Pr

1
Microfinance Bureaus Balancing Vision and
Pragmatic Solutions
Regional Conference on Credit Reporting in
Africa Organized by the World Bank and the
IFC Break-out Session on Microfinance
Bureaus Cape Town, South Africa October 6,
2006 Mehdi Dutheil, Regional Director
2
AGENDA
THE CASE FOR CREDIT REPORTING IN THE MICROFINANCE
SECTOR INTEGRATED CREDIT BUREAUS LONG TERM
VISION AND PRESENT DAY REALITIES CASE STUDIES
INDUSTRY-LED VS. GOVERNMENT-LED MICROCREDIT
BUREAUS PERSPECTIVES
3
AGENDA
THE CASE FOR CREDIT REPORTING IN THE MICROFINANCE
SECTOR INTEGRATED CREDIT BUREAUS LONG TERM
VISION AND PRESENT DAY REALITIES CASE STUDIES
INDUSTRY-LED VS. GOVERNMENT-LED MICROCREDIT
BUREAUS PERSPECTIVES
4
ARE CREDIT BUREAUS REALLY USEFUL IN MICROFINANCE ?

• Banks make consumer loans using credit-bureau
  data for salaried borrowers with an automated,
  high-tech credit scoring approach
• ? Quantitative data are essential, hence credit
  bureaus are a priority
• On the contrary, microfinance enterprise loans
  are based on an individualized, labor-intensive
  high-touch approach to get data directly from
  the applicant and analyze the cash flows and
  personal character of the microentrepreneur
• self-employed poor cannot document income and
  credit history
• to compensate, MFIs send out credit officers to
  applicants homes
• ? Consequently, credit reporting and credit
  bureaus used to be deemed secondary in the
  microfinance sector

() Hans Dellien, WWB, and Mark Schreiner, MRM,
December 2005
5
BEFORE CREDIT BUREAUS THE BOLIVIAN CRISIS
IMPACT OF NOT HAVING CREDIT REPORTING THE
EXAMPLE OF BOLIVIA
Bad Debt Rate - Bolivia
In the late 1990s there was a crisis in the
Bolivian microfinance sector due to over
indebtedness of the clients The main cause for
the crisis was the lack of information sharing
tools.
COOPERATIVES
6
HOW IS CREDIT INFORMATION USED?
In credit underwriting
In pre-selection
In portfolio management

• To identify bad borrowers
• To price risk accordingly
• To use automated / semi-automated underwriting
  tools like credit scoring

• To identify deterioration of existing borrowers
• To avoid aggregation of bad debt among a number
  of financial institutions
• To collect the most risky debt first

• To sort out bad borrowers up-front
• To offer better conditions to good borrowers
• To reduce the cost of credit

7
BENEFITS OF CREDIT BUREAUS IN THE MICROFINANCE
SECTOR
AT THE CLIENTS LEVEL

• Increased efficiency in the evaluation of a loan
  can result in faster loan processing
• Clients with a good record can get preferential
  services and lower prices
• Clients are empowered to apply for credit in
  another location
• Default prone clients have the desire to obtain a
  good report and will hence be encouraged to pay
  their bad debts
• Lower risk of over-indebtedness by clients

8
BENEFITS OF CREDIT BUREAUS IN THE MICROFINANCE
SECTOR
AT THE INSTITUTIONAL LEVEL BETTER INFORMATION
SHARING AND DECISION MAKING

• More reliable decision making
• Minimizing risk because of a better visibility on
  borrowers past and ongoing default history and
  on their current outstanding balance of payments
  in different institutions
• Reducing transaction costs as it facilitates the
  analysis and quantification of credit risk
• Avoiding the aggregation of bad debt by borrowers
  among a number of financial institutions
• Increasing the number of loans granted as
  potential borrowers who were before excluded from
  the system because of the lack of information on
  their concern become beneficiaries

9
BENEFITS OF CREDIT BUREAUS IN THE MICROFINANCE
SECTOR
AT THE SECTOR LEVEL BETTER REGULATION AND
INCREASED COMPETITION

• Facilitates financial sectors supervision
  (Public Credit Registries in particular)
• Provides data for economic research and
  microfinance regulation improvement
• Facilitates the entry of new players in the
  market, such as banks willing to downscale into
  microfinance

10
BENEFITS OF SHARING NEGATIVE AND POSITIVE
INFORMATION
REDUCTION OF DEFAULT RISK AT THE INSTITUTIONAL
LEVEL
INCREASED ACCESS TO CREDIT AT THE CLIENTS LEVEL
Percent decrease in default rate
Percent of applicants who obtain a loan
12 decrease in default rate
90 increase in access
Negative information only
Negative positive information
Negative information only
Negative positive information
Simulated credit defaults assuming an acceptance
rate of 60
Simulated credit availability assuming a target
default rate of 3
Source Barron and Staten (2000)
11
BENEFITS OF CREDIT BUREAUS IN THE MICROFINANCE
SECTOR
CREDIT BUREAUS AND SUSTAINABLE POVERTY ALLEVIATION
IMPROVEMENT OF MFIs FINANCIAL VIABILITY
SETTING UP A CREDIT BUREAU
INITIAL FUNDING
MFIs SUBSCRIPTION TO CB
CB SELF- FINANCING
REDUCED POVERTY
REPAYMENT
SUSTAINABILITY OF THE CREDIT BUREAU
INCREASE IN THE TOTAL NUMBER OF MICROLENDING
BENEFICIARIES
INCREASE IN THE PROPORTION OF BENEFICIARIES WITH
REPAYMENT CAPACITY
Source Développement de la première Centrale des
Risques sur Internet pour les Institutions de MF
au Bénin, PlaNet Finance
12
AGENDA
THE CASE FOR CREDIT REPORTING IN THE MICROFINANCE
SECTOR INTEGRATED CREDIT BUREAUS LONG TERM
VISION AND PRESENT DAY REALITIES CASE STUDIES
INDUSTRY-LED VS. GOVERNMENT-LED MICROCREDIT
BUREAUS PERSPECTIVES
13
THE VISION OF AN INTEGRATED CREDIT BUREAU

• Full information sharing between financial
  (banks, consumer credit institutions, MFIs) and
  non financial institutions (retailers, telecom
  operators, utilities, etc.)
• prevents clients over-indebtedness
• fosters profitability at the financial system
  level
• In the long run, global and specialized private
  operators should be better positioned to ensure
  maintenance and evolution of the credit reporting
  system and bring technological and marketing
  innovations real time updates, mobile access,
  applicant scoring, payment default alerts, etc.
• Depending on countries, the supervision of the
  credit bureau can be taken care of either by the
  industry itself or by a public entity (Central
  Bank, Supervision Commission, etc.)

14
THE CASE FOR INTEGRATED CREDIT BUREAUS
THE IMPORTANCE OF SHARING INFORMATION ACROSS
SECTORS
Positive Negative
Types of Information
Negative Only
Sources ofInformation
Highpredictiveness (e.g. US, UK, Italy, South
Africa)
Full (information shared by banks, retailers,
NBFIs, mobile operators)
Lower predictiveness (e.g. Australia, Brazil)
Lowest predictiveness (e.g. Morocco, South Korea)
Lower predictiveness (e.g. Poland, Czech
Republic)
Fragmented (e.g. information shared among
banks only or retail only)
Source Microfinance and Credit Bureaus, Peer
Stein (IFC)
15
PRESENT DAY REALITIES OFTEN HAMPER INTEGRATED
CREDIT BUREAUS IMPLEMENTATION

• Many MFIs have very basic information systems
  (some are simply not computerized), which cannot
  compare with those of banks
• Microcredit bureaus business models are usually
  based on a big number of inquiries for small
  loans. Therefore, the amount charged /inquiry
  cannot be the same as the one charged for banks
• Because of short term loan cycles, MFIs need more
  frequent data updates and payment default after
  30 days, rather than 6 months
• A large part of MFIs staff being poorly
  educated, the ease of use of the Credit Bureaus
  application is more important than the number of
  functionalities
• The key for inquiries is different between MFIs
  (informal businesses identified by name and ID
  number) and banks (mostly formal businesses
  identified by corporate number)
• In a multi-sector credit bureau initiative,
  achieving a wide-ranging buy-in by all the
  players in a country is possible only if credit
  reporting has already reached sufficient maturity

16
AGENDA
THE CASE FOR CREDIT REPORTING IN THE MICROFINANCE
SECTOR INTEGRATED CREDIT BUREAUS LONG TERM
VISION AND PRESENT DAY REALITIES CASE STUDIES
INDUSTRY-LED VS. GOVERNMENT-LED MICROCREDIT
BUREAUS PERSPECTIVES
17
INDUSTRY-LED VS. GOVERNMENT-LED MICROCREDIT
BUREAUS

• The role of the National Regulator differs widely
  according to countries
• In many countries, Microcredit Bureaus
  initiatives stem from the industry and do not
  require investments in IT in the short run
• In Malis  Fleuve Niger  region, MFI executives
  hold an informal meeting and exchange their lists
  of delinquent clients. No software was developed
• In Haïti, MFIs have developed an Microcredit
  Bureau based on an Access database restricted to
  negative information
• In Mozambique, there is both
• a Credit Bureau based on SAP and supervised by
  the Central Bank, to which leading MFIs
  contribute monthly as well as banks, and
• an informal exchange of delinquent clients lists
  (Excel) between MFIs directors (more-up to-date)
• In Jordan, Tunisia and Egypt, legal difficulties
  impede the set up of banking / microfinance
  credit bureaus

18
Source World Bank 2002
19
CASE STUDY BENIN MICROCREDIT BUREAU
IMPLEMENTATION
CONTEXT OF MICROFINANCE IN BENIN

• 400,000 microfinance beneficiaries out of a total
  potential number of 4,000,000 beneficiaries
• High level of competition with around 400 MFIs
  competing on the same segments of the market ?
  risks over-indebtedness of clients and increased
  default on payments

20
A PIONEERING INITIATIVE
INTRODUCTION TO THE CREDIT BUREAU CREATION PROJECT

• The idea of a credit bureau was first launched by
  5 local MFIs including PADME, FINADEV and Vital
  Finance which realized in 2001 that their
  portfolio was deteriorating
• The objective was to consolidate portfolio
  quality by sharing information on payments
  overdue for more than 30 days via an internet
  system

21
(No Transcript)
22
MAIN AREAS OF ACTION IN THE PROJECT

• PROJECT ROADMAP
• Office set-up and equipment
• Design of the data base
• Establishment of procedures
• Training of credit bureaus administrator
• Creation of the website
• Training of MFI personnel
• Choice of legal statute and registration of the
  statute
• Maintenance of the website
• Awareness campaigns to convince other MFIs on the
  advantages of becoming a member of the credit
  bureau

• MAIN CHALLENGES ENCOUNTERED
• Technology challenge not all MFIs in Benin are
  equipped with the adequate technology (internet)
  ? necessity to provide relevant technology and
  diversify the information channels
• Data collection challenge it took time to make
  sure that the MFIs had a common vision on the
  information they need and the way to collect it

23
HOW THE BENIN CREDIT BUREAU WORKS
MFIs
MFIs

• 3 tools to access information
• Internet
• CD-ROM
• PDA (Personal Digital Assistant)

• Monthly filling and actualization of INFORMATION
  GRIDS listing the number of overdue payments of
  more than 30 days.

Files are sent to the CB
Transmission of the information required
CREDIT BUREAU

• Confidential information gathered on each
  borrower
• Personal information on each borrower
• Occupation
• Number of credits obtained
• Nature of outstanding credit

• Centralization of the information
• Consistency check of the information collected
• Information processing
• Information storage in the CB database

24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
EARLY TANGIBLE RESULTS

• LOAN OFFICERS
• Better knowledge of the applicants enabling a
  better decision making

An average of 150 inquiries per loan officer each
month

• MFIs
• Better quality of lending portfolio

• The number of applications rejected has
  significantly decreased
• Participating MFIs have reported better
  discipline amongst the clients, as they became
  aware that a bad credit history will deny them
  future access to credit

28
DECREASE IN REJECTED APPLICATIONS AFTER THE
IMPLEMENTATION OF THE CREDIT BUREAU
MFI Jan. Feb. Mar. Apr. Comments of MFIs
PADME 2 1 0.5 0.5  Beneficiaries are more disciplined 
FINADEV 6.8 3 1 0
CFAD 5 3 1 0
29
CONSEQUENCES OF THE SUCCESS OF THE PROJECT

• The World Bank has granted the budget required
  to extend the project to 17 MFIs in 2002
• Following the success of this experiment, it was
  decided to further develop the credit bureau in a
  3-year program, with the objective of bringing it
  to financial viability and increasing the number
  of members to 40
• Today, the ownership and responsibility of the
  project has been transferred to Consortium
  Alafia, the Benin Microfinance Association
• The Credit Bureau is now operating in 6
  provinces, through local agencies
• In July 2006, there were 20,000 clients in the
  database
• Discussions have been held with the BCEAO in
  order to transfer the Credit Bureau in the
  context of a regional Credit Bureau project

30
KEY LEARNINGS

• Microfinance credit bureaus are effective tools
  to prevent delinquency, even when the budget
  allows only for the sharing of negative
  information
• Very basic technologies can be sufficient in the
  short-to-mid-term
• Progressive buy-in of MFIs can be ensured by
• offering a highly professional service
• using technologies adapted to MFIs capacity
• proposing an adequate fee structure
• In case of management of the credit bureau by
  the professional association and moderate
  maintenance costs, the fee can be included into
  the membership fee to the association
• Ownership and leadership issues must be tackled
  at the start of the project

31
CASE STUDY THE MOROCCAN CREDIT BUREAU
IMPLEMENTATION
CONTEXT OF MICROFINANCE IN MOROCCO

• 12 MFIs in Morocco, with a microfinance market
  characterized by a high rate of repayment 99
  average
• This rate is deteriorating due to increased
  competition between MFIs covering the same areas
  and an increase in the number of cross-debts

32
INTRODUCTION TO THE CREDIT BUREAU CREATION PROJECT

• The Moroccan Credit Bureau, which began its
  operations in 2005, is mostly aimed at preventing
  crossed loans to clients whose loans are not
  delinquent yet. Hence the need for
• negative and positive information
• enabling access to the database for all the
  Moroccan MFIs (from the largest to the smallest)
• The project was supervised by a work group
  comprising the MFIs, their federation (FNAM) and
  PlaNet Finance Maroc. The Grameen Foundation USA
  and USAID also took part to the project design
• An estimated 1,000,000 yearly inquiries are
  needed to make the project viable

33
CREDIT BUREAU DEVELOPMENT INITIAL STEPS

• Development phase
• Analysis of each of Moroccos MFIs capacity to
  provide the Credit Bureau with the required data,
  and identification of the actions needed to
  develop this capacity (beg. 2004)
• Identification of information that can be
  exchanged
• Establishment of the conditions of contract
• Evaluation of the financial viability of the CB /
  business plan
• Call for tender for the choice of an information
  system provider of services
• Choice of the IS service provider in cooperation
  with the MFIs
• Parameter settings (beg. 2005)

34
CRITERIA USED IN THE CHOICE OF AN INFORMATION
SYSTEM SERVICE PROVIDER

• Choice of a provider with existing experience in
  the management of data
• Choice of a provider not using sub-contractors,
  which guarantees the continuity of the assistance
• Choice of a partner proposing an IS capable of
  evolution, in order to process one million loans
  in the near future
• Choice of simple tools that can be used by all
  MFIs even with basic IT equipment

35
HOW THE MOROCCAN CREDIT BUREAU WORKS ARCHITECTURE
Clients
Database
Credit Bureau Server
MySQL 4.1
Software - Application server
TOMCAT 4.1 - Application CB-ADMIN
- FTP SERVER
Software - Application server TOMCAT
4.1 - Application CB-CLIENT
- FTP SERVER
36
HOW THE MOROCCAN CREDIT BUREAU WORKS UPDATES

• Initially up-loading of all relevant
  information by all MFIs participating in the
  project
• Regularly
• Entry of all new borrowers and loans
• Changes in current borrowers and loans details
• Loan cancellation
• How
• From the MFI Head Office or authorized branch
  locations
• Interactive direct uploading of information
  from MFIs databases by the Credit Bureau
• Batch Preparation and sending of data by the
  MFIs for treatment by the Credit Bureau
• Follow-up of operations by delivery of
  notification with identification number

37
HOW THE MOROCCAN CREDIT BUREAU WORKS INQUIRIES

• The credit Bureau can be searched at all times
  from the Head Office or authorized MFIs branches
• The National Identification Number is the default
  search key
• Inquiries can be carried out
• On the web
• By batch after information is sent to the
  Credit Bureau by the MFIs, detailed reports are
  sent back
• By SMS
• Contents of Results Page
• Identification of the borrowers information
• Identification of loan information
• Various tools for visualization of results

38
HOW THE MOROCCAN CREDIT BUREAU WORKS
ADMINISTRATION

• Security
• Login and encrypted password
• Verification of contents
• Encrypting of exchanged information
• Server protected from external intrusion
• Level of Interaction
• The role of each user is clearly specified
  manager, administrator, updater, enquirer
• Archives of exchanged information
• Reporting on Credit Bureau usage frequency

39
Operational phase Once the project is fully
operational, PF Morocco will help
institutionalize the Credit Bureau and will share
the code with the selected CB manager
Central Bank
Options of management for the CB
Consortium of member MFIs (Economic Interest
Group)
National Federation of Microcredit Associations
Specialized private entity or third party
(Experian, etc.)
40
KEY LEARNINGS

• The institutional framework must be set up
  precisely even before setting up the technical
  framework
• The technical side of it is quite simple
• The buy-in from MFIs top management is essential
• The ease of use of the Credit Bureaus
  application is important, as a large part of the
  MFIs staff is not highly educated

41
AGENDA
THE CASE FOR CREDIT REPORTING IN THE MICROFINANCE
SECTOR INTEGRATED CREDIT BUREAUS LONG TERM
VISION AND PRESENT DAY REALITIES CASE STUDIES
INDUSTRY-LED VS. GOVERNMENT-LED MICROCREDIT
BUREAUS PERSPECTIVES
42
CREDIT SCORING, A POTENTIAL FOR MICROFINANCE
A quantitative method used to predict repayment
risk based on the performance of past loans with
characteristics similar to current loans. By use
of a scorecard, points are assigned to the
attributes of an applicant, and the sum of the
points is the score, with more points meaning
more risk.
DEFINITION
OBJECTIVES
Evaluate the risk from all potential customers
when applying for credit, through the forecast of
delinquent accounts or default of payment
43
BENEFITS OF CREDIT SCORING (1/2)
AT THE CLIENTS LEVEL A FAIR EVALUATION SYSTEM

• Clients are evaluated on non-subjective data
  through a well defined methodology
• Better pricing of loans
• Increased efficiency in evaluating loans can
  result in faster loan processing
• Default prone clients who wish to obtain a good
  report will have an incentive to pay their bad
  debts
• Lower risk of over-indebtedness by beneficiaries

44
BENEFITS OF CREDIT SCORING (2/2)
AT THE INSTITUTIONAL LEVEL

• More reliable decision making through better
  knowledge of the clients past behavior
• Better pricing of loans and provision against
  loan losses through the analysis of individual
  client risks
• Clear segmentation of population by score and
  delinquencies that helps design better strategies
  for delinquency prevention and for marketing
• Increase in the transferability of borrowers from
  one institution to another

45
CASE STUDY OF CREDIT SCORING MEXICO
THE CONTEXT

• Mexican MFI with more than 100,000 clients.
• 40 branches
• Assets over 100 millions USD

PRECONDITIONS FOR THE SUCCESS OF THE PROJECT

• Consolidated MIS
• Commitment of Top Management

46
CASE STUDY OF CREDIT SCORING MEXICO
SITUATION BEFORE SCORING

• A fragmented credit process
• Lack of standardization in decision making
• Authorization delays (up to 10 days)
• Impossible to measure ex - ante risk

47
CASE STUDY OF CREDIT SCORING MEXICO
The scorecard can identify ex-ante risk from
groups where the ratio of good to bad clients is
almost 35/1 to those high risk groups where the
ratio is 2/1
48
CASE STUDY OF CREDIT SCORING MEXICO

• A scorecard was implemented in the MIS of the MFI
• The MFI also have an Excel tool for testing the
  model

49
CASE STUDY OF CREDIT SCORING MEXICO
RESULTS OF THE CREDIT SCORING PROJECT

• Credit in 24 hours
• 80 of the applications with immediate results
• Reduction of 35 of credit cost
• Reduction of bad debt rate as analysts only focus
  on relevant applications (medium or high risk,
  leaving the rest to the score)
• Standardization of risk

50
CONCLUSION KEY SUCCESS FACTORS
KEY SUCCESS FACTORS FOR MICRO CREDIT BUREAUS

• Ensure that MFIs are ready for a credit bureau
  based on their IT systems and credit underwriting
  processes
• Work with experienced credit bureau operators and
  consultants
• Ensure that MFIs are given sufficient support and
  training to include credit reports and
  value-added services into their credit
  underwriting processes
• Collect both positive and negative information
  about borrowers in order to reduce information
  asymmetry
• Adjust credit bureau inquiry prices to the MFIs
  financial capacities

51
CONCLUSION KEY SUCCESS FACTORS
KEY SUCCESS FACTORS FOR MICRO CREDIT BUREAUS

• Ensure that the information is actively shared
  between all involved institutions
• To foster the Credit Bureaus success, a law
  requiring its use by all relevant players can be
  put in place
• To ensure coherency in policy and administration,
  a Credit Bureau should have one single overseeing
  body

52
CONCLUSION PLANET FINANCE CONTRIBUTIONS
Like in Benin or Morocco, the objective of PlaNet
Finance is to be technical and institutional
advisor to the Credit Bureaus project teams. Our
philosophy is to build sustainable credit bureaus
managed by local operators using open-source
technologies. Our credit bureau software has been
designed in order to be easily adapted. The
technologies used are widely known. PlaNet
Finance carefully selects the local technical
operator for the project development and
administration through a formal invitation to
tender followed by a transparent process of bid
selection. PlaNet Finance also ensures that MFIs
are given sufficient support and
training. Implementing a credit bureau is a long
process (over one year usually) but not
necessarily a very costly one. Most often, the
major issues are not the technology but the
institutionnal framework. Once this solved,
PlaNet Finance can lobby to help gather the
needed financial support from potential partners
and donors.
53
Thanks for your attention.
To know more, feel free to contact PlaNet
Finance mdutheil_at_planetfinance.org www.planetfi
nance.org