M-PESA
has made quite some headlines since its introduction in 2007. After its
launch, the mobile money technology became the payment method of choice
across Kenya, given that its use poses lower risks than informal
payment methods, storing money in mobile form implies lower risk than
holding cash, and using M-PESA for payment purposes costs less than bank
transfers. As of 2011, 70% of the adult population in Kenya had an
M-PESA account (Jack and Suri 2011, 2014) and M-PESA has contributed to
financial inclusion. There is evidence that the use of mobile money
increases the use of formal savings accounts and allows for more
effective risk sharing. In recent research, we show another important
channel through which mobile money can enhance economic development.
Namely, by allowing easier access to larger amounts of trade credit,
mobile money allows firms higher production, with important
macroeconomic repercussions.
M-PESA – a successful mobile money technology
In
Kenya, M-PESA is the most commonly utilised electronic money service
allowing users to send money to any cell phone owner via SMS messages.
Cash can be transferred into M-PESA deposits and vice versa via
specialised agents, which are widespread across the country. After being
introduced in 2007 by Safaricom, mobile money usage has grown rapidly.
As of December 2014, the number of total M-PESA agent outlets reached
124,000 and the number of customers 25 million (out of a total
population of 45 million). During 2013, 282.5 million transactions were
conducted in total, and the total value of money transferred was 1.9
million Kenyan shillings (US$22 billion) – equivalent to 40% of Kenyan
GDP in 2013. There are many reasons for this rapid growth and success
of M-PESA, including the dominant market position of Safaricom, the
aggressive marketing and expansion campaign in the early days, and the
need for payment services for families with members often spread across
the country (Eijkman
et al. 2010).
Data
In
this study we use the Kenya FinAccess Business Survey 2014, designed by
the Financial Sector Deepening Trust Kenya (FSD-K) together with
Tilburg University. It includes novel business mobile money usage
questions. The survey data were collected in 2014 by FSD-K from a
representative cross-section of 1,047 mainly small and medium
enterprises in Nairobi.
Using a formal regression analysis we explore which businesses are more likely to use M-PESA when buying supplies.
- The estimate shows that, ceteris paribus, businesses that purchase supplies on credit are 17 percentage points more likely to use M-PESA when purchasing inputs.
It is important to note that these estimates do not imply any
causality, but rather the result may imply that having a trade credit
relationship leads to mobile money usage between businesses and/or using
mobile money facilitates credit relationships between businesses.
Figure 1 illustrates this relationship.
Figure 1. Mobile money and trade credit

Figure
1 shows the predicted share of businesses considering mobile money as a
common method of payment to pay to suppliers according to whether the
businesses have credit relationship with input suppliers, and the 95%
statistical confidence levels for those shares after controlling for
other business characteristics.
A general equilibrium model
We
construct a model that explains the bi-directional causality between
the use of mobile money rather than cash when dealing with suppliers and
access to credit from these suppliers. Specifically, we model
entrepreneurs with access to a production technology that converts
supplier provided inputs into consumption goods. Suppliers sell inputs
to entrepreneurs in return for an immediate payment and/or for a credit
repayment to be made after the production is finalised. Access to
supplier credit, in turn, allows for a higher production scale.
Entrepreneurs only have access to supplier credit if they are part of a
network, which can enforce social sanctions in the form of excluding
defaulters from accessing trade credit. Such enforcement mechanisms
could be effective in a variety of contexts where formal institutions
are weak (Greif 1993, Platteau 2006). Especially, in the context of
development economics, the case for inducing repayment through
exclusion-sanctions has been well argued (Besley and Coate 1995).
On
the one hand, using cash for purchases and credit repayment carries the
risk of theft, which will not only negatively affect credit-constrained
entrepreneurs but will also limit access to credit for those
entrepreneurs who are connected to the network. Using mobile money, on
the other hand, comes with certain transaction costs but reduces the
risk of theft to zero.
Mobile money increases both the share of
entrepreneurs with access to and who are willing to use trade credit and
the amount of trade credit provided.
There are two reasons for
that. First, for a borrower the cost of theft is higher when purchasing
inputs, because if the initial endowment is stolen, the entrepreneur
suffers not only the endowment loss, but due to credit market frictions
the ability to borrow during that particular period. Similarly, for a
debtor at the trade credit repayment stage, theft is not only associated
with the loss of current value of cash but also the highly important
loss of future credit market access. The use of mobile money and the use
of trade credit thus complement each other, as the benefits from
avoiding the risk of theft are higher for trade credit borrowers. Lower
costs of the mobile money technology, and thus wider take-up, will
therefore broaden the use of trade credit and ultimately production of
firms.
We calibrate the stationary equilibrium of the model to
match a set of moments that we observe in the Kenyan FinAccess Business
Survey 2014. The parameterised model matches the Kenyan SME data well
along the dimensions that we calibrate. Most importantly, the
calibrated model predicts a similar fraction of M-PESA users among
entrepreneurs that use supplier credit as we find in the actual data,
thus supporting the assumptions and mechanisms of our theoretical model.
The economic effect
Using
the firm-level survey mentioned above and our model, we gauge the
economic significance of this effect. Comparing an economy with and
without mobile money, we find a difference in macro output of 0.47%.
Finally, to assess the economic importance of the mechanism we have
proposed, we calculate its contribution to economic growth of Kenya
since the introduction of M-PESA in 2007. Kenyan total factor
productivity (TFP) and real per capita income grew 3.3% and 14%,
respectively, between 2006 and 2013.
- The quantitative exercise result from the endogenous model implies
that M-PESA generates 0.5% TFP growth for the Kenyan economy through the
trade credit channel on an annualised basis.
This implies that the mechanism we have proposed can explain 14%
of TFP growth and 3.4% of per capita real income growth over the same
period, suggesting quite a large economic impact of mobile money
technology.
Conclusions
Our findings have
important policy implications. For a long time, the focus of the
financial inclusion debate has been on credit and savings services. Our
results contribute to an expanding literature that shows not only the
importance of effective payment services but also the promise that
digital payment systems can hold. Additionally, while an extensive
literature has focused on the lack of access to credit services by
enterprises as important growth constraint in developing countries, we
show the importance of effective payment services for expanding economic
and financial transactions in an economy.
References
Beck,
T, H Pumak, R Ramrattan and B Uras (2015), “Mobile Money, Trade Credit
and Economic Development: Theory and Evidence”, CentER Discussion Paper
2015-023.
Besley, T and S Coate (1995), “Group lending, repayment incentives and social collateral,”
Journal of Development Economics, 46(1), 1–18.
Eijkman, F, J Kendall, and I Mas (2010), “Bridges to Cash: The Retail End of M-PESA”,
Savings and development, 219-252.
Greif, A (1993), “Contract enforceability and economic institutions in early trade: The Maghribi traders’ coalition,”
The American economic review, pp. 525–548.
Jack, W and T Suri (2011), “
Mobile Money,” VoxEU.org 16 March
Jack, W and T Suri (2014), “Risk Sharing and Transactions Costs: Evidence from Kenya’s Mobile Money Revolution,”
The American Economic Review 104.1: 183-223.
Platteau, J (2006), “Solidarity norms and institutions in village societies: Static and dynamic considerations,”
Handbook on the Economics of Giving, Reciprocity and Altruism, 1, 819–886.
This article is published in collaboration with VoxEU. Publication does not imply endorsement of views by the World Economic Forum.
To keep up with the Agenda subscribe to our weekly newsletter.
Author:
Thorsten Beck is Professor of Banking and Finance at Cass Business
School in London. Haki Pamuk is a PhD student at the Department of
Economics at Tilburg University. Ravindra Ramrattan (1983 – 2013) was
the research economist at Financial Sector Deepening (FSD) Kenya between
2011 and 2013. Burak Uras is an Assistant Professor at Tilburg
University specializing in Financial Economics and Macroeconomics.
Image: A woman uses her Apple iPhone while waiting to cross 5th Avenue in New York. REUTERS/Lucas Jackson.