cgap technology program - agents marketreach
DESCRIPTION
Siedek, H. (2010). CGAP Technology Program - Agents marketreach. Available: http://www.cgap.org/gm/document-1.9.2115/agents_marketreach.pdf. Last accessed 27th Nov 2010.TRANSCRIPT
1DRAFT – for discussion
1
© CGAP, 2007
CGAP Technology ProgramCGAP Technology Program
Banking AgentsBanking AgentsPart IIIPart III –– Market ReachMarket Reach
Hannah Siedek ([email protected])
2DRAFT – for discussion
2
© CGAP, 2007
Banking agents, or banking correspondents, are a key component of the diverse technology approaches being studied by the CGAP Technology Program.
This document provides an overview of banking agents -- retail shops, pharmacies, air time dealers, etc. – that deliver financial services in countries like Brazil, Peru, Colombia, and others.
It explains CGAP’s current knowledge about banking agents. We would be happy to include your analysis or experience with banking agent models in this document and provide all appropriate references.
Please send all your comments, questions, and ideas to Hannah Siedekat [email protected]. We look forward to hearing from you!
CGAP research on banking agents
3DRAFT – for discussion
3
© CGAP, 2007
Part I: The approach
CGAP research on banking agents
Part II: Implementation choices
Part III: Adoption/Market reach
Part IV: Business Model - forthcoming
Part V: Social Impact - forthcoming
Part VI: Regulation - forthcoming
To learn more about CGAP’s Technology Program and view the other parts of this research, please visit http://www.cgap.org/technology.
4DRAFT – for discussion
4
© CGAP, 2007
Present findings of CGAP’s research in Brazil to understand:
If clients are using banking agents for their transactions.
What kinds of transactions they are conducting at the banking agent.
Why and why not people are using banking agents.
Purpose of this presentation
5DRAFT – for discussion
5
© CGAP, 2007
Sample:Twelve municipalities in the state of Pernambuco (poverty levels similar to northeastern Brazil) Municipalities with critical number of banking agents in 50% rural and 50% urban locations750 users and non-users randomly selected from households with average per-capita income equal to or less than half the minimum wage (i.e. less than $175)Marginal error: 3.6%
Questionnaire:45 minutesQuestions:
Sociodemographic profileFinancial products/services usedPoints of sale (POS) usedPerceptions regarding points of sale
CGAP research on customers in Brazil (June 2006)
6DRAFT – for discussion
6
© CGAP, 2007
Summary findings
I. People accept banking agents
II. Agents reach low-income clients
III. Limited range of transactions conducted at banking agents
7DRAFT – for discussion
7
© CGAP, 2007
I. People accept banking agents
Banking agents are popular: almost everybody (90%) uses banking agents, including illiterate clients.
Uncertain profile of a typical “user” but skews towards:Urban population (70%)Up to 50 years old
Low-income CLIENTS chose banking agents since they are clos to their home, have short lines, and offer good service.
NON-USERS who do not use agents do not know about them, or do not think they need them.
8DRAFT – for discussion
8
© CGAP, 2007
Banking agents are popular!
Banked51%
Unbanked49%
POS user83%
Non-POS user17%
POS user91%
Non-POS user9%
POS user90%
Non-POS user10%
Banked47%
Unbanked53%
Banked56%
Unbanked44%
Who uses banking agents?Nine in ten households (90%)
No correlation between use of agent and having a bank account.*
* This might be a result very specific to Brazil since everybody receiving benefits has to have a bank account. Therefore, large parts of the population are banked, but often only to receive their payments.
9DRAFT – for discussion
9
© CGAP, 2007
61
41
28
11
6
5
0 10 20 30 40 50 60 70
Lottery house
Bank branch
Supermarkets,shops
Payments bank
Post office
ATM
%
Six out of 10 people claim to use lottery house as their primary source for some transaction
Four out of ten people claim to use a bank branch as their primary source for some transaction
(n=752)
What are people’s favorite channel?
Lottery houses have largest coverage and are people’s favorite point of sale. However, 40% prefer banking at the branch!
10DRAFT – for discussion
10
© CGAP, 2007
What do clients like about different channels?
1914
1110
74
32
291
0 10 20 30 40
Only facility near my houseGood serviceConvenience
SecurityNo lines
TrustNear my house
Because I feel comfortableOther
No answer
Lottery houses Branches
Independent shops21
1815
1311
913
0 5 10 15 20 25
Good serviceNear my house
No linesConvenience
Only facility near my houseFast
Other
3120
1110
74
21
14
0 10 20 30 40
Near my houseNo lines
Good serviceOnly facility
FastConvenience
TrustSecurity
Other
Postal bank29
1211
722
135
1
0 10 20 30 40
Only facility near my houseNear my house
No linesGood serviceConvenience
FastTrustOther
No answer
% %
% %
Banking agents are often closer to poor peoples’ homes, provide good service, and clients to do not have queue.
11DRAFT – for discussion
11
© CGAP, 2007
Why don´t they use some channels?
4527
1911
43
0 10 20 30 40 50
Do not know itDo not need service
None near my houseDo not trust it
Long linesOther
No answer
Lottery houses Branches
Independent shops Postal bank48
158
3211
166
0 10 20 30 40 50 60
Do not need serviceNone near my house
Long linesDo not know it
ExpensivePoor service
Difficult to use the machineOther
No answer
3926
75
41
162
0 10 20 30 40 50
Do not need serviceNone near my house
Long linesPoor service
Do not know itDo not trust it
OtherNo answer
4126
17222
18
1
0 10 20 30 40 50
Do not need serviceDo not know it
None near my houseLong lines
Poor serviceDo not trust it
ExpensiveOther
No answer
Low-income clients do not use banking agents because they think they do not need them, or because they do not know about them.
12DRAFT – for discussion
12
© CGAP, 2007
50
9
4
19
14
0 10 20 30 40 50 60
Bank branch
Supermarkets,shops
Cooperative
Other
None
Poor clients trust banking agents and some prefer them to branches. Fifty percent of agent users claimed to have used a bank branch as a primarily transaction source, before using the agent. Only 6% of banking agent users still use the branch as primary point of sale.
Where did you conduct transactions before using the banking agent?
%
Agents might substitute the branch?
13DRAFT – for discussion
13
© CGAP, 2007
Who are the 10% not using agents?
Banked and not using agents:Half is older than 40 years.Either with income below US$100/month (23%) or above US$701/month (22%)56% is banked
Unbanked and not using agents:77% is either below 24 or over 50 years old. 92% claim to conduct payments mostly at branchesAs reasons for not using other point of sale, they state that they do not know about the service, or that they do not need it.
(n=91)
Non-POS user10%
Banked56%
Unbanked44%
(n=44)
14DRAFT – for discussion
14
© CGAP, 2007
II. Banking agents reach poor clients64
42
3
5
1
5
7
60
40
15
12
7
2
1
55
47
23
21
11
0
11
0 10 20 30 40 50 60 70
Lottery house
Bank branch
Payments bank
Supermarkets/shops
ATM
Post office
Postal Bank
%
SEC E(n=288)
SEC D(n=419)
SEC C(n=44*)
Poorer segment E clients are more likely to have a “simplified” account (34%) to receive their welfare payments which they access at lottery houses.
32% of segment C has credit cards and 11% use ATMs.
Around 40% of all segments still use the branch.
Agents can reach low income clients. In previous years, Brazilian banks have primarily targeted client segments A-C*. By providing poorer clients with bank accounts, agents play an important role in serving these groups. * The Brazilian economic classification ranges from A to D and is based on household assets, income, and education level of the head of household
15DRAFT – for discussion
15
© CGAP, 2007
Poor clients use agents to receive their benefits
61
6
6
4
4
1
15
1
0 10 20 30 40 50 60 70
Caixa EconomicaFederal
Banco Popular doBrasil
Banco Matriz
Banco Bradesco
Multibank
Lemon Bank
Other
No answer
%
Caixa Economica Federal is ten times more likely to be used most often than any other POS.
62% of Caixa users, receive benefits.
Poor clients seem to choose the type of agent based on convenience.Lottery houses of Caixa Economica have the largest coverage, i.e., in all municipalities, and clients have to go anyway to receive their benefits.
(n=661)
16DRAFT – for discussion
16
© CGAP, 2007
What about illiterate clients?
Financial access:Majority is banked (62%) and over 50 years old (77%)More likely to have a “simplified account”*, i.e. receive government benefits (55% vs 27% of total sample).
Use of agents:82% claim to use banking agentsAlmost same profile with regard to what outlet (e.g., lottery, retail, etc.) is used, with slightly more usage of independent shops, rather than lottery houses.
(n=115)
* A “simplified account” is a government subsidized account for clients with less than US$400 income/month who do not own another bank account. This account is free of charge and offers a limited number of free transactions per month.
17DRAFT – for discussion
17
© CGAP, 2007
III. Range of transactions is limited
Banking agents are almost exclusively (99%) used for bill payments
No transaction conducted significantly more than once a month
Little use of financial service in general: The majority (87%) of low-income clients has a bank account to receive their benefits, only 6% to save money
Less than one in ten has a credit card13% have an active loan2% are part of an insurance schemeOnly 4% send or receive remittances (international and national)
18DRAFT – for discussion
18
© CGAP, 2007
Clients only conduct payments
99
3223 18 17 13
60
20
40
60
80
100
Payments Receive socialbenefits
Withdrawals Receive funeral,retirement,
pension
Receive wages Get loan orfinancing
Deposit insavings acc
%
%
Uptake of a wide range of financial services at the agents by poor clients is still a challenge. Almost half the population is banked, but often only to receive their salary and other benefits.
19DRAFT – for discussion
19
© CGAP, 2007
..and almost only once per month
99
3223 18 17 13
60
20
40
60
80
100
Payments Receive socialbenefits
Withdrawals Receive funeral,retirement,
pension
Receive wages Get loan orfinancing
Deposit insavings acc
%
No transaction happens significantly more than once a month.
Average frequencyper month
1.03 0.93 0.88 0.96 0.95 0.48 0.4
20DRAFT – for discussion
20
© CGAP, 2007
Areas for future learning…How to cross-sell financial services to low-income clients through the agent? Can the agent take over this role, or does the bank have to make the effort?
What are demand-driven products and services which are easy to be offered by the agent, and rapidly adopted by clients?
What other marketing activity (e.g., media, direct sales, financial literacy training) can strengthen the relationship of the client with the bank?
What is the appropriate regulatory environment to support client uptake (e.g., no taxation for transaction below a certain amount, less stringent account opening requirements, etc.)
21DRAFT – for discussion
21
© CGAP, 2007
Besides in-depth research in Brazil, CGAP will use its learning ventures to learn more about how poor clients respond to banking agents:
Research projects under way
Colombia
Kenya
Maldives
Mongolia
Pakistan
The Philippines
22DRAFT – for discussion
22
© CGAP, 2007
23DRAFT – for discussion
23
© CGAP, 2007
© 2007 Consultative Group to Assist the Poor/The World Bank1818 H Street, N.W.Washington, DC 20433Internet : www.cgap.orgEmail: [email protected]: +1.202.473.9594
All rights reserved.The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The Consultative Group to Assist the Poor encourages dissemination of its work and will normally grant permission promptly.
24DRAFT – for discussion
24
© CGAP, 2007
In Brazil, 74 financial institutions (57 banks and 17 financial companies) managed 90,000 banking agents* in all 5,561 municipalities of the country.
Access is not the problem!
Government delivers benefits through lottery houses of state bank Caixa Economica. Every recipient of benefits, but also salaries, has to have a bank account, often a “simplified account.”**
Having a bank account does not imply full use financial services.
Annex: Specifics that impacted our research
Source: Kumar, Anjali. Expanding Bank Outreach through Retail Partnerships, CGAP analysis
* The 90,000 number includes banking agents in shops, but also special attendance points in public and private companies, as well as mini-branch-like outlets. When only counting the banking agents in retail and postal outlets of the largest financial institutions, the number of points shrinks to around 58,000.
** A “simplified account” is a government subsidized account for clients with less than US$400 income/month who do not own another bank account. This account is free of charge and offers a limited number of free transactions per month. This special account can be offered by any financial institutions, but has primarily beenoffered by public banks, since it is often not profitable.
25DRAFT – for discussion
25
© CGAP, 2007
Annex: Details of sample
Sociodemographic: 71% female, 29% male60% with household income less than US$400/month95% of poverty levels D and E*
Education:50% of elementary school or less 13% illiterates
C5%
D59%
E36%
Economic classification:
** The Brazilian economic classification ranges from A to D and is based on household assets, income, and education level of the head of household
Access to finance: 51% have a bank account. 99% transact not more than once a month. Urban populations need 16min, rural populations 34min to the next point of sale.