The GiveWell Blog

Mistargeted microfinance?

There are many studies attempting to gauge microloans’ impact on borrowers, but most suffer from the problem of selection bias: by comparing participants and non-participants, they may be picking up other differences between these groups (for example, people who participate in microloan programs may be wealthier to begin with, so a study showing that they have higher incomes does not really show that microloans help).

One of the better-known attempts to avoid this problem was a 1997 study that attempted to take advantage of a “landholding requirement”: a program that allowed people to borrow only if they held less than a half-acre of land. This study is one of the few empirically encouraging studies of microloan programs, as we discuss here, although there was some dispute over whether the landholding requirement was strictly enforced.

New analysis by the Center for Global Development’s David Roodman shows that even though the microfinance program attempted to target people with less land, in fact people with more land borrowed significantly more. I agree with Roodman’s general conclusion:

This is not necessarily bad: serving richer clients with larger loans may make it more economical to reach poorer people in the same villages–an example of cross-subsidization. And someone whose main asset is an acre or two or riceland in rural Bangladesh is hardly rich by western standards–and might even make better use of the credit. Still, the “mistargeting” of landed families contradicts the public image of microfinance in Bangladesh as targeting the poorest, and probably needs to be better understood.

Additional GiveWell materials related to microfinance:

  • Our full review of the evidence for microfinance programs is available here
  • We first blogged about microfinance here and here.
  • Those blog posts rely heavily on a white paper published by the Grameen Foundation, available here (PDF).


  • on March 26, 2009 at 1:22 pm said:

    Landless people or people with less size of land often feel threatened when they have to compete with those who own larger pieces of land for seeking micro-loans from the same source or scheme. Also, comparatively, they have less skills and knowledge for understanding the process of applying for a loan.

  • George on March 26, 2009 at 3:15 pm said:

    As you noted microfinance results have been clouded by selection bias problems. However, isn’t it a bit of a leap to draw serious conclusions from a study that uses data from 17 years ago and that included a sample size of

  • George on March 26, 2009 at 3:17 pm said:

    **Comment got cut off**

    As you noted microfinance results have been clouded by selection bias problems. However, isn’t it a bit of a leap to draw serious conclusions from a study that uses data from 17 years ago and that included a sample size of

  • Eric Hemel on March 30, 2009 at 8:08 pm said:

    Your discussion of micro-finance, while well informed, leaves out two critical points:

    (1) Most MFIs have grown dramatically over the last decade, creating the possibility that debtors may be borrowing, implicitly or explicitly, to repay their loans. Bernard Madoff paid off all of his investors, who redeemed their funds, while his assets under management were growing. That does not attest to his investment performance. Similarly, the repayment record of any financial intermediary which is growing rapidly is similarly suspect.

    (2) You fail to mention one key aspect of most micro-finance loans: the interest ratew charged. The typical interest rates that I have seen attributed to reputable MFIs range between 30 and 100 percent. I have a much harder time believing in the benefits of “consumption smoothing” when interest rates are usurious (by conventional standards) than when they are in the single digits.

  • Holden on March 31, 2009 at 1:59 pm said:

    George: I’m not sure why your comment is getting cut off; emailed you about this.

  • Holden on March 31, 2009 at 2:03 pm said:

    Eric: thanks for the comment. I think these are valid concerns.

    Regarding #1, one of the few data sources we have on how people are using loans implies that a significant proportion could be using them to repay other debts – see this chart (based on this paper (PDF)). Note that because people could be borrowing from some MFIs to pay back others, the concern about misleading repayment rates applies to all MFIs, not just those that are growing.

    Regarding #2, with volatile enough incomes even a high-interest-rate loan could be helpful for consumption smoothing, but like you, I do find that these high interest rates make the idea at least a bit harder to swallow.

  • Sommerhus Her on April 1, 2009 at 5:04 am said:

    Microfinance should be free from any bias as it should be easily availed even by the poor. I think the problem lies in the loan interest. It is still steep although it is projected to have low interest.

Comments are closed.