M-PESA Marches On

This is a guest post by Jake Kendall of the Gates Foundation

New data shows M-PESA is now reaching most of the poor, unbanked, and rural populations in Kenya – and measurably improving their lives.

M-PESA is a mobile-phone based electronic payment and store of value system which in just over three years has managed to acquire 57% of the Kenyan adult population as customers, and who between them do more money transfers domestically than Western Union does globally. Though money transfer services and savings accounts are nothing new, M-PESA’s real innovation is that customers can deposit and withdraw cash at any of 20,000 stores – that’s 20 times the number of bank branches in the country!

Many in the field of financial inclusion believe M-PESA has potential as a development tool because it significantly lowers transactions costs for intra-family remittances, commerce, utility bill payments and government welfare transfers and because it has the same functionality as a basic bank account, giving it great potential to expand financial inclusion. The numbers cited above have made M-PESA a sensation in the development field, but to date, many still wonder whether M-PESA is really reaching the poor, rural and unbanked; whether it would really be used to save and store money; or the extent to which it could really improve the lives of the poor.

New data

Well, new data has just come in, and the news is even better than we might have hoped. Two researchers (Billy Jack of Georgetown and Tavneet Suri of MIT, with funding provided by the Consortium on Financial Systems and Poverty, a grantee of the Bill & Melinda Gates Foundation) have new results here from a survey of 3000 Kenyan households who were queried once in August of 2008 and again in December 2009.

Their data show that adoption of M-PESA has continued apace going from 44% of households in 2008 to 70% in 2009 – and since December 2009 when the last wave of the survey was conducted, the number of M-PESA customers has grown 40% from 9m to 12.6m adults.

As M-PESA has spread to greater numbers of Kenyans, the true development potential of mobile money has begun to appear in the data. M-PESA has advanced down market to reach a majority of the poor, rural, and unbanked populations.  The share of poor households who have M-PESA has gone from 28% in 2008 to 51% in 2009 (Here, the poor are defined poorest 50% of Kenyan households who on average earn about $3.40 per capita per day, though similar advances have been made in the lowest 25%).  Similarly, the percent of rural households with an M-PESA account went from 29% to 59%, and the percent of unbanked using M-PESA has gone from 25% to 50%. We can now put to bed the question “will M-PESA reach the poor?” It has.

A place to save

M-PESA has also increasingly been seen as a place to save. The percent of users who save using M-PESA has gone from 76% to 81%. Most of the uptick in saving behavior is due to early adopters saving more over time, while the households who had already adopted M-PESA in 2008 (early adopters) went from 79% savers to 89% savers by 2009, only 70% of those who joined between 2008 and 2009 (late adopters) are saving. People seem to be starting with M-PESA as a money transfer service, and then moving toward using it as a savings option.

Preliminary results on impact

The researchers have also shared preliminary (unpublished) results with me that show that households who have access to M-PESA and are nearby an agent point are better able to maintain the level of consumption expenditures, and in particular food consumption, in the face of negative income shocks, such as job loss, livestock death, harvest or business failure, poor health, etc.  On the other hand, households without access to M-PESA appear to be less able to protect themselves from such adverse events.  The researchers have been careful to rule out many possible explanations based coincidence – i.e. they have good evidence that the M-PESA is the thing improving the households ability to keep their consumption stable – and they are currently investigating the precise mechanisms that underlie this. If these results survive further scrutiny, they will be the first rigorous quantification of the welfare benefits of mobile money.

These results are truly excellent news on almost every count. In just a few short years, M-PESA has reached over half the unbanked Kenyan population with a viable savings option and provides the poor with a tool that has a measurable positive effect on their welfare. This is really a major advance in the cause of financial inclusion and reconfirms my belief that mobile money products have the potential to transform the landscape of financial options available to the poor.