How We Talk About Microfinance: A Reply to Dean Karlan

> Posted by Alex Counts, Founder, Grameen Foundation

On Sunday, August 23, as I was enjoying some of the final days of summer visiting friends in New Hampshire, I noticed that I had been tagged in a tweet by Dean Karlan, the founder and president of Innovations for Poverty Action.  He provided a link to an article about FINCA that included extensive quotes from its CEO, Rupert Scofield.  He asked Rupert if he really believed microfinance could reduce terrorism, and asked me what I thought (“whatcha think?” was the precise formulation of his question).  He tweeted again on Monday, asking whether I was “still going to stand by [my] claim that no microcredit leaders make grandiose and overselling impact claims?”

First of all, I have never said that no microcredit leaders have ever exaggerated impact claims.  I believe that those exaggerated claims have been rare and atypical, especially in recent years.  In other words, the tendency for practitioners and advocates to make exaggerated claims not backed up by data has itself been quite exaggerated.

But I don’t think Twitter is the best medium for exploring such topics.  So I was grateful when the Center for Financial Inclusion agreed to publish this response to Dean’s public queries of me, in which I could address some related issues about microfinance advocacy and research.  (This post builds upon some of the observations I made in reviewing Dean Karlan and Jacob Appel’s impressive but flawed book, More Than Good Intentions.)

Regarding the article Dean tweeted about, I am supportive of Rupert’s statements and encourage others to read it and come to their own conclusions.  (Having been the public face of an international humanitarian organization for 18 years, I also realize that journalists sometimes focus on a very small part of what someone says in an interview, often on those things that are potentially the most controversial.)  For the most part, Rupert comments on specific microfinance clients he and the journalist met and on his past experiences and how they shaped his view of microfinance.  It’s impossible to challenge any of those observations and recollections.  They are statements of personal experience and opinion.

At one point he makes the case that microfinance – which research has shown to have consistent though modest positive impacts on human well-being and in a few cases, such as in Bangladesh and through microsavings in multiple countries, more pronounced positive impacts – is part of a larger movement to address the human suffering and poverty that can, as a byproduct, decrease the attractiveness of ideologies that encourage terrorism.

If Rupert wants to make the leap to say that successfully addressing poverty also addresses terrorism, that is a statement of policy or ideology, rather than one of direct claims about microfinance or any other anti-poverty strategy.  Rupert’s effort to provide nuance – saying specifically that microfinance is not “the only solution” – was wise.  I assume he offered the journalist even more qualifiers and nuance that never made their way into the article.  So I don’t find much to disagree with here, nor do I find anything that is grandiose or overselling, even though I have chosen not to make this microfinance/poverty/terrorism argument in public myself.

But let me take a moment to make a few related points.  Like people in any other field, microfinance practitioners and advocates aggressively market their products.  What else would we do?  Researchers do that about research, automotive companies about cars, universities about the education they provide (and education in general), and so on.

There have been scores of studies of microfinance that have shown positive impact, and we have catalogued and cited them.  When a leading study of microfinance, summarized in an article in a peer-reviewed academic journal in 1998, found significant positive impacts on poverty and empowerment in Bangladesh, people like me did not keep quiet about such credible, positive and encouraging findings, as they seemed to confirm earlier, less rigorous research.  Why would we?  [After much effort and the passage of 16 (!) years, two respected academics were able to publish a peer-reviewed critique of the Bangladesh study in 2014, muddying the issue further and causing most practitioners and advocates to exercise additional caution when citing the original findings from Bangladesh, and also prompting more head scratching about how they should digest and react to disagreements among academics.]

Later studies, on a handful of microfinance institutions among the thousands that exist today, pointed to more muted impact (at least from microcredit), causing nearly all practitioners and advocates to gradually recalibrate their marketing to be consistent with the evolving body of knowledge.  While I am sure that one could find individual quotes from industry leaders including myself that stretched the bounds of what the research said at the time, in general I think we have little to be defensive about if you look beyond anecdotes – isolated statements, often taken out of context – and focus on the typical statements of industry leaders and how they have logically evolved as the research literature itself has evolved.

In fact, Grameen Foundation has tried to systematically capture the findings of the literature for our supporters and others by commissioning academics every five years to independently summarize, in layman’s terms, what successive waves of studies have concluded.  Our last such effort was described by the thoughtful microfinance skeptic David Roodman in a blog post as “a model of public communication about social science research.”  A third in this series, written like the second one by Professor Kathleen Odell of Dominican University, is due out early this fall.

I think it is arguable that the variation of microfinance practiced in Bangladesh and perhaps a few other countries does have significant positive impact on poverty while microfinance in many other parts of the world, as currently practiced, brings more limited human development benefits.  While the practitioners of RCTs are very rigorous about random assignment at the level of the program participants, their choice of MFIs to study among the thousands in operation today has been anything but scientific.  Rather than attempting to choose a truly representative sample of MFIs for study, their selection has been uncoordinated and opportunistic, and has to date completely avoided the country where microfinance is most intensively practiced.

For this and other reasons, there is real debate even within academia about what can be learned from RCTs, especially what can be applied to other places and similar programs.  For now, I believe humility and curiosity are called for, as well as constructive efforts that help us get beyond “proving” whether microfinance works and refocus on a more constructive topic: leveraging research to further improve this social innovation that is clearly here to stay.

As I mentioned in my comments at a recent World Bank conference on the impact of microcredit, I continue to be puzzled by the tendency of some researchers (including the authors of the American Economic Journal summary article) to suggest that the microfinance industry somehow made a collective “promise” to “transform” the lives of the average client in under two years, and then failed to deliver on it – and then put this forward as the most compelling narrative about our sector.  (See, for example, the misleading title of this press release.)

Leaving aside the fact that “transformation” is never rigorously defined, this argument requires one to believe that the most aggressive marketing claims made over the last 30 years were actually typical, and that they were collectively agreed upon by this fragmented, decentralized industry some point in the past and regularly reaffirmed thereafter (rather than being periodically modified based on new research).  In both cases, I find the evidence lacking.

One way of interpreting this dynamic, expressed to me by another industry leader as I prepared this response, is that certain research organizations are “aggressively marketing their work by ‘overclaiming’ [or in other words] exaggerating the claims [made by] microfinance [advocates] initially” in order to heighten interest in their research.

So, in summary, I find Rupert’s statements quoted by the journalist to be descriptive, at times quite nuanced, and accurate, as far as they go.  In one case he expresses his personal belief about the link between poverty reduction and terrorism, one that I don’t believe can be conclusively supported or refuted by the existing evidence, but that intuitively makes sense to many people, including me.

Shaped by Rupert’s vast experience in Central America, where the interviews also took place, I suspect he may have been using the terms “violence” and “terrorism” almost interchangeably in the context of a free-wheeling conversation.  Perhaps everyone can agree that reducing poverty and terrorism/violence are two positive goals where the causal relationship between them is tenuous and uncertain, but hardly implausible.

The bottom line for me is that Rupert’s comments are not part of a discussion of impact evidence.  They are intended to communicate why FINCA – and Rupert – find their work meaningful, and that has a lot to do with the legacy of conflict in Central America.  Most readers of the article would understand this, and interpret the claims in exactly the way they were intended.

Image credit: Accion

Have you read?

Reflections on the Latest Studies on Microcredit

Considerations for the Pursuit of Financial Inclusion – Beyond Account Access

Alex Counts on Karlan and Appel’s “More Than Good Intentions”