this recent article in the New York Times for an example). They need to try a bit harder – the actual gender impact of MF appears more complex and interesting, but needs research.  The study by QK Khan found that only 10% of women fully controlled decisions on how the money was spent (but they still all bear the responsibility for repayment). Even more alarmingly, it found that one of MF’s unintended consequences is to aggravate the problem of dowry (the money, goods, or estate that a woman brings to her husband in marriage). 82% of the 2,500 women interviewed in Khan’s study reported that dowry had increased since enrolment. This would presumably (the book isn’t clear on this point) mean that the loan is wrapped up into the dowry payment, while the bride’s family is expected to meet the repayments. MF loans are in effect being diverted into boosting a discredited practice, rather than creating new businesses.

On the other hand, there is evidence that micro-credit stengthens bonds between women in borrowing groups, leads to reduced incidence of domestic violence and increases in community involvement. This reminds me of some of the research on the impact of fairtrade, which found that the main benefit is from its encouragement for cooperative action, rather than the direct benefits from a premium price for coffee, cocoa etc.

Fund-raising via MF – a conflict of interest?: In Bangladesh in particular, numerous NGOs now run MF businesses as a way to cross-subsidise their core costs and other activities, through what Mukta calls ‘primitve accumulation’. The danger is that financial dependence on the proceeds from microcredit will make it harder for them to take an objective view of the developmental pros and cons of microfinance. It could also mean (shades of the fairtrade movement, here ) that MFIs spend far more time making their business models work than using their clout to press for changes in public policy.

Reforming microcredit: Unlike some of the commenters, I am not inherently anti-bank, or for that matter, anti micro-finance – poor people need a range of financial services and they are not getting them. Sso I was interested in some of the good ideas for reform. These include, to quote Nicholas, ‘placing specific and regulatory obligations on interest, credit recovery, community reinvestment and community participation in governance.’ Requiring both banks and MFIs to publish accurate information on charges and real interest rates (as Russia has done) would be a good start. As would rebalancing microcredit with other microfinance products. The codes of conduct for MFIs that I’ve seen are incredibly vague (see here and here) – anyone got better examples? The best thing I’ve seen on this very superficial look at the issue comes from some MF industry leaders who, aware of these criticisms and concerns, met in 2008 and drew up the ‘Pocantico Declaration‘.

Some fair criticism of both the original blogpost, and the studies it reported. On the latter, Alan points out that the Poverty Action Lab study, as with most (all?) others, ignores the issue of whether setting up a business via MF actually destroys another business in the vicinity – it measures gross business creation, not net. And Suyash is absolutely right to say it’s very harsh to draw conclusions after only 18 months – let’s hope MIT has plans to return and continue the research in Hyderabad, but I can find no suggestion of this in their paper.

Finally, if, as Chris pointed out, these criticisms have been circulating for well over 10 years, how has microfinance retained its Teflon magic bullet status? This is where ideology trumps evidence, I think – it was just too good a fit with the neoliberal spirits of the age to let a little thing like evidence stand in the way. Might the Teflon start to peel off in these new Post-Washington Consensus times? President Obama’s recent award of the Medal of Freedom to Grameen Founder Muhammed Yunus and the appointment earlier this month of MFI Accion International’s President and CEO, Maria Otero, as US Under Secretary of State for Democracy and Global Affairs suggest that view is at best premature.

13 comments

  1. Just to correct things (not just the usual problem with my first name!). First, from what I know of Saving for Change (from Jeff and from googling) it is not like China’s RCCs at all. China’s RCCs were all about mobilising savings to invest in substantive local businesses with real growth and development prospects. Careful analysis went into deciding which ones (generally which Township and Village Enterprises, TVEs) to invest in, which technologies,what was minimum efficient scale to start at, and so on adn so forth. Saving for Change, so far as I can see, is about efficient (quick, secure, etc) savings mobilization for the poor, which is needed. But then thats basically it. The money is loaned out to tiny microprojects once more on a client-led basis as in other MFIs. So most still goes into petty trade, subsistence agriculture, selling mobile phone time on the street corner, etc. There is no real development beef planned so far as I can see, compared to the RCCs which planned to, and succeeded in underpinning China’s rural industrialization drive in the southern part of the country.

    Second, the RCT methodology does really not cater well to important spill-over effects. In fact, the RCT methodology may distort further the impact in favor of the chosen instrument – microfinance. Say you identify your control individual and see what happened to her without microfinance. She saw a loss of $5 a month in income. Turns out she’s working in a microenterprise too (a typical situation in developing countries) but she’s had it bad of late because some MFI has been pumping money in locally into new microenterprises doing exactly what she is already struggling to do, and so she has seen her turnover down and her income too. The RCT finds clients, meanwhile, have a gain of $5 from their new microbusiness. In the comparison with the client group, then, the difference between them looks like $10 and so quite large, making it look as though new clients have been doing fantastically better than non-clients thanks to microfinance. Microfinance must be really brill! In fact, things are actually worse. Local incomes generally fall because of the new supply and the poor community is worse off, although the new microenterprises might be better off than before. This is what the ILO is saying must not happen anymore! Average incomes in poor communities has been falling this last decade, largely because we are cramming them full with microenterprises. This is why the ILO says IFI plans to accelerate this microenterprise strategy now as a response to the current the financial crisis, are wrong. Also opportunity costs here – what else we could have done with that money? Maybe Chinese-style solid investment programs? All told, this link (spill-over) is not really registered by RCT. The problem arises because RCT methodology originated in medical trials where if I get a new cancer drug to try out but you get the placebo or nothing, any success I experience in curing cancer has absolutely no impact on your chances of being cured or not. So the RCT methodology in medicine is fine. But not in microfinance where local economies are in the short term finite and also causally inter-connected in ways that medical patients are not.

    Milford

  2. Duncan, I think you are spot on about ideology, trumping evidence. That was indeed what I was trying to illustrate by pointing to Ben Rogaly’s perceptive piece, prepared in the run up to the first Micro finance Summit in 1997.

    Another piece that I feel is also relevant to this debate is Naila Kabeer’s excellent paper ‘Money Can’t Buy Me Love? Re-evaluating Gender, Credit and Empowerment in Rural Bangladesh’ penned in 1998. see http://www.ntd.co.uk/idsbookshop/details.asp?id=401

    This explained very eloquently why it was possible to have a number of highly contradictory evaluations of the impact of credit on women’s lives, not least because most evaluations used externally determined notions of success – in part ideology based – and tended to ignore womens’ lived experience.

  3. Lovely, provacative posting, Duncan. I feel like there are a lot of exploratory thoughts here that are 80% right. A few points:
    * To think about microfinance coherently I think you need a strong understanding of how poor people manage money. A lot of this discussion seems to view the poor household like a black box: if we inject the microcredit intervention, what comes out? Being poor in a poor country means that your income is not only low but volatile and unpredictable and that you are exposed to more uninsured risks to life and limb . Thus the poor need intermediation—ways to put money aside, access it when needed–more than the rich. Savings, credit, insurance all help. Cf. Portfolios of the Poor, a key author of which, Stuart Rutherford, knows Bangladesh very well. But being poor also means that you can only afford low-quality services, such as group microcredit. So the best you can do is patch together a diverse portfolio of financial services—microsavings, moneylender, borrowing from your neighbors. Microfinance can contribute to that portfolio particularly by being among the most reliable of financial services available to the poor. Notice, there’s nothing here about microenterprise and rising out of poverty—that story is oversold and deserves skepticism. But to focus just on criticizing that story, like Bateman and Chang, misses important things.
    * If microfinance is a development fad, why has its appeal lasted so long?—“teflon,” as you say. I think it has to be acknowledged that millions of people are voluntarily partaking of it. That has to be taken seriously…though still with a dose of skepticism since the same can be said for cigarettes. I do take seriously worries about overborrowing. Commonsense says that credit pushing is dangerous.
    * A big question here is: if microfinance ought realistically to be seen as a way of building self-sustaining institutions to deliver poverty-ameliorating financial services to millions of people, does it deserve support? Or do other forms of aid do better? I’d be interested in your thoughts.
    * A fine CGAP study argues rather persuasively that microcredit rate are mostly reasonable and falling, with a few prominent exceptions. I blog it here.

    On that note, you might check my blog for critical thought on the impacts of microfinance. I’m blogging the process of writing a book on this topic.

  4. I want to share two comments in the light of this discussion and particularly the point that Duncan makes regards to people living in poverty need a range of services.

    In many contexts particularly in South Asia-the availability of hand loans or sometimes loans accessed by pawning whatever little assests provides immediate relief to the borrower irrespective of its wider implication.Rural hand-loans in that sense are the most flexible- no questions asked but cash provided upfront irrespective of time/place( Obviously very high interests too).There are incidences that such loans can transend generations and so the providers are also a standard and generational feature of the landscape( Rural and Urban- at times futur harvest could be given as a guarantee atvery low prices).These loans sometimes becomes so critical for instance money borrowed during the planting season to get food( Households will have least food at planting time). Financial institutions might see such loans as unproductive but more recently it is recoqnized that such consumtion loans are integral part of the production cycle.However such loans many times tend to accumulate because of non-payment or delayed payment and consequently it can sqeeze the loanee to extremes. It is one of the reason that such micro credits if available irrespective of the interest rates is accessed by poor people hurriedly because they try to pay- off the hand loans from micro credit they receive-the reason why no clear and visible results of positive spends might not be seen (Health and education for example).In this scenario- a massive appropriation of any surplus that is created by the people living in poverty is the standard feature. So, unless we are able to compleatly obsorb the historically accumalated loans -the micro credit will not provide us the positive outcomes we look for. Indian govt -Central and state has intervened in populistic ways to write -off such loans by law only making the people more vulverable to shocks becuase any alternate cannot substititute the simplicity, flexibility and accessibility of hand-loans.

    The other comment is that if the larger economy is insignificant in its vitality and not growing micro- credit becomes pocket-money and will literally wallows in that petty -insignificance of localized activity with very little larger economic significance.

  5. Microcredit vs. Microfinance

    Most of the MFIs now provide other financial services besides credit. With the exception of Grameen Bank, MFIs in Bangladesh cannot collect savings from non-members. So, it is regulatory hurdle that prevents them from promoting savings. On the other hand, Grameen Bank has ramped up savings collection from members and non-members alike, as allowed by the law. For every dollar that it lends out, it has $1.39 in savings. The interest rate offered on these savings are higher than the commercial rate. The savings are mobilized locally and reinvested locally. Most of the branches of the bank are self-sufficient and do not need to borrow from the Head Office. You can find more about Grameen’s savings mobilization efforts in my co-authored book, The Poor Always Pay Back: The Grameen II Story (sorry for the self-promotion).

    Interest Rate and Equity

    I do not know how Q K Ahmed is coming up with the effective interest rate numbers. For Grameen Bank, the effective interest rate is 20 percent. Grameen computes its interest rate on a declining balance. For example, for every 1000 Taka loan the borrowers pay 10 taka per installment. Comparing MFI interest rate with Commercial Bank rate is meaningless. If commercial bank’s gave loans to poor women, we would not need MFIs. You have to compare MFI interest rate with the next best alternative: interest rate charged by the money lenders. The average money lenders change 120 percent annually in Bangladesh (10% per month). In a remarkable new book, The Portfolios of the Poor, the authors pointed out that MFI clients value the credit they receive because of their predictability and repay that loan first before any other loan. To use a finance jargon, loans from MFIs has a ‘super senior status.’

    Gender Impact

    I agree with Duncan that the actual gender impact is more complex. The usual criticism is that microfinance does not lead to empowerment because women hand over the loan to their husband. This presupposes empowerment as binary: if you have the loan you are empowered and if you don’t you are oppressed. Empowerment is more complex. What matters is what happens to decision making in the household after receiving credit. There has been plenty of research on the empowerment issue and the best one is by Naila Kabeer, Money can’t buy me love’ from IDS Sussex. The critics do not realize that in a society where women has mobility restrictions, they have to use the help of the male in the household to execute an investment project. Take for example, if a women receives a loan for a irrigation pump, she will have to find clients most of whom are male. She cannot go to the local Bazar to find the clients. Her husband or the son has to find her clients.

    The point about dowry somewhat true. Despite sincere attempts by MFIs, dowry payment does not seem to go away. However, there is no empirical evidence that suggests that dowry payment has increased because of proliferation of microfinance. My guess is that the increase divorce rate is prompting the families to buy insurance in the form of dowry to guarantee happiness for their daughters.

    Fund-raising via MF – a conflict of interest?

    There is a complete firewall between Grameen’s Banking and for-profit operations. In other words, the profits from banking is reinvested in making more loans. Grameen’s long term plan is allow its borrowers to swap their shares in Grameen Bank for these for-profit companies. By the way, Grameen Borrowers received 100% and 20% dividends for their shares in the bank in 2006 and 2007 respectively.

    Reforming microcredit

    I agree that MFIs need to make sure the clients understand the full cost of borrowing from them. Another initiative to promote transparency http://www.mftransparency.org/. This initiative was launched by Chuck Waterfield and Muahmmad Yunus after the Compartamos IPO of 2007.

    RCT–the new fad?

    I think you missed the biggest criticism of RCT type study: the random exclusion of people. How can you justify not providing service to someone in the name of rigorous science. This sort of exclusion is possible in individualist society in the West. In a close knit society such as Bangladesh where everyone knows everyone else, it will be impossible to do such a study. Actually, you might get beaten up if you do that. More importantly, RCT are implemented by research institutions in rich countries using the clients of small NGOs who need donor money desperately. What does that say about power relations–a topic close to heart for Oxfam?

    QK Ahmed’s criticism of microfinance

    By the way, Dr. Ahmed’s criticism is not motivated by academic reasons only–it is a long story.

  6. In response to the last post , I think it’s unfair to critise RCT studies because some people are excluded. If you haven’t enough funds to include everyone , some one is bound to be excluded. What you are doing in RCT is funding the same number of people as before but evaluating more people to see the effect. Of course , you don’t want to spend large % of your funds in evaluation but seems only sensible to check outcomes as rigorously as we can. Western individualist donors and taxpayers would demand it of NGOs anyway even if in principle we didn’t want to do it.

  7. On a final personal note, Mr Obama’s mother was an employee of Women’s World Banking so microfinance credibility with the Obama administration has the maternal touch!

  8. In response to Ken’s argument, if the MFIs do not have sufficient funds to cover everyone, it need to focus on the poorest of the poor. They cannot randomly chose poor/nonpoor and poorest of the poor just to meet the scientific standard.

    Nic;

    Actually, Obama’s mother worked for Ford Foundation and in that capacity she worked with BRI in Indonesia and implemented microfinance project for Asian Development Bank in Gujranwala in Pakistan. Obama visited her when she was implemented credit project for women in Pakistan.

  9. In response to Bateman’s comment that “the RCT methodology does really not cater well to important spill-over effects,” the issue is not that the methodology itself does not cater well to them. RCTs can cater well to spillover effects if the unit of randomization is designed to to include the area where you will see spillovers. For example, with the Saving for Change RCT, the unit of analysis is village rather than individual, and spillover effects occuring within the village can be taken into account in the analysis. The bigger issue is being able to identify where you may see spillovers in order to build it into the design of the study. If effects are expected in communes or regions instead of villages, then the unit of randomization can be based on those instead.

  10. Hi – I don’t think it’s a unarguable fact that an MFI with limited funds has to focus on the poorest of the poor.

    What would happen if your $1000 of funds , could help 10,000 of the slightly less poor but only 100 of the poorest of the poor, what would be best ?

    We need to know the effect of our interventions to decide what is the best use of limited funds

  11. SECONDARY EFFECTS OF MICROCREDITS: A NOISELESS ADVOCACY INSTRUMENT

    I have realized that the name of microcredits includes several types of credits with common characteristics:
    · they are small amounts of money,
    · normally they are paid back in a huge percentage(97-98%),
    · given to women
    · with a high interest rate

    Apart from those details, they could be obtained by only one person straight from an official bank, or from special banks, or from an NGO………..or it could be decided by a group of people, trained previously for that purpose.
    I’m focussing my point in this last type of credit: a group of people (self help group), who has received education on this matter, and being part of a pyramid of groups, they decide which woman among them is receiving the credit, and they are, in a certain way, responsible for the payback of the loan.
    A lot of discussion is being held about the pros and cons of the microcredits, but I have hardly heard about the main virtues of these credits.
    I believe, with Duncan Green, that the only way to change the world is through the actions of active citizens.
    And I believe that these actions should be done where the problems are. I mean that protests in the North scarcely have results in the South, so the active citizens should exert pressure in their area, town, country, etc…..
    If you want to push the authorities to act , normally you need a group of people organized, and some leaders able to motivate and create the climax against the unfair situation.
    Coming back to microcredits, in my opinion, what will happen to them could be similar to Viagra. Viagra was designed to attenuate ischemic cardiomyopathy, but a very well known secondary effect turned it into one of the most famous medicines in the world.

    I think that a “secondary effect” of microcredits is the capacity to create a group of organized people, with several leaders in every pyramid, and able to fight against the unjustice they live in. They can become active citizens pressing local authorities to get changes in their political decisions.And that is, to me, the best profit of microcredits, although I haven’t seen comments pointing at it.
    We are trying to get active citizens, leaders, local activists in developing countries and we could be blind to see that we could receive a lot of help from the microcredit groups.

  12. Adding to Miguel’s point above, I think it’s important to consider that different microfinance programs are aimed at different audiences and can often be different from each other.

    I know one of BRAC’s findings was that microcredit was not a suitable mechanism for the poorest of the poor. Their “Challenging the Frontiers of Poverty Reduction” (CFPR) program takes a holistic approach by providing capital assets, training, a food stipend, savings options, healthcare and legal awareness programs, and doesn’t even include microcredit initially (after the program, participants are encouraged to join the mainstream microcredit options). It’s not sustainable in the microcredit-paying-for-itself sense (entirely donor-funded), but it’s an innovative approach that tries to overcome some of microcredit’s usual defects.

    Another excellent innovative microfinance initiative is SafeSave, led by Stuart Rutherford mentioned above, which provides savings opportunities to slum dwellers.

    Just a few thoughts…

Leave a comment