MARISCAL, JUDITH; FLORES-ROUX, ERNESTO - The enigma of mobile banking

1. The enigma of mobile banking
Judith Mariscal
Ernesto Flores-Roux
IV Congress “ICTs and its Social and Economic
Impact in the Americas”
Brasilia, May 15, 2010

2. Mobile money systems: “holy grail” to bank the unbanked...
“The potential of mobile phones to revolutionize access to financial
services in developing countries is exemplified powerfully by the
success of M-Pesa moile money service in Kenya”
GSMA, Bill & Melinda Gates Foundation, “Seeking
Fertile Grounds for Mobile Money”, 2009
“New uses for mobile phones could launch another wave of
development.”
“The mobile service that is delivering the most obvious economic
benefits is money transfer, otherwise know as mobile banking.”
The Economist, September 26th-October 2nd , 2009
2

3. ... but so far the global experience is patchy
Source: Datamonitor, taken from Melin, Niclas (Marketing Director Revenue Management, Ericsson), no date specified, at
http://telecomcity.se/Upload/DocumentManager/MobileMonday/MobileMoney.pdf 3

4. The two most cited examples are M-Pesa in Kenya...
 Launched in 2007 by Safaricom (35% owned by the State, 40% Vodafone, 25% public)
 Services offered:
― Top up
― Cash in/cash out
― Domestic transfers/remittances
― International transfers (only from UK; recently launched)
― Bill payment* (moving into actual purchases)
 In 14 months, it had 2.7 MM users and 3,000 agents
 In less than 3 years, it had 7 MM users and 10,000 agents
 Based on a simple application on the phone and SMS with PIN
 Maximum amount s allowed: KShs. 35,000 (USD 450) per transaction, KShs. 70,000 per day
 Tariffs**:
― Transfers (intra M-Pesa): USD 0.50
― Transfers (non M-Pesa): 1%-4$ (minimum
charge USD 1.00)
― Cash withdrawal: 0.5%-2.0%
* Schools, financial institutions, some government agencies, heald management, insurance agencies, NGOs and religious
organizations, transportation, utilities 4
** Tariffs ranges are for typical amounts; very small transfers (e.g. KShsh. 100 [USD 1.3] are allowed with a hefty charge of KShs. 30

5. ... and Gcash and SmartMoney in the Philippines
 Launched in 2003 (SmartMoney) and 2004 (Gcash)
 Smart partnered with Banco de Ouro, Gcash has no banking partner
 Mostly based on SMS
 Services offered
― Top up
― Cash in/cash out
― Domestic transfers/remittances
― International transfers (original killer application; now Smart partnered with
MoneyGram)
― Bill payments
― Purchases (physical and online)
― Loan disbursements
 Tariffs:
― Transfers (intra M-Pesa): USD 0.50
― Transfers (non M-Pesa): 1%-4$ (minimum
charge USD 1.00)
― Cash withdrawal: 0.5%-2.0%
5

6. OiPaggo in Brazil has made some inroads, but it has not
achieved enough scale
 Launched in 2007
 Only offered in 12 cities in the Northeast of Brazil
 Credit based (ie, need preapproval): work under a credit card administration regulatory
regime
 Oi Paggo assumes all the risk of non-payment
 Products
― Top up
― Purchases (physical)
― Transfers (in pilot phase)
― Payments (utilities, in pilot phase)
 Built a merchant network of over 75,000 establishments
 User base of approximately 250,000
 Charge retailers typical credit card feed (2.99%) for purchases, no monthly fees nor POS
device rental
6

7. Conceptually, mobile money systems are very simple
Money in Money out
Network
(inlet) (outlet)
 Mobile money systems require a way for money (usually cash) to enter the system and a way
for money to exit the system
 The system is usually though of as the “mobile network”
In reality, the ecosystem is
significantly more complex
7

8. Building the “inlet network” is a significant channel in itself,
as it requires recruiting and managing thousands of “points”
Inlet network
Mobile company’s
own network
Merchant network Money out
Network
(outlet)
Banks
Other
correspondants
 “Ubiquity” (decent footprint)
 Cash-management skills (for inlet, mainly
Building the inlet channel safety)
implies building a (physical)  Right incentives (fees)
network of thousands of  Simplicity of transaction
points where money can  Reasonable homogeneity of network
come into the system (standardized process, branding, fees, etc.)
[predictability]
8

9. More challenging than building the inlet network is building a
good value proposition for the outlet network
Key variables that influence
Driver the driver
 Widespread  (Mobile company’s typical -
Airtime value proposition)
 User-friendly  System
Payments
Degree of difficulty
 Receiving network in place  Regulation
(“chicken and egg problem”)  Scale of accepting network
 System considered safe (trust)  Security/safety
Cash  Use of cash is “dangerous”  Migration
 Distance (outlet at different place  Brand equity of provider
from inlet): Transfers and
remittances
 Except for temporary repository  Regulation (banking)
Savings of cash, not explored  Banking penetration +
 Regulation (enters heavily into
banking arena)
Certain combinations of these key variables are
esential for determining success or failure
9

10. More holistically, the complete ecosystem is quite complex
Credit
Mobile company’s
own network
Airtime
Merchant network Closed
“e-wallet”
system
(repository)
Banks (top-up)
Savings
Other
correspondants
Payments Cash
Mobile company’s
Merchant network own network
Utilities/ Merchant network
Government
Banks
Loan payments
Other
correspondants
10

11. M-Pesa has concentrated on “cash in – cash out” (transfers)
Credit
Mobile company’s
own network
Airtime
Merchant network Closed
“e-wallet”
system
(repository)
Banks (top-up)
Savings
Other
correspondants
Payments Cash
Mobile company’s
Merchant network own network
Utilities/ Merchant network
Government
Banks
Loan payments
Other
correspondants
11

12. In the Philippines, all players have extended into most parts of
the ecosystem, except credit and savings
Credit
Mobile company’s
own network
Airtime
Merchant network Closed
“e-wallet”
system
(repository)
Banks (top-up)
Savings
Other
correspondants
Payments Cash
Mobile company’s
Merchant network own network
Utilities/ Merchant network
Government
Banks
Loan payments
Other
correspondants
12

13. Oi Paggo is currently building its offer based on credit
Credit
Mobile company’s
own network
Airtime
Merchant network Closed
“e-wallet”
system
(repository)
Banks (top-up)
Savings
Other
correspondants
Payments Cash
Mobile company’s
Merchant network own network
Utilities/ Merchant network
Government
Banks
Loan payments
Other
correspondants
13

14. Literature has identified certain enablers of mobile payment
systems
Decent mobile  Footprint/coverage
infrastructure  Penetration
 High penetration of formal financial Scalability
Low penetration of services (savings, credit, transfers
formal financial and payment options) substantially  Success depends on
services diminished the value of mobile volume
banking  “Chicken and egg”
problem
 Determines what can be done and
who can do it  Virtuous vs. vicious
cycle
 Flexible regulations allow for more
Regulation entrepreneurship, risk taking and  “Scalability” is
innovative ideas determined by these
four “enablers”
 Determines the extent to which a
partnership with a bank is required
 Intrinsically linked to the
Inlet/outlet convenience of mobile money
partnerships systems
 Basis for “scalability” 14

15. The first enabler – reasonable wireless infrastructure and use –
does not appear to be a determining factor. M-Pesa began offering
mobile money services when penetration was around 20%
Wireless penetration
Year 1 = Year of launch of mobile money services
100
Brazil
South Africa
90
80
Philippines
70
60
50
Kenya
40
30 Tanzania Afghanistan
20
Colombia: 93.2%
10 Mexico: 78.2%
Rwanda: 24.7%
0
1 2 3 4 5 6 7
* The “Strength of legal rights index” measures the degree to which collateral and bankruptcy laws protect the rights of borrowers
and lenders and thus facilitate lending. The Legal Rights Index ranges from 0-10, with higher scores indicating that those laws are
15
better designed to expand access to credit; Source: WorldBank, 2010, “Doing business: Measuring business regulations”

16. In countries where m-money initiatives have been launched,
penetration of financial services has varied widely
Banked population
Percentage of population (16 and over) with a bank account
100
90
80
70
60
50
South Africa
Brazil Colombia
40
30 Mexico
Philippines
20
Kenya
10
Tanzania
0
Source: WorldBank, 2009 16

17. Even the strength of the financial system and the availability
of credit shows no consistent pattern
Strength of legal rights index*
Index (0-10)
10 Kenya
South Africa
8
6
Colombia
Mexico
4
Philippines
Brazil
2
0
* The “Strength of legal rights index” measures the degree to which collateral and bankruptcy laws protect the rights of borrowers
and lenders and thus facilitate lending. The Legal Rights Index ranges from 0-10, with higher scores indicating that those laws are
17
better designed to expand access to credit; Source: WorldBank, 2010, “Doing business: Measuring business regulations”

18. The cost of transfers (outgoing) also varies significantly from
country to country
Cost of transfers
Percentage on a 250 USD international transfer (outgoing)
9%
8%
7%
6% Brazil
5%
Colombia
4%
South Africa
Mexico Kenya
3%
2%
1% Philippines
0%
Source: WorldBank, 2009 18

19. Also, the cost to the sender of an (incoming) international
transfer varies significantly
Cost of transfers by country of origin (2 largest
per country considered)
Percentage on a 200 USD remiitance
US to Brazil 15.7
Japan to Brazil 15.2
UK to South Africa 10.2
Canada to Mexico 9.5
UK to Kenya 8.5
US to Colombia 7.6
US to Mexico 7.4
US to Kenya 7.4
US to Philippines 6.6
US to South Africa 6.4
Spain to Colombia 5.4
Saudi Arabia to Philippinnes 4.9
Source: WorldBank, 2009; Western Union; MoneyGram 19

20. Also, international incoming transfers (remittances) show a
significant variance. They are relevant in the case of the
Philippines and less so in the case of Kenya
International incoming remittances
As a percentage of GDP
25%
 18 countries* have remittances
over $1 bn, representing more
20% than 10% of GDP
 Except for the Philippines, none
has built an m-money transfer
mechanism
15%
Philippines: $18,600, 11%
10%
Kenya
5%
Colombia
Mexico: $26,300, 2%
Tanzania South Africa Brazil
0%
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000
US million (2008)
* Bolivia, Kyrgyz, Senegal, Haiti, Albania, Moldova, Jamaica, Tajikistan, Nepal, Bosnia and Herzgovina, Honduras, Jordan,
El Salvador, Guatemala, Servia, Lebanon, Bangladesh 20
Source: WorldBank, 2009

21. Domestic remittances (measured by internal migration and
the size of rural population), also varies significantly within
our case studies
Rural population and migration
Growth of rural population (2000-2008)
10%
0%
- 10 20 30 40 50 60 70 Kenya80 90 100
Colombia
-10% Mexico South Africa
Philippines
-20%
-30% Brazil
-40%
-50%
-60%
Percentage of rural population (2008)
Source: WorldBank, 2009 21

22. Security (as measured by the homicides ratio) also does not
show a pattern
Homicides ratio
2008, per 100,000 inhabitants
50
45
Colombia
40
South Africa
35
30 Brazil
25
20
15
Mexico
10
Philippines
5 Kenya
0
* The “Strength of legal rights index” measures the degree to which collateral and bankruptcy laws protect the rights of borrowers
and lenders and thus facilitate lending. The Legal Rights Index ranges from 0-10, with higher scores indicating that those laws are
22
better designed to expand access to credit; Source: WorldBank, 2010, “Doing business: Measuring business regulations”

23. Our hypothesis:
None of the factors identified by the prevailing literature on the subject
explains the existence and success of a mobile money system
Except for very basic wireless penetration, none of these factors
appears to be essential nor enabling
As same companies have mixed experiencies (M-Pesa in Kenya ,
Tanzania and Afghanistan), it is not solely dependent on the mobile
company’s abilities nor on its entrepreneurial skills
A set of unique charactistics of each country, together with non-
hindering regulation, will probably determine success on a case by case
basis
As more companies launch these services, some of them will manage to
turn them into successes.
There appears to be no set of clear identifiable and replicable variables
that serve as a basis for success
23

24. 24