The Impact of Mobile Money on Poor Rural Households: Experimental Evidence from Uganda

How to cite

Bruhn, Miriam; Wieser, Christina. 2019. "The Impact of Mobile Money on Poor Rural Households: Experimental Evidence from Uganda." AEA RCT Registry. AEARCTR-0004605. https://doi.org/10.1257/rct.4605-1.0

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

Miriam Bruhn* , Christina Wieser

* Primary investigator

Registry ID
AEARCTR-0004605
DOI
https://doi.org/10.1257/rct.4605-1.0
First registered
Last updated
Status
Completed
Trial dates
2015-09-01 – 2018-02-28
Countries
Uganda (Africa)
JEL codes
C93; E42; G23; O12; O18
Registry record
http://www.socialscienceregistry.org/trials/4605

Abstract

We ask whether poor rural households benefit from better access to mobile money agents. In a randomized experiment, we assigned 168 areas in Northern Uganda to receive an agent in 2017, with 163 areas serving as a control group. Data from a 2018 household survey show that the agent rollout doubled the nonfarm self-employment rate, from 3.4 to 6.4 percent, and reduced the fraction of households with very low food security from 62.9 to 47.2 percent, in more remote areas. These increases appear to be due to more peer-to-peer transfers and cost savings for remittance transactions.

Intervention

The intervention consisted in rolling out mobile money agents for Airtel Money. The rollout took place in treatment clusters between January and June 2017. A professional services firm was hired to assist with identifying potential agents. This firm also helped agents with the logistics of signing-up to become Airtel Money agents and provided them with the necessary equipment, training, and marketing materials.

Experimental design

The study was conducted in rural areas in the North of Uganda, covering the three regions West Nile, Mid-North and Karamoja. The Ugandan Bureau of Statistics (UBOS) helped define the study sample. As a starting point, all rural enumeration areas in all sub-counties in the West Nile, Mid-North and Karamoja regions were used. From this list, sub-counties that had an Airtel Money agent were dropped, based on an agent database provided by the Bill and Melinda Gates Foundation. A random sample of 1,200 enumeration areas (EAs) was drawn from all rural enumeration areas in the remaining 249 sub-counties using population proportional to size sampling. The sample was stratified by sub-county to ensure that it included enumeration areas from all sub-counties.

In September and October 2015, a household listing exercise was conducted in the EAs selected for the study to generate a sampling frame that could be used for the baseline survey. At the same time, all businesses in each EA were listed to gain information on potential mobile money agents in the EA. To minimize potential spillovers of agents to control group EAs, the selected EAs were mapped and grouped into clusters. A 0.5km buffer was drawn around the boundary of each EA and EAs whose buffers overlapped were grouped. Clusters were thus at least 1km apart from each other.

In this process, some EAs were dropped from the sample since either the listing exercise could not be conducted (for logistical or security reasons), the listing did not yield any businesses and thus no potential agents, or the maps were missing from the software to group EAs into clusters. As a result, there were 929 EAs that were grouped into 658 clusters.

Using a computer assisted stratified randomization approach, 329 clusters were assigned to a treatment group and 329 clusters to a control group. The treatment clusters formed the list of EA clusters where Airtel Money agents were to be rolled out. No such agents were to be rolled out in control clusters.

Design details

Randomization method
The randomization was done in office by a computer. The randomization strata were based on three variables. First, a variable equal to one if the cluster included more than one EA (which was the case for 16 percent of clusters) and equal to zero if the cluster included only one EA. Second, a variable equal to one if the distance to the nearest bank branch was greater than the median distance across all clusters (25.2km) and equal to zero otherwise. These bank branches refer to Airtel partner banks and the distance was calculated as the closest distance from any point on the cluster boundary. The main reason for stratifying on this variable was that Airtel raised concerns as to whether agents that are further away from these banks can operate well since they may have a harder time maintaining liquidity. On the other hand, distance to an Airtel partner bank is likely also correlated with distance to urban centers and other financial services, so that clusters that are further away from a branch may see greater benefits from obtaining an agent. The third stratification variable was based on a strategic priority
Randomization unit
Enumeration area clusters
Clusters
658 enumeration area clusters
Observations
9600 households
Arms
329 enumeration area clusters control, 329 enumeration area clusters treatment
Primary outcomes
Agricultural investments, non-farm self-employment, food security, poverty
Secondary outcomes
Peer-to-peer (P2P) transfers, savings

Cited by

No citation data available for this registration yet.

Keywords