CGAP is the World Bank’s (not-quite-so-)arm’s-length sub-organisation whose role is to promote microfinance. CGAP (“see-gap“) once stood for “Consultative Group to Assist the Poorest“, now it officially stands for “Consultative Group to Assist the Poor”. Actually, if nomen were to be omen, it should probably stand for “Consultative Group to assist (those who lend to) the Poor (and not-so poor)”.
So many functions … but can it cut? Image: Slartibartfass (CC BY-SA 3.0)
I don’t expect CGAP to function as an independent evaluator of microfinance. What I do expect is that CGAP publications have minimum standards of research quality and logic.
The most recent CGAP report, entitled “Latest Findings from Randomized Evaluations of Microfinance” (Bauchet et. al.), however, is appalling on both counts. Nearly everything about this report is problematic. It is racked by wishful thinking – to paraphrase: “we may not have evidence that microfinance does what it was supposed to, but we still believe it works” – and it has a disturbing feel about it, which derives from: (1) what the authors have left out, and (2) the heavy tension between concern for the poor and patronising them.
Ignorance or ignoring?
The first main problem is that a uniquely important piece of literature is left out. Given that their report is supposed to give an overview of the available knowledge from RCTs, I was surprised – to say the least – that Bauchet and colleagues did without a single mention, let alone a discussion, of the systematic review by Duvendack et. al.
Published just this summer for the British DFID, Duvendack and her co-authors went through a whopping 2,643 publications on microfinance in order to systematically assess what is and what isn’t known about the impacts (Bauchet et. al., in comparison, looked at a total of 20). In what is the most comprehensive study to date, Duvendack et. al. found dismayingly that (after over 30 years) “there is no good evidence for” microfinance actually working – contrary to the message Bauchet et. al. wanted to convey. How could the CGAP report have missed this recent, huge and hotly-debated publication?
Either: the authors were genuinely ignorant of the Duvendack study, in which case they really have no idea where the research on microfinance’s impact currently stands. Or: they didn’t agree with the conclusions, and therefore conveniently chose to ignore them.
It doesn’t do what it’s supposed to do… but it certainly works!
And on to the second problem, which is about how to measure success (I’m picking out examples here, but I have many more in case it seems I am being selective)… The key message of the new CGAP report is: Microfinance works, but not in a miraculous fashion, and not in the ways it was expected to. It doesn’t reduce poverty, but it does practically everything else that we didn’t try to prove. A good example along these lines (emphasis added): “No evidence was found that microcredit was empowering women, at least along measured dimensions“.
“Now with nearly 200 million borrowers, microcredit has been successful in bringing formal financial services to the poor”, Bauchet et. al. proclaim. Success!! 200 million people have loans. But are these loans doing anything useful? Well… As was reported after the publication of the first RCT results, actually borrowers in India didn‘t earn higher profits, hire more employes, or see more revenues – but they did turn out to be 1.7 percentage points more likely to start a business. Wow: receiving a loan after telling a bank that I want to start a business makes me marginally more likely to actually start one. Come on. Shouldn’t the standards of success be higher than that?
Pausing to think about these results, in actual fact this means: people with small businesses, who voluntarily came forward to get a loan for whatever reasons they had, and who did not receive a loan, were no better or worse-off than their counterparts who had received a loan. What stronger proof could there be for microcredit being useless?
At least Bauchet et. al. are honest about this much:
Microcredit is not transforming informal markets and generating significantly higher incomes on average for enterprises.
But they immediately add:
And yet the industry has focused almost exclusively on the rhetoric of entrepreneurship and has overlooked the many important benefits to households that are using loans to accelerate consumption, absorb shocks, or make household investments, such as investments in durable goods, home improvements, or education for their children. …
While these uses of financial services are different from the uses initially anticipated, they are still valuable, and the ability to manage finances is a fundamental part of everyday life for all people.
If anything, this text is evidence of an industry searching for a new source of legitimacy. Microfinance used to be about “empowering women”, then it was about “microentrepreneurship”, now the slogan is “financial inclusion”; which may be little more than a catchphrase for the idea that bringing the poor access to finance is an end in and of itself.
Thank you for not smoking, thanks to microfinance
Worse yet, the new focus seems to be on making microborrowers better people. Now the measure of success is that borrowers should adhere to a Protestant work ethic:
Those who started a new business cut back on temptation goods (tobacco, alcohol, tea, betel leaves, gambling, and food consumed outside the home) and invested more – tightening their belts to make the most of the new opportunity. This switch from temptation goods to investment and durable consumption in the groups with businesses is an encouraging finding.
Perhaps the economist who wrote that sentence doesn’t smoke, and never eats a samosa at a roadside stall. Yet, the patronising audacity of categorising a cup of tea as a “temptation good”, as if it were a frivolous luxury (in India, mind you), is unsettling. While to a homo oeconomicus a foregone cigarette is merely one small step towards rationality, one can imagine the real-life borrower’s view of this: “Working to pay off this loan has even taken my daily cup of tea from me.” Only in the economist’s world is this an improvement thanks to microcredit.
Unsurprisingly, the authors of the report don’t dwell on significant findings (from the Philippines) that “subjective well-being slightly declined”. Instead, they contently note:
The new businesses created and the shift away from small “wasteful” expenditures implied that access to loans enabled households to make clear choices to reprioritize, invest, and make the most of the new opportunity: “The main objective of microfinance seemed to have been achieved. It was not miraculous, but it was working” (Banerjee and Duflo 2011, p.171.).
Is this really a positive thing? When read in a the cold light of reason, the evidence presented here shows: because of microcredit, poor people consume fewer locally-produced goods and services (like tea, hand-rolled cigarettes, and roadside food), thereby hollowing out the local market, and they now work harder (or do something, anything) to pay off loans for a microfinance bank with foreign shareholders; and (unsurprisingly) as a result they are less happy.
In this way, consistently throughout the report, Bauchet and co-authors (who are, in fact, mostly affiliated with the same institutes producing the RCTs) reproduce the subtle spin put on the evidence. The main objective, as they interpret it, is not to make the poor materially better-off: it is to get them to make sensible choices (as Western economists define them) and to get them to work. The new success of microfinance: re-programming people? Microfinance then clearly becomes a tool of governmentality, a sublte but forceful way of changing people’s behavior. In the section on savings products, they present poor people as irrational with regards to the future. Microfinance organisations should work to create savings discipline.
Despite the lack of evidence for positive effects on welfare from credit, the studies so far offer tantalizing evidence that there could be important potential benefits for some poor households to be gained by helping the poor reprioritize their expenditures.
When neither a lack of positive evidence, nor a fair amount of negative evidence can break the optimistic outlook, it seems that in reality, the irrational wishful thinkers are the evaluators themselves – not the borrowers.
Finally, the mind boggles as to what the minimum requirement is for being allowed to publish reports for CGAP are:
- is it high standards of evidence, solid logic, transparency of assumptions? High standards, good logic and transparency are a bonus, perhaps, since very insightful reports sometimes emerge from CGAP…
- or is the minimum requirement being “on message”? No matter what, microfinance has to be the solution. After reading this report, I am inclined to think it is that.
(phil)
5 comments
Comments feed for this article
December 12, 2011 at 18:17
Ramesh S Arunachalam
Great Phil, Brilliant Stuff and could not agree more! Read their piece on SKS and you know what the reality is and you can see what they do to their reports. Simply put, CGAP is not accountable to anyone and they do what they like. Times are changing as change and revolution are sweeping the world, albeit slowly. One day, I am sure that CGAP will become accountable and I wait for that day! Look at the conflicts of interest on their executive committee and board – how can they be an independent evaluator of Micro-Finance! Keep up the excellent work. Cheers Ramesh S Arunachalam
December 12, 2011 at 18:32
philmader
Thank you Ramesh. Your kind words are very welcome, not least after seeing Roodman’s opinion of this as a “thorough, careful, and clear summary” (in what looks like an unacknowledged response to my dissection of the report). Congratulations, by the way, on your very insightful blogging for Moneylife; I see a common theme between our work, which is trying to unmask (and challenge) the wishful thinking in microfinance, which so often has nourished the hope that problems will go away if you just don’t acknowledge them as problems.
March 23, 2014 at 23:20
Ali Mehmood
Hi Phil: Could I remind you that Duvendack et al is based on 58 studies only and not 2,000 plus, a figure you have highlighted so many a times in your blog/book; it gives a mischievous feeling unless you really do not know the difference between the number of items resulting from searching databases from the one being used for a review. [in lay man terms, no one looked at more than few lines of abstract or summary for over 2000 studies]. I know this difference because I am about to complete a systematic review myself which started with more than 5,000 records.
On a side note, having dismissed lots of valid research using a heuristic screening criteria, findings from Duvendack et al, at best, are heavily biased in the favour of quantitative methodologies. Besides that they conveniently changed the original review question to fit their pre-conceived notions about microfinance impact which is very bad academic practice (you can see that change from impact on income to impact on well-being stated on page 5).
It’s a shame to see that studies having limited to no generalizability are being scaled to represent the global trends in research through flawed systematic review! Needless to say, if it was not DfID sponsored, this publication should have been dismissed by many people in academia. But we all know the acceptance of research now-a-days is not driven by quality rather it’s based on how well someone is connected.
Have a nice evening,
March 23, 2014 at 23:40
philmader
Hi Ali, thanks for your insightful comment – and for reading the blogbook. What you say is, of course, true: Duvendack et al only studied 58 studies in depth. But the 2,500 plus figure is still a very important one, and not some mischief of mine, because it is the population which the sample represents (or seeks to). And if you start with 5,000, that’s all the more comprehensive and impressive.
As far as I understand, for the purposes of an SR it is often enough possible to tell from the abstract how valid any results of the study may be. In any case I look forward to reading your SR, and if I do a post on it here, will deal in greater depth with how the sample is whittled down from the population of available studies.
With your assertion that the DFID name was responsible for any attention the Duvendack et al study received, I’m not sure if you do the study justice, however. In fact my impression has always been the study got surprisingly little attention, and was hardly promoted by DFID; especially in light of the gap which it filled.
Either way it’s important that there are more systematic studies out there, so I’m keenly awaiting yours. Regards.
August 25, 2015 at 15:17
Out now: The Political Economy of Microfinance: Financializing Poverty |
[…] offers new insights into which have particular significance in light of the continually unresolved issues around poverty impact. More than 35 years into the microfinance experiment, the fact is we still […]