Impact Evaluation of a Microcredit Program in Morocco

How to cite

Duflo, Esther; Haas, Ralph de; Crépon, Bruno; Devoto, Florencia; Parienté, William. 2014. "Impact Evaluation of a Microcredit Program in Morocco." AEA RCT Registry. AEARCTR-0000371. https://doi.org/10.1257/rct.371-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Esther Duflo* , Ralph de Haas , Bruno Crépon , Florencia Devoto , William Parienté

* Primary investigator

Registry ID
AEARCTR-0000371
DOI
https://doi.org/10.1257/rct.371-2.0
First registered
Last updated
Status
Completed
Trial dates
2006-03-01 – 2014-12-31
Countries
Morocco (Rural areas, nationwide )
JEL codes
O12, O16, G21, I32
Registry record
http://www.socialscienceregistry.org/trials/371

Abstract

This randomized evaluation measures the impact of a microcredit program introduced in rural areas of Morocco starting in 2006 by Al Amana, the country’s largest microfinance institution. The study takes advantage of Al Amana's expansion into rural areas with no previous microcredit penetration, before the introduction of the product and for the duration of the study. We designed a sampling strategy that allows us to estimate the impact on borrowers, and also to capture impacts representative at the village level. Within the catchment areas of newly-opened branches, 81 pairs of matched villages were included in the study. The treatment villages, randomly selected within each pair, received access to Al Amana loans immediately, while villages in the control group were only offered microcredit two years later. Following Al Amana's timeline of branch openings between 2006 and 2007, a baseline survey grouped in four waves was administered to collect data on socio-economic characteristics, household production, members’ outside work, consumption, credit and women’s role in the household. A follow-up survey conducted two years later found that, among the thirteen percent of households in treatment villages who took a loan, microcredit had large, albeit heterogeneous, impacts on assets and profits from self-employment activities, but that any potential gains in consumption were offset by a reduction in income from casual labor. A third round of data collection is currently underway to measure whether, now that a much longer time period has elapsed, the investment in business assets paid off in the longer run.

Intervention

Al Amana, the largest Moroccan microfinance organization, opened around 60 new branches in non-densely populated rural areas of Morocco between 2006 and 2007. This expansion of microcredit into previously un-served areas created a favorable context for the use of an experimental methodology to evaluate the impact of the access to microcredit. Among the new villages now covered by Al Amana, 81 sites or “pairs” located at the periphery of the catchment area of 47 new branches were included in the study. Within each pair, the village to receive microcredit services just after the branch opening and the village to be served two years later were randomly selected.

In treatment villages, credit agents started to promote microcredit and to provide loans immediately after the baseline survey. They visited villages once a week and performed various promotional activities: door-to-door campaigns, meetings with current and potential clients, contact with village associations, cooperatives and women's centers, etc. The main product Al Amana offers in rural areas is a group-liability loan. Disbursed amounts range from 1,000 to 15,000 MAD (124 to 1,855 USD) per member. It can take three to 18 months to reimburse loans, through payments made weekly, twice a month, or monthly. For animal husbandry, a two-month grace period is granted. Interest rates on rural loans ranged between 12.5% and 14.5% by the time of the follow-up study (i.e. from 2006 to 2009).

Experimental design

The design of our study tracked the expansion of Al Amana into rural areas between 2006 and 2007. Before each branch was opened, data was collected from at least six villages located on the periphery of the intervention areas - villages that could either have been included or excluded in the branch's catchment area. Selected villages were then matched in pairs based on observable characteristics (number of households, accessibility to the center of the community, existing infrastructure, type of activities carried out by the households, type of agricultural activities). In each pair, one village was randomly assigned to treatment and the other to control.

Our sampling strategy followed a novel approach to maximize the evaluation's power to detect both direct and population-level effects of microcredit access. Specifically, we selected two samples of households: one containing those with the highest probability to become clients of the microfinance institution and one containing a random selection of households from the rest of the population. Using the first sample increases the probability to detect an effect on those who are the most likely to borrow, if there is one. Using both samples together, with appropriate weights, allows us to measure the effect on the whole population of offering access to microcredit.

Design details

Randomization method
In office with random number generation in Stata
Randomization unit
Village
Clusters
162 villages, divided into 81 pairs of similar villages
Observations
5,551 households (4,465 households at baseline, plus 1,433 households added at follow-up)
Arms
81 villages control, 81 villages treatment
Primary outcomes
Access to credit, income levels and composition, labor allocation, consumption, education and female empowerment.

Materials

Papers from this trial

  1. https://www.povertyactionlab.org/sites/default/files/publications/47_97%20Crepon%20et%20al%20Morocco%20Jan2015.pdf

Cited by

No citation data available for this registration yet.

Keywords