microfinance insights vol.12: teaser

InvestmentBalancing Social and Financial

CritiqueIt’s No Magic Bullet

GenderThe Women’s Empowerment Myth

IndexSocial Impact Tools

Features

A bimonthly publication from Intellecap www.microfinanceinsights.com

Cover Price: US$12/EUR 8/INR 75

TM

Evaluating the Impact of Microfinance Panacea or Placebo?

Vol.12, May/June 2009

Proudly presents the Inaugural Issue of Beyond Profit

A global magazine that presents Ideasand People from the Social Enterprisesector. Through Beyond Profit, Intellecapsupports new approaches and models,and gives unique insights into a marketthat has the potential to draw millions ofpeople out of poverty.

Intellecap is a globally recognized social investment advisory firmthat catalyzes the growth of for-profit social businesses and createsan ecosystem that nurtures social entrepreneurship worldwide.

www.intellecap.net

Impact Investing

Bold Models to Drive Development at Scale

Can We Teach

Social Entrepreneurship?

10 New Ideas that Could

Change the World

Social Enterprise. Ideas. People.

May/June 2009

An Intellecap Publication

Shaping For-Profit Enterprises

Through Disruptive Innovation

www.beyondpro

fitmag.com

Complimentary CopyInaugural Issue

1cherryred.indd 1

23/04/2009 2:39:45 PM

Visit www.beyondprofitmag.com before June 1 to sign up for a free subscription.

Proudly presents the Inaugural Issue of Beyond Profit

5may/june 2009 l microfinance insights lwww.microfinanceinsights.com 5

l vOluME 12 l MAy/JunE 2009

In our cover story, Sam Dailey Harris, Director of the Microcredit Sum-mit Campaign, takes on the quantity versus quality debate: does the number of clients reached matter more than the poverty reduced? While microfinance has often met with criticism, the fact that it is crucial to breaking the poverty cycle is too apparent to be ignored.

Regular Featuresnews Board 8

Commentary 10A Double Bottom Line Vision for the FutureNigel Bigger and Jennifer Meehan

Global viewpoints 18Perspectives from Azerbaijan, the Philippines, Sudan and more

Centerfold 29An Index of Social Impact Measuring and Reporting Tools

Resources 50Recommended Readings and Calendar of Events

Trends 52Indicators on Global Poverty and the Microfinance Rating Industry

Survey 54Microfinance and Social Impact

Books 56Book Excerpt and Book Review

last Word 58A Dose of Realism: Tempering ExpectationsVineet Rai

Interview The Proof is in the People: Impacting Women through Relationships and Trust 15 Ela Bhatt

Critique It’s No Magic Bullet: Why Microfinance is Unlikely to Move People out of Poverty 22 Thomas Dichter

Research Opening the Black Box: How the Poor Use Credit in India 24 Alexandra Kobishyn

Randomized Experiments in Microfinance 27 Jonathan Bauchet and Aparna Dalal

Basics Social Performance Management in Practice 33Frances Sinha and Ragini Chaudhury

Investment The Challenge of Duality: 35An Investor’s Perspective on Balancing Social and FinancialGoalsHugo Couderé

An Equity Investor Takes Stock: Integrating Social 37Performance Measurement and Microfinance InvestmentPaul DiLeo

HousingMortar and Mortgage: Low Income Housing Solutions for 39Urban IndiaAffordable Housing Team, Monitor Group

Human ResourcesAltruism Reloaded: Seeking Work with Impact 41Sarika Bansal

StudentsCredit for Credits: Student Groups Get Involved inMicrofinance 44Ranjit Koshi

ResultsElusive Impact or Concrete Change: How Social 45Performance Assessment is Helping to Change the Face of Reporting in MicrofinanceMicol Guarneri

Back to Basics: Getting on Track after Mission Drift 47Cecilia del Castillo

GenderThe Women’s Empowerment Myth 48Radhika Desai

TM

Mission (Im)possible:Can Microfinance Really Change the lives of the Masses?

12Cover StoryCredit: Russ Bowling

7may/june 2009 l microfinance insights lwww.microfinanceinsights.com

Dear Reader,

What is that quote about that incorrigible beast with a powerful bite? “If you think you’re too small to be effective, you have never been in bed with a mosquito.” As several of my mosquito-hating friends would tell you, small interventions (positive or negative) can have a big impact.

Microfinance is based on this premise: a little slice of something, like credit, can go a long way. Microfinance has brought women around the world together in harmonious groups. Transformed low-income men and women into wealth-creating entrepreneurs that control their own destiny. Put a roof over the heads of the homeless. Or at least that is what we’ve always thought. Did anyone bother to check? No one usually does when something is so small. However, in the past several years, as the sector has grown and matured, it has been subject to more scrutiny. Now, it is no longer a given that micro-finance is a salve to the poverty stricken, nor can it be reliably thought of as a steroid to boost a borrower out of poverty. What we had hoped was a cure, may be merely palliative or quite possibly only a placebo.

And that is what we chose to explore in this issue. Are women truly being empowered through the provision of credit? Because access to money is not in and of itself empow-ering, is it?

Can we optimize financial return as we maximize social return? How do we balance the two, and are they mutually exclusive?

If microfinance is indeed making an impact, what is the best way to measure it, and who should do so?

If we provide 100 million people with microcredit, does it matter if none of them move out of poverty?

Should we blame MFIs for not achieving poverty reduction, when really, their goal has initially been to create an environment of inclusivity? Or did these goals change along the way?

Last week, I had a meeting with a social investor that struck me as a sign of the times. His firm, based out of Norway, has invested several million dollars in poverty inter-ventions—from a solar lighting company to an NGO in India that is bringing women together in self help groups and giving them business skills. I use the word “invested” loosely, because his company is actually doling out grants. But, this firm views it as investing, because they are fastidious about calculating their ROI—in terms of jobs cre-ated, children educated, healthier lives lived, etc. However, the investor is hesitant about microfinance, because, he said, “With microfinance, we know the repayment rate, but we don’t know what value is created for the final recipient. We just don’t know the social impact of microfinance.”

In this issue, we explore this conundrum. Is microfinance making a difference? We leave it to you to be the final judge.

Enjoy the issue,Lindsay Clinton

Managing Editor

Managing EditorLindsay Clinton

Editorial TeamAparajita Agrawal, Ranjit Koshi, Asako Matsukawa, Vibha Mehta

Advisory BoardVineet Rai, Wim van der Beek

Cover & Page DesignToonPillz

Cover PhotographHerval Freire

For editorial, contributions,subscriptions, advertisements and

other queries, please contact:Microfinance Insights

C/o. Intellectual Capital Advisory Services Pvt. Ltd (Intellecap)

512, Palm Spring, Beside D-Mart,Link Road, Malad (W),Mumbai 400 064, India

Tel: 91-22-40359222, Fax: 91-22-28801572

DisclaimerThe views and opinions expressed

herein by authors are not necessarilythose of Microfinance Insights magazine,

its Staff or its Editor, and they assumeno responsibility for them.

Printed and published byIntellectual Capital AdvisoryServices Pvt. Ltd. (Intellecap)

512, Palm Spring, Beside D-Mart,Link Road, Malad (W),Mumbai 400 064, India

Printed at Bell Graphics,

Lower Parel, Mumbai

© Intellectual Capital Advisory Services Pvt. Ltd. All rights reserved throughoutthe world. Reproduction in any manner

without permission is prohibited.

Cover Price (Within India): INR 75 Cover Price (Outside India): US$12/EUR 8

For more currencies and subscription details,visit www.micofinanceinsights.com

vol. 7, July 2008

From the Editor

TM

Panacea or Placebo?

12 l microfinance insights l may/june 2009

cOver stOry

In 1996, I traveled the world speaking about the first Microcredit Summit that was to be

held the following year in Washington, DC. I remember a particularly difficult briefing for development specialists in London that year that played a key role in setting the course for the Microcredit Summit Campaign on the is-sue of social progress. During that rough and tumble session I laid out the Summit’s goal to reach 100 million of the world’s poorest fami-lies, especially the women of those families, with credit for self-employment and other fi-nancial and business services by 2005. I also discussed the Summit’s three core themes: 1) reaching the poorest families, 2) reaching and empowering women, and 3) building finan-cially self-sufficient institutions. The three core themes would soon grow to four, based

on the heated discussion that day. People in the group were upset about a

number of things, including concerns that 1) microfinance was too business-like and did not adhere to the proud charitable traditions of the UK, 2) microfinance would become the new darling of funders and push out much-needed support for health and education, and 3) the goal of reaching 100 million of the world’s poorest families was too outlandish (read “too American”). One academic paper out of the UK that year talked about the “hard selling of a new anti-poverty formula” by the “microfinance evangelists.”

But the criticism that rang truest was this one, “What difference does it make if you reach the poorest families, you reach women, and the microfinance institutions (MFIs) are

financially sound, if there is little or no im-provement in the lives of the clients?” I took this omission to heart and back to the Sum-mit’s Organizing Committee. A quick con-sensus emerged, and we added a fourth core theme - ensuring a positive measurable impact on the lives of the clients and their families.

With the Microcredit Summit Campaign’s announcement in January 2009 that we had surpassed our goal of reaching 100 million of the world’s poorest families with microloans in 20071, existential questions arose about the “quantity versus quality” debate. That is, does the number of clients reached matter more than the poverty reduced, is microfi-nance really having a social impact, and how to respond to critics who say microfinance is not working. Answering the Tough QuestionsI believe these questions are all interrelated. Let me start with my own personal heresy on the issue of microfinance impact.

Do I believe that microfinance alone can end a family’s poverty? Absolutely! I have seen it and probably many readers have as well. Take the case of Sufia Khatum, the Bangla-deshi stool maker from Jobra village (near Chittagong) who received a loan of less than US$1 in the mid-1970s from a young econom-ics professor named Muhammad Yunus. Her income immediately jumped from two cents a day to US$1.25 a day as a result of this loan. But a string of simple anecdotes like this can be dismissed without concurring evidence from respected researchers.

This forces me to ask a series of more im-portant questions. Do I believe that microfi-nance alone can always end a family’s poverty? Absolutely not! Do I believe that we should have to demonstrate that microfinance alone ends poverty for the international community to redouble its efforts to extend its reach and impact? Absolutely not! Poverty is much too

Mission (Im)possible: Can Microfinance really Change the Lives of the Masses? Microfinance has its share of skeptics who criticize it for not being the panacea for poverty. While microfinance cannot always remove poverty, it is, if properly implemented, a powerful tool to help the poor begin their climb out of poverty. Proponents, for their part, focus on outreach and sustainability as indicators of success. The author, Sam Daley-Harris, Director of the Microcredit Summit Campaign, adds one more dimension to define success in microfinance – poverty reduction. It is, after all, the raison d’etre for the sector.

Credit: Russ Bowling


complex to always be cured by one interven-tion. We do not ask if basic education alone ends poverty. We do not ask if vaccinations alone end poverty. So why should microfi-nance be held to the standard of reducing poverty alone?

This logically leads to a third question. Do I believe that if we cannot demonstrate that microfinance alone ends poverty, we should denigrate its importance as a financial and anti-poverty intervention? Absolutely not! When implemented well and focused on the poor, it is one of the most powerful tools we have to help a family begin a dignified route out of poverty – something that dozens of studies confirm.2

Even though I do not believe that microfinance alone can always end poverty, nor should we expect it to do so, I do believe that MFIs should find cost-effective ways to measure social progress in order to create a sense of accountability and set performance standards. Just as any credible MFI would track its financial health; it should also track the social progress of its clients if poverty reduction is one of its stated objectives. This discussion brings us to the “quantity versus quality” debate—does the number of clients reached matter more than the poverty reduced? I believe that they both matter, but “poverty reduction” matters more.

Who is Poor?As a result, one of the Microcredit Summit Campaign’s two goals for 2015 is to ensure that 100 million families rise above the US$1 a day threshold. The Campaign has encour-aged practitioners to use tools like the progress out of poverty index (PPI),3 a cost-effective way to determine whether clients move above the US$1 a day marker. Even simpler tools are available to track social progress, tools that do not link to an absolute measure such as US$1 a day. Grameen Bank’s 10 poverty indicators make tremendous sense for clients in Bangla-desh, and similar indicators can be selected for other countries. Sample their simplicity:A member is considered to have moved out of poverty if her family fulfills the following

criteria:The family lives in a house worth at least •Taka 25,000 (US$370) or a house with a tin roof, and each member of the family is able to sleep on a bed instead of on the floor. Family members drink pure water from •tube-wells, boiled water or water purified by using alum, arsenic-free, purifying tablets or pitcher filters.All children in the family over six years of •age are going to school or have finished pri-mary school.Minimum weekly loan installment of the •borrower is Taka 200 (US$3) or more.Family uses sanitary latrine.•Family has adequate clothing for everyday •use, warm clothing for winter, and mosquito-nets to protect them from mosquitoes.Family has sources of additional income, •such as a vegetable garden, fruit-bearing trees etc., so that they are able to fall back on these sources of income if needed.The borrower maintains an average annual •balance of Taka 5,000 (US$75) in her savings accounts.Family experiences no difficulty in having •three square meals a day throughout the year, i.e. no member of the family goes hungry any time of the year.Family can take care of health needs. If any •member of the family falls ill, the family can afford to seek adequate healthcare.For other countries, these indicators would

need to be adapted to local conditions by those who understand them.

Is Microfinance in Ship Shape?What about those who say that microfinance

is not working? Over a recent four week pe-riod I have felt surrounded by some of those critics. One suggested, in an online debate, that MFIs rarely serve the very poor and an-other suggested there is very little change in clients’ lives. Let me begin by addressing the first charge. While I believe that many MFIs do serve those who earn less than US$1/day, I would like to see more do so.

From the very start, the Microcredit Sum-mit Campaign took strong measures to work against that development tendency –not unique to microfinance— of directing re-sources primarily to the better-off poor and non-poor. The original Microcredit Summit goal was established to reach 100 million of the world’s poorest families precisely because of that failure of international development, a failure that begins with the bilateral and mul-tilateral aid agencies. Clearly, by reaching that goal the community of microfinance practi-tioners, especially in Asia, have given some priority to reaching the very poor – particu-larly since 80% of the MFIs surveyed had their outreach to this group verified by a third party. But more can and must be done.

That is why another of the organizations I founded, RESULTS, a citizens’ lobby com-mitted to creating the political will to end

“Just as any credible MFI would track its financial health; it should also track the social progress of its clients if poverty reduction is one of its stated objectives.”

“RESULTS has encouraged more than 1,500 members of parlia-ments from around the world to push successive World Bank Presi-dents to get more Bank microfi-nance funds to families living on less than US$1 a day, a clear area of failure for the World Bank.”

cOver stOry

Critics charge that microfinance is for the economically active poor. But isn’t every family economically active?


global poverty, successfully lobbied to get a law passed requiring that half of US Agency for International Development (USAID) mi-crofinance resources go to families living on less than US$1 a day. How is USAID doing in fulfilling this requirement? Not well at all.4

This is also why RESULTS has encouraged more than 1,500 members of parliaments from around the world to push successive World Bank Presidents (Wolfensohn, Wolfowitz and Zoellick) to get more Bank microfinance funds to families living on less than US$1 a day, a clear area of failure for the World Bank.

So, do many MFIs reach the ‘upper poor’ in much greater numbers than the ‘very poor’? Probably, but not the largest MFIs in Asia and that is not the point anyway. I assume that one could find 1,000 MFIs with an average of 5,000 clients each and find that “most [of those] MFIs reach the upper poor.” But of what use is that information if BRAC Bangladesh, for example, has one million more clients than all of those 1,000 MFIs put together and is contributing (along with Grameen Bank and other MFIs in Bangladesh) to a profound reduction in that country’s levels of poverty? Should we not be focused on those successfully leading the way and push the others to reach that level of performance?

In his book The End of Poverty, Jeffrey Sachs describes a visit with BRAC microcredit clients where he learns that the women had, or planned to have, no more than two children each. He wrote: “Perhaps more amazing than the stories of how microfinance was fueling small-scale businesses, were the women’s attitudes to child rearing...Here was a group where the average number of children for these mothers was between one and two children...This social norm was new, a demonstration of a change of outlook and possibility so dramatic that the late Dr. Allan Rosenfield [the Dean of the Columbia University School of Public Health at the time] dwelt on it throughout the rest of his visit . . . he remembered vividly the days when Bangladeshi rural women would

typically have had six or seven children [not one or two].5

This is a profound change. Of course, BRAC was not solely responsible for this change, but studies of Grameen Bank and BRAC found that they were more effective in reducing fertility rates than were family planning programs. Should we not be try-ing to figure out what BRAC and others in Bangladesh are doing right so that it can be emulated, rather than ignoring it or writing it off as a special case?

When Microsoft Founder Bill Gates visited a group of BRAC microfinance clients he told a journalist, “It was like a religious experience.” Was Gates deluded or was he seeing some-thing the researchers who denigrate microfi-nance fail to see?

Another misleading charge is that microfinance is not for the very poor, but for the economically active poor. In my mind, virtually every family is “economically active” – even begging is defined by some to be an economic activity. If we feel an urgency about reducing extreme poverty, should we not put aside pointless debates and identify those MFIs that are targeting and benefitting the poorest, and then apply those best practices as widely and as quickly as possible?

Good Borrowers Make Good Families Jamii Bora, which means ‘good families,’ is a case in point. This Kenyan microfinance institution has grown from working with 50 women beggars ten years ago to serving more than 200,000 members today. One of those entrepreneurs is Joyce Wairimu. Wairimu

was one of the 50 women beggars who started Jamii Bora with founder Ingrid Munro [who was interviewed in this magazine’s Nov/Dec 08 issue] in 1999. Munro calls Wairimu one of the fast climbers out of poverty. How fast? In ten years, Wairimu has built six businesses and employs 62 people.

Munro started Jamii Bora 20 years ago when she and her husband adopted three street children. It was in the fertile ground of Munro’s relationship with the mothers of her sons’ friends in the streets—women who were beggars— that her profound insights would grow. When Munro, a Swedish trained architect and urban planner, retired from the African Housing Fund in 1999, she thought she would also retire from the little group of 50 beggar women with whom she had been working. But when the women pleaded with her not to leave them, Munro agreed to stay and insisted that they must lift themselves out of poverty. For Munro that meant the women had to start developing the discipline of saving on a regular basis.

She had them come every Saturday with about 50 cents in savings. When they deposited their 50 cents she would give each of them two scoops of corn and one scoop of beans for free. She admits now that for those first two months she was tricking them into saving. After two months, the bags were empty, but the beggars continued to save. What is remarkable about Jamii Bora is its people: all MFI staff are former members, previously destitute themselves.

Numbers and profitability matter, but pov-erty reduction matters more – at least to me, and it should to all those who have commit-ted to halving the number of extreme poor by 2015. And while the critics are busy stirring up doubts, millions of microfinance clients are finding better lives for themselves and their families. n

“Another misleading charge is that microfinance is not for the very poor, but for the economically active poor. In my mind, virtually every family is “economically ac-tive” – even begging is defined by some to be an economic ac-tivity.”

cOver stOry

Sam Daley-Harris is Founder of the Microcredit Summit Campaign which seeks to reach 175 million poorest families with microcredit (www.microcreditsummit.org) and of RESULTS which seeks to create the political will to end poverty (www.results.org). He can be reached [email protected]

See http://www.microcreditsummit.org/SOCRs/SOCR2009_English.pdf1. For a summary of the findings of more than 90 studies completed through 2005, see “Measuring the Impact of 2. Microfinance: Taking Stock of What We Know,” by Nathanael Goldberg, Grameen Foundation White Paper Series, December 2005.For information about the PPI: http://www.progressoutofpoverty.org/3. For more about USAID’s compliance, visit: http://www.mrreporting.org/Pub/AnnualReports/Report2007.4. aspx)Sachs, Jeffrey. The End of Poverty. The Penguin Press, pp. 13–14, 2005.5.


Randomized experiments help practitioners, donors and academics evaluate the impact of microfinance interventions. The simple and adaptable methodol-ogy can accommodate variations to test different impacts. While randomized experiments have a unique ability to measure the impact of microfi-nance programs, they are not without challenges for the implementing institution. Jonathan Bauchet and Aparna Dalal from the Financial Access Initia-tive discuss the merits and limitations of randomized experiments for microfinance evaluations.

Randomized Experiments in Microfinance

Microfinance institutions can use randomized control trials to find out if their customers, such as farmers, are truly benefiting from their loans.Credit: The International Rice Research Institute

research

When medical researchers want to prove the effectiveness of a new car-

diac medicine or nutrition protocol, they set up randomized trials. The idea is simple and clean. The researchers form a list of patients requiring treatment and randomly choose re-cipients of the new treatment and those who will receive the standard treatment. Random selection ensures that the new treatment is not diverted to the most promising patients, and the study can yield a reliable measure of efficacy.

The same approach is increasingly popu-lar among evaluators of development inter-ventions, including microfinance. Donors, academics and practitioners are turning to randomized experiments to address the fun-damental challenge in impact evaluations - the need to clearly establish that the intervention caused the impact.

Imagine that we are a farmers’ cooperative that wishes to measure the impact of a new loan product designed to help farmers buy fertilizer. We could compare the income of farmers before and after receiving the loan. But we would worry that changes in income might not be wholly attributable to the new loans. What if, say, the farmers experienced a good monsoon or if prices of their produce happened to rise? Surely these factors had an effect on their income. To assess the real im-pact of the loan, we would need to separate the effect of these environmental factors from that of the loan. Ideally, we would want to compare the farmer’s actual income to what his income would have been had he not borrowed at all (the “counterfactual”).

But this seems impossible. Our farmers did borrow, so how can we measure the counter-factual? A common approach is to compare

the average income of farmers who received the fertilizer loan (the treatment group) to that of farmers who did not (the control group). The reasoning is that farmers in the control group are susceptible to the same environmen-tal factors, hence their income is representa-tive of our borrowing farmers had they not borrowed. This comparative technique works if the two groups were similar before the in-troduction of the loan. It would be unfair, for example, to compare the income of farmers who own large tracts of land to that of farm-ers who rent small patches, or the income of veteran farmers to that of inexperienced ones. In evaluation terms, we want to measure the unique impact of the fertilizer loan on income, net of the effect of land ownership, abilities and such other confounding factors.

In microfinance, it is difficult to assume that the treatment and control groups are similar at the start of a program. Individuals who choose to borrow and are approved by an MFI are undoubtedly different from those who chose not to borrow or are denied credit. Borrowers and non-borrowers may differ based on ob-servable characteristics like education, health and pre-loan income. They may also differ on unobservable characteristics like entre-preneurial ability or innate motivation. Fail-ing to account for these characteristics could lead to overestimating or underestimating the impact of the loan. This error in attribution

is called “selection bias.” Fortunately, statisti-cal techniques make it possible to isolate the confounding effect of measurable characteris-tics such as age or education. We have limited tools, however, to isolate the effect of unob-servable characteristics such as motivation. Randomized trials can help us do so.

Eliminating Selection BiasRandomized experiments solve the selec-tion problem in a simple, yet powerful way. In a randomized experiment, the two groups (borrowers and non-borrowers) are selected at random. Hence, with a sufficiently large sample, one is certain that the two groups are similar in both observable and unobservable characteristics, since chance assigned individ-uals to each group. As a result, access to the intervention in question (e.g., being randomly chosen to get the new fertilizer loan) is the only systematic difference between groups, al-lowing us to conclude that it must be the cause of any difference in the average outcomes in each group.

The Mechanics Implementing a randomized impact evalua-tion is a complex operation. In Figure 1, we provide a basic outline. As in the fertilizer loan example, the institution first identifies the intervention (program or product) they want to evaluate. This could be the fertilizer loan, a new savings device, a financial literacy pro-gram, etc. Randomized evaluations can also test variations in existing programs, such as marketing strategies, loan terms, interest rates and their impact on the take-up, repayment rates and profitability. As the second step, the lender identifies a set of potential clients for the loan. This set could consist of existing clients

“Practitioners, donors and aca-demics are turning to random-ized experiments to address the fundamental challenge in impact evaluations - the need to clearly establish that the intervention caused the impact.”


Figure 1. Five Steps in aRandomized Evaluation

Identify the intervention (fertilizer 1. loan)Identify potential clients2. Conduct baseline survey 3. Randomize Treatment and Control4. Perform the intervention (provide 5. fertilizer loans)Conduct follow-up survey (measure 6. impacts)

Jonathan Bauchet and Aparna Dalal work at the Financial Access Initiative. Through consortium member Innovations for Poverty Action, FAI is supporting randomized experiments to evaluate the effect of microfinance interventions in fifteen countries. A complete list of projects and publications are available at www.financialaccess.org and www.poverty-action.org. To access a list of readings related to randomized trials, please see www.microfinanceinsights.com.

research

of other products and/or potential new clients. Third, a baseline survey of these potential cli-ents is conducted to measure characteristics that might influence the outcomes of interest. Fourth, some of the potential clients, usually 50%, are randomly selected to receive the new loan. The rest must wait until after the study to be eligible for the product. By offering the loan only to the treatment group, the lender is ran-domizing access to credit. If the lender wished to randomize the use of credit, he could offer the loan to the entire set of potential clients, and only randomly select who actually is as-signed a loan from the group of individuals who applied. Randomizing access to credit or the use of credit matters for the lessons that can be drawn from the evaluation. Does the lender want the results to apply to all potential borrowers (access), or only to those who apply for the new loan product (use)? The fifth step is for the lender to perform the intervention, in this case to make the fertilizer loans. Last, after a period of time judged necessary for the loans to produce impact, the outcomes of interest are measured by conducting another survey of individuals in both the treatment and the control groups. At the core, the anal-ysis involves nothing more than comparing the mean of the outcome in the treatment and control groups. More elaborate techniques are occasionally necessary, particularly when the evaluation includes complex randomization schemes or design elements. In all cases, the analysis must be done by comparing partici-pants based on their initial assignment into the two groups, and not based on their actual participation, becausethe latter would reintro-duce a selection bias.

The Decisions and ChallengesInstitutions need to make a number of key decisions and face several challenges when designing and implementing such an experi-ment. The first decision pertains to the the sample size. While larger samples are more expensive and time consuming, they yield a more reliable measure of impact. This is re-ferred to as the “power” of a study, and can be determined arithmetically. Second, institu-tions must consider the level of randomization. For instance, institutions can randomize at the group level (and not at the individual level) by assigning village banks, villages or other groups to selectively receive an intervention.

Assuming all other design elements as fixed, group-level randomization requires a larger sample to maintain power. Third, institutions interested in using randomized impact evalu-ations must be willing to temporarily exclude clients from benefiting from its services. This could represent a departure from the institu-tion’s mission or strategy, and can sometimes raise ethical questions about who gets access to new products and services.

The greatest practical challenge in rand-omized experiments is to enforce the initial random assignment. Randomness is the source of our ability to claim that the inter-vention caused the observed outcomes. Three common threats to maintaining the random-ness in participation are: • Non-compliance: Members of a controlgroup assigned not to receive a loan might obtain loans from other lenders, or members of a treatment group can change their mind and not borrow after all. •Attrition:Individualsintheexperimentmaydrop out of the program or move away so that surveyors can’t measure the outcomes. If those that can’t be surveyed did better (worse) than those who can, the impact of the loan will be underestimated (overestimated).•Spillovers:Participantsassignedtothetreat-ment group might influence the outcomes of those in the control group, e.g., during a business training program, members from the treatment group might share insights from the training with neighbors in the control group. These events can be mitigated with careful design and implementation. Non-compliance

can be reduced by training bank staff carefully, and ensuring that they maintain the integrity of the groups. Attrition can be addressed by monitoring groups, and employing persistent surveyors. Spillovers are more difficult to deal with. One way to minimize this complication is to adjust the level of randomization. It may make sense to randomize larger groups into treatment or control bins rather than rand-omizing across individuals.

The LimitationsRandomized experiments are implemented by a specific organization in a particular set-ting, and therefore, provide limited support to generalizing the findings across other settings. The best way to overcome this limitation is to replicate the evaluations in various settings before drawing any major conclusion about the intervention.

Randomized experiments provide an esti-mate of the average impact of an intervention, but they do not tell us about the distribution of impacts. For example, if the fertilizer loan makes certain types of farmers much better off and all the others a little worse off, a ran-domized experiment (and most other evalu-ations) might yield the result that the average impact was positive if the positive impact is large enough to offset the sum of negative im-pacts. Sub-group analyses can be conducted to measure the impact for each type of farmer, but this exercise requires a sufficiently large sample in each sub-group.

As the debate on the impact of microfi-nance continues, randomized experiments serve as an effective tool to generate highly reliable information on the effect of programs and products. While they are demanding for its designers and implementers, they should be viewed as investments that help donors and policy makers channel resources and help institutions design better programs and products that can increase both their financial standing and social impacts. n


treNds

Microfinance Market IndicatorsStatistics and numbers related to global poverty, and the microfinance rating industry, selected and compiled by theMicrofinance Insights editorial team.

Global Poverty Map – an Illustration of Global Poverty from Multiple Perspectives

MFI Double Bottom Line4

Regional Snapshot of Gender Inequality2

Change in Global Poverty Levels1

Facts and Numbers2

SA and SSA show significantly low GDP per capita: US$3,893 and • US$3,099 respectivelySA spends 2% of GDP on Health, making it the lowest globally. • However, the public expenditure on education is 12% of GDP, while other regions spend about 5% on average. Gender Empowerment Measure index is lowest in MENA region, • scoring 0.34 in average

Cost per loan is one indi-• cator used to determine an MFI’s operational ef-ficiencyWhile MFIs save by build-• ing their credit portfolio rather than client base, cost per loan rises

SA: South Asia LAC: Latin America & CaribbeanEAP: East Asia & Pacific MENA: Middle East & North Africa ECA: Eastern Europe &Central Asia SSA: Sub- Saharan Africa

35

31

22

1819

141613

16

1213

10

0

5

10

15

20

25

30

35

40

1990 2004

SSA

SA

EAP

LAC

MENA

ECA

0102030405060708090

2002 2003 2004 2005 2006 2007 2008

SA and EAP

SSA and MENA

LAC

ECA

90

95

100

105

110

115

120

125

130

2004 2005 2006

Africa Asia ECA LAC MENA

0

50

100

150

200

250

Africa Asia ECA LAC MENA

2004

2005

2006

Slight Decline of the Population Undernourished 2

Financial Self-Sufficiency Cost per Loan

Regional Breakdown of Microfinance Outreach3

SA EAP LAC MENA SSA

Literacy Rate, Female (% aged over 15)

54% 86% 88% 73% 55%

Literacy Rate, Male (% aged over 15)

72% 91% 90% 85% 69%

Estimated Earned Income,Female (PPP US$)

2,069 6,091 5,189 3,934 1,764

Estimated Earned Income, Male (PPP US$)

5,004 12,309 10,323 14,287 3,738

% o

f tot

al p

opul

ation

% o

f abs

olut

e po

or fa

mili

es re

ache

d%

of fi

nanc

ial s

elf s

uffici

ency

US$

Human Development Index 1990 Human Development Index 2008

High Human DevelopmentMed Human DevelopmentLow Human DevelopmentNot Ranked


BOOks

As children, we all aspire to do something different: become a pilot, a dancer, a painter. And then we grow up, and our responses become more conditioned to what is expected of us. Some, however, do not forget their goals, and nothing, not even the highest paying, most com-fortable corporate jobs can shake that belief. It is not the lure of making it big or the fancy of being labeled an “outlier,” it is a belief in their objective and a fire to make it happen.

Rashmi Bansal, a graduate of Indian Institute of Management in Ahmedabad (IIM-A)—In-dia’s Harvard—and the author of Stay Hungry Stay Foolish, captures this self-motivating spirit of 25 IIM-A students who decided to take up off-the-beaten-path careers. The short stories are divided into three sections: believers (entrepreneurship was their chosen path), opportunists, and people with alternative vision (those that use entrepreneurship to create social impact).This excerpt reveals the journey of one such man with an alternative vision, venkat Krishnan, the founder of GiveIndia, who very early in his life, was exposed to clear class discrimination and decided to do something about it.

-R

as

hm

iB

an

sa

l

Rashesh Shah did it.

Sanjeev Bikhchandani did it.

Shantanu Prakash did it.

'Stay Hungry, Stay Foolish' is the

story of 25 such IIM Ahmedabad

graduates who chose the rough

road of entrepreneurship. They

are diverse in age, in outlook and

the industries they made a mark

in. But they have one thing in

common: they believed in the

power of their dreams.

This book seeks to inspire young

Bschool graduates to look

beyond placements and salaries.

To believe in their dreams.

ISBN 978-81-904530-1-1

The Centre for Innovation, Incubation andEntrepreneurship (CIIE) at IIM Ahmedabad

aims at fostering innovation-drivenentrepreneurship through incubation, research

and dissemination of knowledge.

Rs.125/-

Using Entrepreneurship to Bring About Social ChangeBook Excerpt: Stay Hungry Stay Foolish

His first IP (Independent Project) was on the feasibility of private enterprise in

education, especially vocational education. By this time he was quite confident about

wanting to become an entrepreneur, at some stage in life. But it was also clear that even if he became an entrepreneur, it would not be something like IT, but about “making a dif-ference.”

“I remember my first reflective note for the LEM class1 – I see myself as an instrument or tool that is available to the society. And my choices should be guided by maximizing the returns that I will give to the society. So I will not do something just because I like it, but because that is the best use of my time for the society’s benefit.”

Come placement and you know Venkat is not going to go for the usual companies.

Actually, he almost joined an aatawallah2 near Vapi who had participated in placement that year. He was offering a fancy salary, but the chap said that the job is to help the atta chakki3 and help him to save income tax. That put Venkat off completely.

“I would have joined him, if he had been an honest guy. Because he was asking you to run the business as a CEO,” says Venkat wistfully.

And that’s something that makes a lot of sense for any MBA with ambitions of becom-ing an entrepreneur. Joining a company which may not be the biggest or most glamorous name in the business, but a place where you get a hands-on and get a 360 degree experi-ence of actually running a business...

“We need the best minds in the country to think, ‘What is the human ideal that we aspire towards?’ Rather than what is the next 30 crore flat that I can buy for myself, or whatever else that I can do for myself.”

He hastens to add, “Please have a 30 crore flat, but don’t be blind to the world outside your window.”

The time was ripe. A growing number of Indians were beginning to do well for them-selves. They were going to have everything that they could possibly want very early in life. Could we not then start building a culture that helps give back?

And thus was born “GiveIndia,” an organi-zation dedicated to promoting and enabling a culture of “giving.”

When Venkat left Eklavya4, a couple of things fell in place. He had just bought a home PC and was fascinated by the power of the internet.

The net was also a useful source of informa-tion. Venkat found out that in the US “giv-ing” – in all forms – formed 1.8% of GDP or US$180bn in ‘99-00. The corresponding number in India was less than 0.1% or 0.2%.

And here’s a startling fact – the poorest peo-ple give the most, as a percentage of income. This is true not only in the US, but all over the world.

Venkat realized that on the one hand there were organizations and people who are pas-sionate about and doing amazing work which nobody has heard of. On the other hand, there is an opportunity to pay back.

“I used to write to batchmates, friends, peo-ple I know who are doing well asking them, ‘Why don’t you give more?’ The first question was, ‘Who can I give to? I don’t know if my money will be used properly.’ That typical cynicism that we have in our system is per-haps justified.”

So the idea was born that one can create an organization that showcases NGOs do-ing good work and enable those who wish to “give” a platform to connect with them. And thus, help create a culture of giving back...

“I think the next generation has a much greater orientation of giving. We have all seen difficult times in childhood. So there is this fear that something could happen, there could be a recession, etc. But the youngsters today are so confident, so secure. They feel confident that we will be able to take care of ourselves, so let’s share a bit.”

Venkat recalls GiveIndia’s first ever annual report, which contained a paragraph which he wished to see in the 2020 annual report. It’s a letter which read as follows:Dear Stakeholders,We are delighted to inform you GiveIndia has closed down. Donors are now active, they are finding NGOs, they are engaging with them, they are giving money directly and they don’t need GiveIndia. n

Ordering InformationStay Hungry Stay FoolishBy: Rashmi BansalPublished by: Centre for Innovation, In-cubation and Entrepreneurship, February 2009Pages: 325ISBN: 9788190453011 Price: INR 125, US$ 2.5

Laboratory in Entrepreneurial Motivation, a course offered to IIM-A students1. An owner/worker at the flour mill2. Flour mill3. A day school for the poor started by a few IIM-A graduates under the guidance of Professor Sunil Handa4.

InvestmentBalancing Social and Financial

CritiqueIt’s No Magic Bullet

GenderThe Women’s Empowerment Myth

IndexSocial Impact Tools

Features

A bimonthly publication from Intellecap www.microfinanceinsights.com

Cover Price: US$12/EUR 8/INR 75

TM

Evaluating the Impact of Microfinance Panacea or Placebo?

Vol.12, May/June 2009

Notes: *Rates applicable according to country of delivery. Listed rates do not include postage, which varies by country. ** Shipping cost for Indian customers. All rates are online at

www.microfinanceinsights.com.

Online payment options are available at www.microfiinanceinsights.comFor details on payment through wire transfer, e-mail us at [email protected]

Mail this form to:Editor – Microfinance Insights, Intellectual Capital Advisory Services Pvt. Ltd,512, 5th Floor, Palm Spring, Link Road, Malad (W), Mumbai 400064, India.

Phone: +91-22-40359222/210

Name of bank: Address of bank:

Favoring “Intellectual Capital Advisory Services Pvt. Ltd.” payable at Mumbai, India

Phone No.:Mobile No.:

E-mail 1:E-mail 2:

Name:

We accept checks/ demand drafts/ pay orders drawn only on Indian banksCheck/ Demand Draft/ Pay Order No.:

Organization:

Mailing Address:

City:State:Pin:Country:

We’ve gone bimonthly! Now, instead of 4 issues a year, you’ll receive 6. Check out our new rates below. To subscribe, visit us online at www.microfinanceinsights.com or fill out

the form and mail us your check!Subscription Options

(Choose your currency)Single Issue 1 year:

6 issues2 years:

12 issues

INR Rs 75 + 10** Rs 400 + 60** Rs 800 + 100**

USD* $12 $60 $120

EUR* €8 €40 €80

GBP* £6 £30 £60

microfinance insights vol.12: teaser

Documents