Microcredit: Mainstream Questions Mainstream
By Farooque Chowdhury
19 August, 2011
Countercurrents.org
Mainstream microcredit findings are now being questioned by mainstream. Thus, part of mainstream research system, stakes and stakeholders, and vulnerability of mainstream have got exposed.
A DFID funded systematic review concludes: “[A]lmost all impact evaluations of microfinance suffer from weak methodologies and inadequate data […] thus the reliability of impact estimates are adversely affected.” [Duvendack M, Palmer-Jones R, Copestake JG, Hooper L, Loke Y, Rao N (2011) What is the evidence of the impact of microfinance on the well-being of poor people? London : EPPI-Centre, Social Science Research Unit, Institute of Education , University of London . ISBN: 978-1-907345-19-7]
Issues of weak methodologies and inadequate data have also been argued by D W Adams and J D von Pischke in their “Microenterprise credit programmes: déja vu” (1)
Citing B Armendáriz de Aghion and J Morduch's The Economics of Microfinance (2), the review said: “Overall, it is widely acknowledged that no well-known study robustly shows any strong impacts of microfinance.” “We find no robust evidence of positive impacts on women's status, or girl's enrolments […]”, said the review. As reason for having no evidence it said: “[T]his may be partly due to these topics not being addressed.” “Well-known studies which claim to have found positive impacts on females are based on weak research designs and problematic [Instrumental variable] analyses which may not have survived replication or re-analysis using other methods, i.e. PSM [Propensity score matching]”, said the review. It should be mentioned that the one of the overwhelming facts of microcredit has been the targeting of women as they are “better clients”.
Citing again The Economics of Microfinance (3), the review said: “Despite the apparent success and popularity of microfinance, no clear evidence yet exists that microfinance programmes have positive impacts.”
Referring to “four major reviews examining impacts of microfinance” (4) the review said: “These reviews concluded that, while anecdotes and other inspiring stories (H Todd, Women at the Centre , Dhaka: University Press Ltd., 1996) purported to show that microfinance can make a real difference in the lives of those served, rigorous quantitative evidence on the nature, magnitude and balance of microfinance impact is still scarce and inconclusive.” In making the observation it again referred to Armendáriz de Aghion and Morduch's study mentioned above.
The review said: “This can lead to misconceptions about the actual effects of a microfinance programme, thereby diverting attention from the search for perhaps more pro-poor interventions.” It suggested: “[I]t is of interest to […] engage with evaluation techniques and to understand their limitations, so that more reliable evidence of impact can be provided in order to lead to better outcomes for the poor.”
The review emphasized the need to “re-investigate existing microfinance impact evaluations”, and scrutinize more carefully the “robustness of claims that microfinance successfully alleviates poverty and empowers women”. The review “re-visited the evidence of microfinance evaluations focusing on the technical challenges of conducting rigorous microfinance impact evaluations.”
The original objective of this review was to assess the impact of microcredit. However, the study reviewed “not only ‘credit', but also ‘credit plus' and ‘credit plus plus' interventions”.
One of the substantial aspects of the systematic review is the methodology followed. The review based its “methods on the Centre for Evidence Based Conservation and EPPI-centre guidelines as these are suited to the quantitative and mixed methods used in microfinance evaluations.”
The review assessed validity of available evaluations. Initially it focused on the “intervention (e.g. provision of micro credit), the measurement of outcomes (e.g. income, expenditure, assets, health and education, empowerment, and so on) and contextual factors likely to affect differences in outcomes in different contexts, including other microfinance services.” It considered “different categories of persons (impact heterogeneity), and the potential existence, as well as the likely significance of factors which might confound observed relationships to undermine claims of a causal relationship with microfinance.” The review searched eleven academic databases, four microfinance aggregators and eight non-governmental or aid organization websites. The reviewers consulted bibliographies of reviewed books, journal articles, etc. They screened articles in two further stages, reducing 2,643 items to 58 which they examined in detail. They classified the research designs used in microfinance impact evaluations into five broad categories: randomized control trials (RCTs), pipeline designs, with/without comparisons, natural experiments and general purpose surveys. These five categories were cross-classified with three categories of statistical methods of analysis, which in descending order of internal validity are two-stage instrumental variables methods (IV) and propensity score matching (PSM), multivariate (control function) and tabulation methods.
“As very few RCTs were available,” the review included “many studies with weak designs that have been analysed with sophisticated methods”. They, however, noted that studies “in general weak design cannot be fully compensated by sophisticated analysis.” They “adopted a heuristic scoring of research designs and methods of analysis, combining these scores into a single value and defining a cut-off exclusion value.” A few articles that they “marginally exclude[d] by this approach were included based on [their] judgment, resulting in a final count of 58 papers.” They “devote[d] considerable attention to the most prominent with/without studies which have been highly influential in validating orthodox favourable views of microfinance impacts. These earlier studies have turned out to have low validity with replicated analysis and critical assessment.”
The review found “only two RCTs of relevance to [its] objectives”: “one has low-moderate and the other high risk of bias; neither finds convincing impacts on well-being.”
The study discusses “the two historically most significant studies (5) […], which, partly as a result of their prominence, have been replicated. The replications fail to confirm the original beneficent findings, and conclude that there is no statistically convincing evidence in these studies to either support or contradict the main claims of beneficence of microfinance. This is partly because of their weak research design.”
The original review question was “What is the evidence of the impact of microcredit on the incomes of poor people?” But the review suggested adjusting it to “What is the evidence of the impact of microfinance on the well-being of poor people?”
The study worked with a number of sub-questions: Evidence of impact of microfinance on (i) other money metric indicators such as microenterprise profits and revenues, expenditure, assets, and housing improvements; (ii) other human development indicators i.e., education, health, health behavior, nutrition; (iii) women's empowerment. While investigating these questions the review had to examine, where the evidence allows, the possibility of modification of the outcomes by (i) borrower's gender of borrower, (ii) household's poverty status, (iii) rural/urban setting, (iv) geographical location, (v) presence of second income earner in the household, and (vi) type of product.
The major problem identified in the review was that “few, if any, studies provide reliable evidence of impact using the criteria normally adopted in systematic reviews”. The review “found only two randomised controlled trials (RCTs) and could not conduct a meta-analysis.”
After the review, few questions get generated: why these topics were not “addressed in valid studies”, why the studies were “based on weak research designs and problematic analyses”, was there any rationale to make massive propaganda based on these sort of research, what were the stakes behind these sort of exercises? Was the propaganda beneficial to the poor? There are similar other fundamental questions. Does not the reality/real face of mainstream research come to light with this fact?
The reality microcreditors facing bears signs of accumulating problems beyond resolution. That “story” bears possibilities of coming to light with the development of debtor-creditor contradiction. Now, a mainstream study has put anew a question mark, and fundamentally and essentially that question questions mainstream knowledge system, its “scientific” method of enquiry, basis of its publicity-propaganda and credibility of all these “initiative”: microcredit and studies with weak methodologies.
Note:
(1) World Development , 20 (10), 1992)
(2) Cambridge : MIT Press, 2005
(3)2005 and 2010 editions
(4) J Sebstad and G Chen, Overview of Studies on the Impact of Microenterprise Credit , report submitted to USAID assessing the impact of microenterprise services (AIMS), June,1996; G L Gaile and J Foster, Review of Methodological Approaches to the Study of the Impact of Microenterprise Credit Programmes , report submitted to USAID AIMS, June, 1996; N Goldberg, Measuring the Impact of Microfinance: Taking Stock of What We Know , Grameen Foundation USA publication series, December, 2005; K Odell, Measuring the Impact of Microfinance: Taking Another Look , Grameen Foundation USA publication series, May, 2010; C E Orso, Microcredit and Poverty. An Overview of the Principal Statistical Methods Used to Measure the Programme Net Impacts , POLIS working paper No. 180, February, 2011
(5) M M Pitt and S R Khandker, “The impact of group-based credit programmes on poor households in Bangladesh : does the gender of participants matter?” Journal of Political Economy , 106 (5): 1998, and USAID funded studies in India , Zimbabwe and Peru . USAID produced three panel datasets, each with two waves in the late 1990s 2-3 years apart. The Indian study was on the microfinance operations of the Self Employed Women's Association (SEWA).
Dhaka based free lancer Farooque Chowdhury contributes on socioeconomic issues.
Comments are not moderated. Please be responsible and civil in your postings and stay within the topic discussed in the article too. If you find inappropriate comments, just Flag (Report) them and they will move into moderation que.
