The Impacts of Microfinance: Evidence from Joint-Liability Lending in Mongolia

How to cite

Haas, Ralph De; Harmgart, Heike; Fitzsimons, Emla; Augsburg, Britta; Attanasio, Orazio. 2014. "The Impacts of Microfinance: Evidence from Joint-Liability Lending in Mongolia." AEA RCT Registry. AEARCTR-0000303. https://doi.org/10.1257/rct.303-3.0

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

Ralph De Haas* , Heike Harmgart , Emla Fitzsimons , Britta Augsburg , Orazio Attanasio

* Primary investigator

Registry ID
AEARCTR-0000303
DOI
https://doi.org/10.1257/rct.303-3.0
First registered
Last updated
Status
Completed
Trial dates
2008-02-01 – 2011-11-01
Countries
Mongolia (Uvs, Arkhangay, Khovsgol, Bulgan, Hentii provinces (aimags))
JEL codes
016, G21, D21, I32
Registry record
http://www.socialscienceregistry.org/trials/303

Abstract

We present evidence from a randomized field experiment in rural Mongolia on the comparative poverty impact of group versus individual microcredit. We find a positive impact of group loans but not of individual loans on entrepreneurship and food consumption. Moreover, group borrowers are less likely to make informal transfers to families and friends while the opposite holds true for individual borrowers. This suggests that joint liability may deter borrowers from using loans for non-investment purposes with stronger impacts as a result. We find no difference in repayment rates between both types of microcredit.

Intervention

Joint-liability group lending (without repayment meetings) in 15 villages, individual lending in 15 villages and ten control villages.

Experimental design

The experiment took place in 40 villages across five aimags (provinces) in northern Mongolia. The experiment started in February 2008 when XacBank loan officers and representatives of the Mongolian Women's Federation (MWF) organized information sessions in all 40 villages. After about 30 women had signed up in each village and had declared their interest in receiving a loan, a detailed baseline survey was administered during the second and third week of March 2008. The total number of women interviewed was 1,148. The face-to-face interviews were conducted by a specialized survey firm hired by the research team and independent of XacBank. At the time of the baseline survey we also collected information on the main socioeconomic, demographic, and geographic characteristics of the 40 villages. Randomization took place after completion of the baseline survey so that at the time of the interview respondents did not know whether they would be offered a group loan, an individual loan, or no loan at all. Village-level randomization was performed in a central location, using a random number generator in Stata, as decentralized and public randomization was not feasible given that the villages were spread over a large geographical territory. Randomization across rather than within villages was chosen because it was administratively and politically easier to manage. We stratified at the province level. After randomization, group formation proceeded in the 15 group-lending villages, but not in the individual-lending and control villages. Group formation consisted of the development of internal procedures, the election of a group leader, and the signing of a group charter. Groups were formed by the women themselves not by XacBank. A maximum of two women per group were allowed to be from the same family. After a group had collected enough internal savings it could apply for its first XacBank loan. The treatment period during which XacBank provided loans in the group and individual-lending villages lasted 1.5 years - from March 2008 to September 2009. During this period participating women in treatment villages could apply for (repeat) loans. In October 2009 we conducted a follow-up survey to measure again the poverty status and economic activity of the participating women. At this time, we also conducted a second village-level survey to collect information on village characteristics that may have changed, such as the prices of important consumer goods. XacBank collected repayment information on all loans for the entire period March 2008-June 2011. Lastly, in October 2011, we revisited one individual-lending and two group-lending villages for structured interviews and discussions with a number of borrowers about their experience of the lending programs.

Design details

Randomization method
Village-level randomization was performed in a central location, using a random number generator in Stata, as decentralized and public randomization was not feasible given that the villages were spread over a large geographical territory.
Randomization unit
Randomization across rather than within villages was chosen because it was administratively and politically easier to manage. We stratified at the province level.
Clusters
40 villages
Observations
1,148 women
Arms
Joint-liability group lending (without repayment meetings) in 15 villages, individual lending in 15 villages and ten control villages.
Primary outcomes
Loan take up, female entrepreneurship, profits, household income, consumption, asset ownership, labor supply.

Cited by

No citation data available for this registration yet.

Keywords