On the Impact of Microcredit: Evidence from a Randomized Intervention in Rural Ethiopia

Ethiopia is a poor developing country with deep scars of the civil war that ended in the 90s. While Ethiopia ranks among the countries that receive the most foreign aid, still there is little consensus on efficient ways of spending aid or of ways of making the Ethiopian economy less aid dependent. Typically, one suggested route is microcredit, which once seemed to be the holy grail of development, but now is increasingly viewed with skepticism.

In a recent Barcelona Graduate School of Economics working paper (No. 741) (October 2013), Alessandro Tarozzi, Jaikishan Desai, and Kristin Johnson investigate the impact of microcredits by exploiting the data of a large-scale clustered randomized controlled trial (RCT) conducted in rural Amhara and Oromiya (Ethiopia). The authors study the effects of micro-lending on income-generating activities, livestock ownership, schooling, and measures of women’s empowerment.

Data collection

The authors use the data gathered in study areas during a pre-intervention household survey in 2003 and a post-intervention survey completed three years later composed of about 6000 households each. Given that the interviewers questioned the same villages but not the same households in the follow-up sample the data constitutes a panel of villages, but not of households. The RCT was conducted in poor rural areas where borrowing was not common and agriculture and animal husbandry represented the bulk of the local economic activities.

Impact on borrower behavior

Borrowing increased substantially more in treated relative to control communities. More households borrowed and those that borrowed also borrowed more. The fraction of households with loans increased by 25% and the average loan was close to 80% of the poverty line in Ethiopia in terms of adult consumption per year. Rather than displacing other forms of borrowing, the introduction of micro-lending led to substantial relaxation of credit constraints. Female borrowing in the control areas was only 3%, whereas the estimates show an 11% increase in areas where microcredits where made available. Households with low socio-economic status were less likely to borrow from ACSI/OCSSC.

The increased uptake shows that the first purpose of microcredits, provision of scarce resources, seems to have been achieved. The question that remains is: did borrowing help households to improve their living conditions?

Impact on household economic activity doesn’t seem to extend beyond farm businesses

As a next step the authors look at what the loans were used for. The majority of the loans were utilized for productive purposes, such as to pay for ‘working capital’ or ‘basic investment’, in particular to fund crop cultivation or animal husbandry and to a lesser extent for ‘trading and services’.

It seems that indeed recipients tried to improve their productive capacities. However, no evidence could be found that microcredit led to more non-farm businesses or even female-led non-farm businesses. While this might be seen as a drawback in terms of the effectiveness of microcredit, the estimates suggest that net revenues increased threefold. However, the impact is found to be statistically insignificant. Access to micro-loans increased the stock of animals owned and the value of their sales.

Little to no effect on child labor or gender equality

Looking at the effects on children and other socio-economic outcomes the authors find that the estimated impact on schooling attendance of children is positive, but insignificant and not very large. Similarly, the effect on hours worked by children is small and not significant. Concerning the involvement of females in the decision making of the predominantly male-headed households no evidence of improvement could be identified.

A concerning increase in food insecurity

Surprisingly, access to microcredit is found to significantly increase the number of months of food insecurity by 0.7 months. One of the major criticisms of microcredit is the claim that it can drive poor households into a debt trap, in some cases even associated with suicide. While this would be an exaggerated jump to conclusion from the findings provided, still the fact that microcredit is associated with more food insecurity raises concerns.

So does microfinance transform societies?

Concluding, the authors point out that the introduction of microfinance seems not to have led to an increase in non-farm business activities, but revenues from such activities increased substantially. The results are overall consistent with the broad framework in the survey by Bannerjee (2013), where increased access to microfinance is associated with some improvements in living standards of beneficiary communities, but without compelling evidence of a true transformative power of microfinance.


Banerjee, A. (2013). Microcredit under the microscope: What have we learned in the past two decades, and what do we need to know? Annual Reviews of Economics 5, 487-519.