The GiveWell Blog

Microfinance evidence of impact

David Roodman posts a review of recent high-quality studies of microfinance.

Note that prior to these fairly recent studies, most impact studies in this area had serious flaws, as Mr. Roodman notes and as we argue in our year-old review of this area.

The studies Mr. Roodman discusses are randomized controlled trials, and so their conclusions are far more trustworthy. But the conclusions are also generally less encouraging.

  • A consumer loan program in South Africa (i.e., loans that were not specifically for business expansion) had very positive effects: “Six to twelve months after they applied for the four-month loans, unrejected applicants were 10 percentage points more likely to have a job, 7 points less likely to be below the poverty line, and 6 points less likely to report that someone in the household had gone to bed hungry in the last month than those rejected.” (This study is included in our year-old review of this area.)
  • A Philippines microcredit program targeting the middle class found “no changes in household income, spending, or diet 1–2 years later.”
  • A savings program “appeared to help [women] accumulate money for investments such as stock for their stores, leading eventually to greater prosperity.”
  • The one study of a traditional microfinance program targeting the very poor found “no impact on total income, spending, health, or school enrollment rates” overall, though some subgroups appeared to benefit.

It seems to us that rigorous studies have not shown the impact implied by success stories, and that the most encouraging effects have come from programs that are not centered around business expansion loans.


  • I feel that micro finance may be a victim of all its attention. I feel that it is a tool in the toolkit of (what I think you call) financial empowerment. It has come to such fame because it is (supposedly) an extremely effective tool.

    Shouldn’t it be judged relative to other such programs rather then a stand alone area? What are your thoughts?

  • We believe donors should consider as wide a range of options as possible to achieve goals that are meaningful to them.

    Our current position is that health programs are a more proven and promising path than microfinance for individual donors seeking to help improve lives in the developing world. (See our report for more.) We continually re-examine this position as more information becomes available about the impact of microfinance programs.

  • Holden,

    My understanding of your position is that you class Health & Economic empowerment separately. I think that seperate classes is a necessary compromise. I don’t mean to say comparing & even ranking different classes isn’t worthwhile, just that it is a different type of activity to comparing and ranking within a class.

    My question/comment above relates to classing microfinance projects from other financial empowerment projects.

  • We don’t believe there is a single “correct” way to split charities into categories. Some donors (myself included) may be open to both health and economic empowerment programs, and therefore seek to compare across those two areas. Others may be determined to stick with economic empowerment, but willing to compare microfinance programs to other programs such as agricultural extension services. Still others may be set on microfinance specifically.

    For donors like me who want to compare across areas, my current position is that there are better options in health than in economic empowerment, but I’m not sure yet whether the best economic empowerment options are microfinance programs. From what we’ve seen, all economic empowerment programs have relatively spotty track records.

    Does that clarify?

  • Thanks for that Holden.

    Do you think that there is an inherent bias towards health. I imagine that in health, it is much easier to measure directly or infer impact. For example, an immunisation program can take the number of immunisations and multiply by the decreased likelihood of contracting a disease. You can from this get a reasonably good guess at impact. A health education program, on the other hand, would need to be measured directly to gauge effectiveness. The impact might also be be subject to compounding effects (children’s education programs might increase the effectiveness of parent education programs at a later stage) or be primarily long term effects.

    BTW, this was not what I was referring to in my earlier comment. I was wondering if you think that microfinance has reached a point of diminishing returns. After the nobel prize and all the attention, I assume that a lot of resources were diverted to microfinance. Less specialised organisations may have taken up the task and perhaps less qualified recipients were targeted.

    Is this notion accurate?

  • Bias towards health. Emphasizing proven results will result in a bias (though not an exclusive one) toward activities whose outcomes are easier to measure. We feel this is an appropriate bias.

    Easier-to-evaluate endeavors provide stronger feedback loops to the aid agencies conducting them, which in our view makes them more likely to succeed (indeed, the fact that health outcomes are particularly measurable could partly explain why there has been particularly great improvement in medicine, and thus in health outcomes, over time). Easier-to-evaluate endeavors also generate stronger evidence of effectiveness ultimately, which makes it easier for an individual donor to have confidence and to continually hold the charities they give to accountable. If easier-to-measure activities are easier to succeed at and easier to have confidence in, then it makes sense that they would be more likely to be recommended.

    Our goal is not to give all sectors an equal chance, but to find opportunities that can be confidently recommended to individual donors.

    Has microfinance reached diminishing returns? It’s true that a lot of money seems to have gone into the area, but there also may be a great deal of need in the area. Many believe that microfinance is still vastly underfunded because so many people in the world still can’t access credit. I don’t feel the data are there to have a strong view one way or the other.

  • I find this debate crazy. If you can live your life without a checkbook, a savings account, access to school,mortgage, auto and business loans, insurance, annuities, retirement plans, etc etc, then you are prepared to live well with the same level of access to financial services that the majority of the world’s population enjoys(sic). Until the majority has access to appropriate financial services and therfore the ability to manage their financial risks and assets, the world’s poor will remain vulnerable to every shock — natural disaster, epidemic, war, political upheaval — everything. The best microfinance isn’t some isolated do-gooder effort, it is connecting the economic activity of the poor to the financial system in which they live. That is real empowerment. How many hospitals were built in the 1970’s with international project finance that are now rotten hulks because they could not be maintained and couldn’t build a sustainable business model? When people can manage their financial lives, then they can choose and pay for services. There is no trade off between health and microfinance. They are equally valuable and in fact depend on each other for long-term viability and effect. But health funding without a financially active and financially healthy client base isn’t sustainable over the long term, it will always require charity. We may knock off a disease or two, but primary, acute and tertiary care for all won’t become realities without large scale changes in access to finance for the poor majority.

  • Roy, your comment incorporates several hypotheses about the need for – and more importantly, the likely effect of – credit. It’s possible that more credit would greatly improve the health of economies, solving many of the problems you describe; it’s also possible that improved health would lead to economic growth, more effectively than greater access to credit; it’s also possible that there is no solution to bad economic health on the horizon.

    I like about studies like those above because they take plausible but by no means obvious hypotheses and subject them to empirical tests and critical scrutiny.

Comments are closed.