Saving through the Mobile: A Study of M-PESA in Kenya

Introduction

In the past five years, there have been two interesting shifts within the development finance community. The first is a focus on branchless banking models, which extend the reach of financial services to poor people who cannot be reached through traditional networks of bank branches (Lyman et al., 2006; Ivatury & Lyman, 2007; Ivatury & Mas, 2008). These models rely on third party outlets, such as retail shops or post offices, to act as agents and provide financial services beyond the physical bank branch. They also rely on technology infrastructure (POS, internet, mobile phones) to provide transactional security and to extend the reach of services. Secondly, there is recognition that credit may not be the appropriate tool for all households to manage their financial needs. Rather, appropriate savings mechanisms are also important to sustain and improve livelihoods.

These two trends are coming together. New strategies are being cultivated to mobilize savings via branchless banking models in general, and the mobile phone in particular. There is a recognition that savings, probably more than any other financial service, requires proximity and immediacy for the customer. Loans have built-in discipline mechanisms (regularity of payments, penalties in the event of non-payment), which make people willing to travel some distance or otherwise incur some personal cost to repay a loan that they had taken out in a time of need. But people may forgo the possibility of banking a portion of their cash receipts if doing so entails enough distance and inconvenience. If a formal savings vehicle is not available conveniently at or near the point where money is earned, it is likely that the money will either go into a range of competing uses or will remain stored informally. Therefore, there is increasing recognition that formal small-balance savings can only be mobilized effectively for the bulk of the population if branchless channels are leveraged.

Much of this recent interest towards savings via the mobile phone has been sparked by the stunning success of a Kenyan mobile money service called M-PESA. This service was introduced in March of 2007 by Safaricom, the country’s largest mobile operator. It has since acquired over 7 million users and extended its agent base to over 10,000. Although M-PESA was designed as a money transfer service there is evidence that it is also being used for savings. For example, a study funded by FSD-Kenya of over 3000 households in Kenya revealed that the application was being used to store money (GSMA, 2009).

The portfolio of savings mechanisms is shown in Figure 1. The study further found that M-PESA was one of the most popular mechanisms for savings amongst the sample, even beating out the mattress for M-PESA users. These results raise some interesting questions. In particular, why did M-PESA beat out the mattress and other popular savings mechanisms? Is M-PESA becoming a substitute for these other mechanisms? How are the savings patterns of the poor changing as they begin to store money with M-PESA? This chapter will address these questions. The chapter will begin by reviewing the conceptual arguments around the relevance of savings and mobile money in poor peoples’ daily lives. It will then examine the evidence from a collection of financial diaries, which captured the savings practices of 14 M-PESA users for a period of one month. These diaries were part of a fourteen month ethnographic study that examined the adoption, usage and impact of M-PESA in two locations: an urban slum and a rural village. Finally, the empirical findings will be used to suggest how mobile money applications can be designed to better suit the needs and interests of poor savers.

Figure 1.

Use of savings instruments in Kenya (Source: GSMA (2009), quoting study funded by FSD Kenya