The second post in our series described the importance of demand-side data for understanding consumers and their financial habits and needs. Various organizations are contributing to the global pool of demand-side data in branchless banking and in this post we’ll focus on two of the main sources. The Financial Inclusion Tracking Surveys (FITS) are annual household panel surveys in Uganda, Tanzania, and Pakistan while the Tanzania Mobile Money Tracker Study (TMMT) uses quarterly surveys to track market trends. Both are being carried out by InterMedia and the Bill & Melinda Gates Foundation. In this post, we’ll highlight analysis on rural and urban households to demonstrate the actionable insights that can be gathered from such datasets. We find that rural and urban populations in Tanzania and Uganda may be on the cusp of rapid mobile money growth, agents play differing roles in rural and urban areas of Uganda, and in Tanzania, rural and urban areas are experiencing differing growth rates as well as different use rates among women.
The FITS surveys reveal that 1 in 5 Ugandan households and 1 in 3 Tanzanian households have mobile money users. As we’d expect, the penetration amongst urban households is much higher than amongst rural households. Interestingly, the current urban penetration in these two countries (51% in Uganda and 65% in Tanzania) is significantly higher than M-PESA Kenya’s urban penetration back in 2008 (44%), when Billy Jack (of Georgetown University) and Tavneet Suri (of MIT) did their first round of household panel surveys covering M-PESA in Kenya. In 2008, M-PESA was undergoing exponential growth, which means Tanzania and Uganda’s urban areas could be at the start of a similar upward trajectory.
As the graph below shows, the harder-to-reach rural areas have much lower penetration than urban areas but again are on par with M-PESA Kenya’s 2008 penetration, meaning rural areas could also be set up for a trajectory similar to M-PESA.
Tanzania and Uganda have a much higher proportion of rural households than Kenya, however, leading to lower national penetration (35% and 20% respectively) than M-PESA’s back in 2008 (38%). If Tanzania and Uganda had the same urban/rural structure as Kenya, their overall penetration rates would be on par with M-PESA’s in 2009 and 2010 when it was well on its steep upward trajectory.
Although it took M-PESA much longer to penetrate rural areas, it succeeded and today 80% of rural Kenyan households use M-PESA. The very similar usage rates in rural and urban areas in neighboring Tanzania and Uganda compared with Kenya in 2008 are a positive sign that, given a few years to penetrate rural Tanzania and Uganda, mobile money will saturate rural areas in these countries as well.
The FITS survey
shows that in Uganda, agents play an especially important role in the uptake and use of mobile money in rural areas. Twenty-two percent of rural MTN users (the main provider in Uganda) began using because an agent recommended the services to them. This is second only to a friend’s recommendation.
Agents play an important role in Uganda’s urban areas as well but in different ways, with agent problems often discouraging use. Almost a quarter of mobile money users in urban areas had experienced a rude agent, compared with only 7 percent of users in rural areas. Problems with agents were also far more likely to go unresolved following an official complaint in urban areas than in rural areas. This indicates that in urban areas, better customer service among agents and improved problem resolution could improve the user experience.
The TMMT study
reveals differing trends in rural and urban areas in Tanzania. After no growth in the first part of 2012, mobile money use jumped in the third phase of the study by 6 percentage points, from 24 to 30 percent of all Tanzanian adults. Almost all the growth was in rural areas. These areas also saw increased exposure to advertisements and increased recognition of mobile money’s value-added services, such as using mobile money to save, for business transactions, and to pay bills (Figure 2). In contract, urban areas did not experience any growth.
Targeted marketing and a growing agent network are likely causes of these rural trends. Such findings indicate that, while rural areas are still far behind urban areas in terms of absolute usage, they are able to “catch up” with the right efforts.
The TMMT study also revealed gender differences in rural and urban areas. In urban areas, women are equally likely to use mobile money as men. However, in rural areas women are less likely to use the service. Given that rural women face different barriers, or have varying needs, we need to develop more nuanced marketing strategies, as well as creative products that better suit this segment.
The rich datasets gathered for these studies contain far more insight than what will fit in this post. We encourage all those who are interested in learning more to check out the Mobile Money Data Center
for access to the data and do-it-yourself analysis. We’ll also be adding datasets from additional studies and countries throughout the next year. Feel free to come back and let us know in the comments what you find in the data.
---- Michelle is a research manager at Intermedia and Claudia is a financial sector specialist at CGAP.