Mudit Kapoor, Jonathan Morduch, and Shamika Ravi

From Microfinance to m-Finance

Innovations Case Discussion: M-PESA

In some countries it can take years to get a new telephone line installed. In 1990,
there were just 10 telephone lines installed for every 1000 people in the
Philippines. In Kenya, the ratio was 7 per thousand. In India, 6 per thousand.
Compare that with the United Kingdom with 441 lines per thousand in 1990, or
the United States with 545.1 For decades, public sector telephone companies in
developing economies seldom had incentives or budgets to rapidly expand land
line networks, and the private sector has had even less motivation to serve the cost-
ly-to-reach.

The advent of mobile telephones has changed the equation. Mobile technolo-
gies allow countries to leapfrog over existing technologies, and they are leaping
quickly. In 2006, an industry association counted 800 million telephones sold over
the prior three years in developing countries.2 In January of this year alone, Indian
companies added 7 million new subscribers.3 With cheap mobile telephones wide-
ly available, 40 percent of the population of the Philippines now has a telephone
and over 200 million text messages are sent daily—at just 2 cents a message.4 In
Africa, cell phone penetration jumped from 2 million subscribers in 1998 to 82
million by 2004, and the market continues to expand rapidly. One estimate places
penetration in Africa in mid-2007 at 160 million subscribers.5

The spread of mobile technology offers more than a cheaper way to provide
telephones. The technology also involves, in part, an expansion of human interac-
tion, changing the nature of interaction as well as its level. The “mobile” element
allows you to call your spouse when you’re stuck in traffic and running late; to keep
track of your teenagers when they go out with friends; to form a potentially life-

Mudit Kapoor is Assistant Professor of Economics and Public Policy and a faculty
associate of the Centre for Analytical Finance at the Indian School of Business (ISB)
in Hyderabad.

Jonathan Morduch is Professor of Public Policy and Economics at the Robert F.
Wagner Graduate School of Public Service at New York University. He is also Director
of the Financial Access Initiative, a new consortium of researchers at NYU, Harvard,
Yale, and Innovations for Poverty Action, focused on generating evidence to inform
policy on banking the unbanked.

Shamika Ravi is Assistant Professor of Economics and Public Policy at the Indian
School of Business (ISB) in Hyderabad and a fellow of the Microfinance Management
Institute.

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

From Microfinance to m-Finance

saving link in the case of natural disasters, like earthquakes or floods; and, in more
normal times, to reach colleagues, whether they’re at work or about town. In these
and myriad other mundane ways, mobile technology makes communication more
flexible, reliable, and cheap.

As the technology has spread, it took little time before thoughts turned to
whether mobile telephones could have explicitly economic applications.
Muhammad Yunus pitched Iqbal Quadir’s plan to sell mobile telephones to “tele-
phone-ladies” in rural Bangladesh as a way to multiply the reach of a single phone
(the purchasers then rented air-time to their neighbors).6 Today, nearly half of
Bangladesh’s villages have mobile access via a telephone-lady and over a quarter
million phones have been sold under the program.7 Yunus has spoken often of how
the technology could allow villagers to obtain better information on prices in local
markets and thus to improve their competitive positions when dealing with
traders. In this vision, access to phones increases earning capacity which in turn
makes access to Grameen Bank loans more powerful. In a similar way, a Vodafone
study noted that “62 percent of small businesses in South Africa and 59 percent in
Egypt said they had increased their profits as a result of mobile phones, in spite of
increased call costs.”8

Others are going much further in coupling telephones and banking. Mobile
telephones have been imagined as devices to complete banking transactions direct-
ly. M-Pesa in Kenya is one example (discussed by Nick Hughes and Susie Lonie in
this volume), joining similar innovations like G-Cash and SMART Money in the
Philippines and Suvidha/BEAM in India.9 It is easy to see the appeal: a Finmark
Trust report shows that one third of people in South Africa and Botswana who lack
bank accounts, for example, nevertheless own a mobile telephone or have access to
one (FinMark Trust 2004 and FinMark Trust 2005a). Many of the unbanked are
poor, and mobile technology offers the possibility of both filling financial gaps and
improving the economic lives of customers. The appeal of these “m-Finance” ini-
tiatives will grow with the spread of phones.10

The core of businesses like M-Pesa rests with facilitating financial transactions
via mobile telephones. In the most common application, microfinance customers
can pay loan installments via telephone by entering a code that transfers funds
from a personal account to the bank’s account. But transactions can also flow
between customers directly—as long as the two parties are in the M-Pesa network
(or have access to someone who is). Someone can, say, send money to an uncle
who is not in the network by sending a code to the uncle. The uncle takes the code
to a local shopkeeper who is in the M-Pesa network and the shopkeeper transfers
the funds. In Kenya, these transactions are facilitated by mobile phone agents in
commercial areas.

Banking via mobile phone thus offers features of automatic teller machines,
internet kiosks, and point-of-service devices like debit cards—technologies that
collectively represent a new wave of devices to bank low-income households. While
the focus below is on mobile phones, the other technologies can provide simpler
solutions, though less powerfully. Debit cards, for example, do not require owning

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

Mudit Kapoor, Jonathan Morduch, and Shamika Ravi

a mobile telephone yet offer ways to pay bills, make purchases, and save. ATMs
offer cash without requiring human interaction of any sort. Internet kiosks, which
are usually manned, can be a
platform for a range of finan-
cial transactions but are only
available in fixed locations.

[T]hrough the economist’s lens,
the loan officer is a tough guy
who makes sure rules are
followed. In reality, the loan
officer also plays the roles of
social worker, book-keeper,
mediator, detective, and coach.

Turning to mobile tele-
phones, the m-Finance appeal
rests on two possibilities. The
first is that mobile phones will
allow banks to do their exist-
ing business more cost-effec-
tively. By cutting costs, the
technology can make it feasi-
ble to reach a broader popula-
tion. The second and more
interesting possibility is that
mobile phones and related technologies will alter the nature of banking relation-
ships themselves. Doing so will mean going back to basics and rethinking assump-
tions about the financial demands of poor households.

THE HUMAN TOUCH

Robert Annibale, the head of Citigroup’s Microfinance Group, often notes that
there are two populations in the world who routinely meet with their bankers: the
very rich and the very poor. The very rich are served by private bankers ready to
invest in maintaining the relationships critical to keeping privileged customers
happy. The very poor, meanwhile, have microfinance.11 The typical mode of micro-
finance has been highly hands-on. As first rolled out at Grameen Bank, Bolivia’s
BancoSol, and at replicators worldwide, customers meet with credit officers in
groups each week or two. Group meetings stretch across an hour. Interactions are
face-to-face, and customers who come up short on their required payments must
deal directly and immediately with their neighbors and the bank’s representative.
When loans are successfully repaid in full and new loans are made, the group too
shares in the celebratory moment. It is a model built on a high level of “touch”. Day
by day, loan officers spend their discretionary time dealing with problem cases and
resolving disputes. In this way, reality departs sharply from the economic theorist’s
caricature, in which the loan officer is simply an enforcer of optimized loan con-
tracts: through the economist’s lens, the loan officer is a tough guy who makes sure
rules are followed. In reality, the loan officer also plays the roles of social worker,
book-keeper, mediator, detective, and coach. It is high-touch work.

Banking by mobile telephone—or leaning heavily on related technologies like
automatic teller machines and internet kiosks—is low-touch. A big question for
technology enthusiasts is whether much is lost in translation. It may be that the

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

From Microfinance to m-Finance

high degree of human touch, while valued by some, doesn’t ultimately contribute
much to successful microfinance. Perhaps people are happier without the group
microfinance meetings and prefer greater anonymity. Perhaps that’s not true (i.e.,
touch matters), but, all the same, gains in service quality due to the technology
outweigh losses. Or there may be hybrids in which traditional banking models are
combined with m-Finance.

RE-IMAGINING MICROFINANCE

Truly imagining the potential for m-Finance means re-imagining microfinance.
Muhammad Yunus created a narrative which helped to launch microfinance and,
in many ways, still guides efforts. The narrative rests most fundamentally on
“microcredit for micro-enterprise”: small loans to support the small businesses of
poor entrepreneurs. A first step in assessing possibilities for m-Finance is to open
that box. A growing pile of evidence shows that microfinance customers (includ-
ing Grameen Bank borrowers) use loans to meet widely-ranging needs; the list
includes paying for health emergencies, school fees, and putting food on the table.12
M-Finance holds the potential to create financial products that better fit with these
needs—as well as to create more flexible products to finance small businesses.

With m-Finance, for example, it would become easier for the microbank to
extend emergency loans to its clients. Consider a medical emergency which neces-
sitates that a m-Finance, the family might have to take an emergency loan from a
money lender (who might exploit the circumstances) or to lean on relatives. But
with m-Finance the family can get the loan or draw down savings while in the city,
assuming that their credit history is fine.

This simple example points to a larger way that m-Finance can expand the
nature of microfinance. The leading microfinance models have been highly atom-
ized: dominated at first by nongovernmental organizations which seeded village-
based or neighborhood-based organizations. Even as for-profit organizations have
edged in, the community focus remains. If a villager wants a loan, most often they
must get it from their village organization. If they want to save, they must also do
it locally. Loan officers come to the villages at set times and carry out transactions
locally. Unless you have made arrangements beforehand, if you are out of town,
you are out of luck.

M-Finance can increase the reach of microbanks. Imagine a case where a
microbank has financed a local retail shopkeeper. Once a week he needs to go the
city to make wholesale purchases. Given that the microbank is in the village, he
must carry cash into the city. Lack of security makes it expensive to carry cash,
though. Moreover, his transactions will be limited by the amount of money he has
in his pocket. With m-Finance he could go to the local branch of a bank in the city
to withdraw cash (as long as it is a part of the m-Finance network). Or, more sim-
ply, he could get cash from a networked shopkeeper. In the same way, he could
deposit surplus cash while in the city before heading home.

Consider the story of Ram, who at age sixteen was helping his father on the

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

Mudit Kapoor, Jonathan Morduch, and Shamika Ravi

family farm in rural South India. The farm is just a tenth of an acre, and income
from the farm could not support the family, especially with the prospect of the
upcoming marriage of one of Ram’s sisters. One day in the village, Ram met an old
friend, Shyam, who told him about the opportunities selling ice cream on Juhu
beach, a major tourist attraction in Mumbai. Shyam promised to help Ram to get
set up, but Ram needed an initial investment of Rupees 10,000 (approximately
$250). The local money lender refused to lend, fearing that Ram would run away
with the money. An uncle could not help either. But the uncle introduced Ram to
his employer, and after some persuasion the employer lent Ram the money on the
condition that the loan be repaid within the year in ten monthly installments. Ram
agreed and set off for Mumbai. Today, he is a successful ice-cream vendor; he has
re-paid the initial loan and sends money
regularly to the family by post office money
order.

Truly imagining the
potential for m-Finance
means re-imagining
microfinance.

Now consider the story with m-Finance.
With the ability to transfer funds by tele-
phone, Ram could remit back to the village
instantly, safely, and more frequently. He
might also become a viable prospect as a
microbanking customer in his village. Thus,
he might be able to get a loan in the village
information networks are
where his
strongest—but could repay the loan from Mumbai. Similarly, m-Finance could
create new business opportunities for villagers. Consider a village artisan who sells
her produce in the local village market. She has secured a loan from a local
microbank and wants to expand her business, but the lack of a marketing infra-
structure prevents her from expanding her business to urban centers where profit
margins are much higher. The local microbank hesitates to lend money to the
entrepreneur even though they know from past experience that she is reliable. A
big constraint for the microbank is imposed by the requirement of weekly meet-
ings and repayments. With m-Finance, however, the entrepreneur can repay from
a distance, at least for part of the time, by making repayments from afar. She can
thus earn extra profit while maintaining good standing with the microbank.

In similar ways, m-Finance can facilitate internal remittances. While interna-
tional remittances get increasing attention, within-country remittances also play
important roles in poverty reduction. The Coalition for the Urban Poor (CUP) in
Bangladesh, for example, estimates that migrant workers in Dhaka send approxi-
mately 60 percent of their income to relatives.13 The flows form a significant part
of household budgets and support education, crisis management, and the other
exigencies of daily life. Many migrant workers, though, lack access to formal banks
offering remittance services and must rely on informal mechanisms which tend to
be time-consuming and insecure.14 M-Finance can fill the gap by providing cost-
effective, secure and fast remittance services as a matter of course.

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

From Microfinance to m-Finance

Overcoming these hurdles will be made easier by the shift away from the prac-
tice of “group lending” in microfinance. In South Asia, especially, loans are deliv-
ered through groups—particularly amongst poorer households who typically bor-
row smaller loans—but rigid adherence is softening.15 The shift away from groups
stems from the stresses attached to group guarantees, coupled with the fact that
group meetings consume time. In regions with low population density groups may
also be costly to attend: one microfinance initiative in China eventually closed as
customers dropped out due to the high costs of group meetings.16 Grameen Bank
has sworn off the idea of holding customers liable for the troubles of fellow group
members. So has ASA, a large competitor in Bangladesh. BancoSol now has almost
no loans under group guarantees. CARD in the Philippines is also now experi-
menting with dropping group guarantees. The trend is toward individual loans
with individual responsibility for repayment, and this opens possibilities for m-
Finance given its personalized, individualized mode.

That said, group meetings have advantages. The meetings can be social occa-
sions, and the groups can offer support to customers —economic, emotional, and
strategic. The meetings may also be venues for training and for marketing new
products. The public aspect of loan repayments may also help ensure timely repay-
ments. The local elements and the use of groups also allow loan officers to effi-
ciently acquire “soft information” on customers: Who is energetic? In a stable fam-
ily relationship? Sensitive to peer pressure? Entrepreneurial? Shady? Having a hard
week? Over-extended? The shift to m-Finance facilitates easy access to “hard infor-
mation”—especially on the history and timing of credit and saving transactions—
but at the expense of these kinds of soft information. One question is whether the
hard information, once fed into credit-scoring models and the like, can adequate-
ly substitute for the soft information. Or whether it’s possible to create structures
in which m-Finance is integrated with traditional modes of microbanking—so
that important aspects of “touch” (and soft information) are retained.

The regular repayment schedules favored by most microbanks have advan-
tages, but they add to costs. Most microbanks insist on frequent meetings (and fre-
quent repayments of loan installments) in order to closely monitor clients, provide
necessary training and to pick up ‘early warning signs’of trouble. But, to the extent
it is needed, the monitoring is most helpful in the early phases of relationships.
One possibility is to phase in m-Finance gradually, such that successful clients
“graduate” to individualized mobile phone-based operations.

M-Finance also permits the possibility of less frequent repayment schedules.
Research shows that low-income borrowers tend to prefer to synchronize the tim-
ing of loan repayments to the timing of income inflows.17 So far it has proved cost-
ly for the microbanks to provide fully customized services to its clients, but with
m-Finance it becomes easier. For example, a fruit vendor with a daily flow of
income might prefer to repay loans in small bits on a daily or weekly basis. With
m-Finance technology he can do so, repaying in whatever form he likes (daily,
weekly, etc) as long as the microbank’s targets are met on time (say, that a given
amount be cumulatively repaid monthly by a given date). Alternatively, he could

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

Mudit Kapoor, Jonathan Morduch, and Shamika Ravi

save the income in a personal account and simply transfer the full amount to the
microbank at regular intervals.

[T]he future of m-
Finance will rise or fall
with its cost implications.

An additional prospect for m-Finance is that the telephone can be used to
remind customers of upcoming deadlines—an application that does not rely on
the other parts of the banking infrastruc-
ture. Customers can be aided in saving
through friendly reminders, for example.
Or they can be advised of upcoming
dates for loan repayments, payment of
insurance premiums, utility bills, etc. In a
pilot study in the Philippines, a simple
reminder to save appears to be making a
positive impact on accumulated savings.
In this example, mobile telephone technology replaces the human touch—and
may even be superior to it.

EXPANDING ACCESS

The promise of m-Finance to expand to the unbanked rests in large part on the
specifics of the business model. In the case of M-Pesa in Kenya, the partner
microbank has not yet developed a back-end operation that allows m-Finance
transactions to be handled automatically and electronically. Cost savings (and
functionality) are thus limited. But cost savings are one of the big promises of m-
Finance. Developing workable business models has been challenging for
microbanks (even for organizations relying in part on subsidy), and the future of
m-Finance will rise or fall with its cost implications.

Microbanks have found that the costs of high-touch operations for tradition-
al microbanks have been hard to spread over small-sized loans. Interest rates thus
have often been high (often 30 percent per year and higher), especially for institu-
tions serving the poorest customers. In The MicroBanking Bulletin’s latest data on
386 microbanks, personnel expenses account for more than half of all operating
costs. Moreover, personnel costs (expressed as a fraction of assets) are 50 percent
higher (12.2 percent versus 7.9 percent) for microbanks focused on the poorest
customers versus those focused further up market.18 A hope with m-Finance is to
speed up routine processes so that field staff can focus more heavily on problem
areas and new opportunities. A second hope is to cut the costs of dealing in small
quantities.

A third hope is that m-Finance can help expand operations in remote or
sparsely-populated rural areas.19 But one piece of evidence from the M-Pesa expe-
rience offers a cautionary note about outreach to poorer customers. We begin by
noting that, at present, the charge for sending a text message by telephone is 5
Kenya shillings. Customers must also pay the bank a fee of 30 shillings per trans-
action. Consider the case in which a typical small loan of 20,000 shillings is repaid

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

From Microfinance to m-Finance

weekly over a year in 400 shilling increments. The cost of each transaction for the
customer is thus 35/400 shillings, or nearly 10% of the loan size. The customer
pays interest on top of these fees. The big question here is whether the automation
inherent in m-Finance can eventually bring a compensating drop in interest rates.
Or if the added convenience makes the extra cost worthwhile.20 If not, shifting to
m-Finance could screen out poorer customers.

M-Finance will not undo all the trade-offs of microfinance as practiced today.
Outside of South Asia, microfinance fails to reach very poor households in sub-
stantial numbers. Despite innovation, costs (and thus interest rates) remain rela-
tively high, particularly in Latin America. Still, microfinance proves helpful for
customers, even if it is not the panacea that its most ardent advocates promise. M-
Finance has the potential to take microfinance beyond the bounds of villages and
neighborhoods and create banking services that move with customers. And it can
be a platform for developing new products like emergency loans and flexible sav-
ings accounts. The potential for these applications is ample, as is the demand. M-
Finance will only succeed, though, if it rests on a deep and realistic understanding
of the financial needs, constraints, and opportunities of poor households.

Acknowledgements

We have benefited from discussions with Gautam Ivatury, Kabir Kumar, and Don
Johnston and comments from Javier Bronfman. Aparna Dalal provided expert
assistance. Morduch is grateful for funding from the Bill and Melinda Gates
Foundation through the Financial Access Initiative. The views are those of the
authors only.

1.United Nations Development Program, Human Development Report

2006.

2. GSM Association (2006). “GSM Hits Two Billion Milestone.” London: GSM Association. 16 June.
3. Gautam Ivatury. Interview in CGAP Portfolio, March 2007.
4. Owens, John and Bantug-Herrera, Anna. “Catching the Technology Wave: Mobile Phone Banking
Available at:

and Text-a-Payment

Chemonics International.

in the Philippines.”

5. Kabir Kumar, Consultative Group to Assist the Poor, interview (April 26, 2007).
6. Knowledge@Wharton, “Microfinance 2.0”, Forbes, April 10, 2007.
7. Yunus, Muhammad (2007). “Grameen Bank At A Glance.” February. .

8.Vodafone, “New research reveals economic and social benefits of mobile communications in
Africa” (Press release). March 9, 2005. Available at: .

9.Nick Hughes and Susie Lonie. “M-PESA: Mobile Money for the ‘Unbanked.’” Innovations, this
issue. See also the Safaricom website and .

10. A rich survey of current trends in m-Finance, with an eye to regulation, is available in David
Porteous, “The Enabling Environment for Mobile Banking in Africa: Report Commissioned by
Department for International Development (DFID).” Bankable Frontier. Available from:
.

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023

Mudit Kapoor, Jonathan Morduch, and Shamika Ravi

11. To be precise, when the very poor have bankers, they are often from microbanks; microfinance
today reaches only a small share of the global poor and, by one count, just over 100 million cus-
tomers worldwide. Sam Daley-Harris (2006). State of the Microcredit Summit Campaign 2005.
Washington, DC: Microcredit Summit Campaign. The surprising parallel between modes of
banking the very rich and the very poor was pointed out by Thomas Easton in his survey of
microfinance in the November 5, 2005 Economist magazine.

12. See, for example, Stuart Rutherford, “Uses and users of MFI loans in Bangladesh,” MicroSave
Briefing Notes on Grameen II #7. 2006. Also: Don Johnston and Jonathan Morduch,
“Microcredit versus Microsaving: Evidence from Indonesia,” paper presented at World Bank
Conference on Financial Access, March 2007.

13. See http://www.asia2015conference.org/pdfs/Deshingkar.pdf.
14. ICICI bank is conducting a household survey in several states in India to understand the extent

of internal remittances and develop appropriate financial services.

15.Beatriz Armendáriz and Jonathan Morduch (2006). The Economics of Microfinance. Cambridge,

MA: MIT Press.

16.Albert Park and Changquing Ren (2001). “Microfinance with Chinese Characteristics,” World

Development 29(1): 3 – 62.

17. Armendariz de Aghion, Beatriz and Jonathan Morduch (2000). “Microfinance Beyond Group
Lending,” The Economics of Transition 8 (2) 2000: 401 – 420. Shamika Ravi (2007), “Repay as
you Earn”, Indian School of Business, draft manuscript.

18. MixMarket (2006). “Benchmark Tables,” Microbanking Bulletin. Autumn. Available at:

19. Mitchell A. Petersen and Raghuram G. Rajan (2002). “Does Distance Still Matter? The
Information Revolution in Small Business Lending,” Journal of Finance, 57(6): 2533-2570.
December.

20. The example draws from our reading of the terms and conditions of the Safaricom/M-Pesa con-
tract, described at
(accessed May 4, 2006).

innovations / winter & spring 2007

Downloaded from http://direct.mit.edu/itgg/article-pdf/2/1-2/82/704165/itgg.2007.2.1-2.82.pdf by guest on 08 September 2023
Download pdf