The Use of Client Assessment Scorecard in Buusaa Gonofaa MFI, EthiopiaEuropean Microfinance Week 2009November 24 – 26, 2009, LuxembourgPresentation by Teshome Y. Dayesso, General Manager, bgmfi@ethionet.et

Outline of Presentation

Expectations when developing BG’s ‘social ledger’ or poverty assessment tool

How it works and experience so far Collection of data from clients Training and role of Loan Officers Data capture, data analysis and reporting

Use of the information – how the scorecard guide SP management, better segmentation of clients and better adaptation of products, loan size determination

Interests and challenges – operational and cost implications; other issues triggered with the Award

Perspectives and the way forward

Why ‘Social’ Ledger or Scorecard?

To answer a 5 Million Dollar question asked by Board Members: “what is happening with achieving our social mission, not only financial sustainability?”. Whom do we reach? How poor are they? Is there a change (+ve, -ve) in our clients’ livelihood? Where do we succeed in changing client’s livelihood?

Where do we fail? Why? Who benefits from BG most? Does our loan assist

either survivalists or entrepreneurial poor? Or both?

Poverty Indicator

Measurable Indicators  

Year of Scoring

1 2 3 4 5  Date of scoring as Month/Year:   m1/yy m2/yy m3/yy m4/yy m5/yy  Household wealth          

Household Wealth

 

o      # Oxen 18 0 1 3 3 2o      # Cows 16 1 1 1 3 2o      # Sheep/goats 2 0 1 4 1 1o      # Bed type – Metal/Wood 2/4 2 2 2 2 2o      # Tape recorder 2 0 1 1 1 1o      # TV 24 1 1 1 1 1

Total Score of HH wealth:   100 127 183 213 188

Growth in Business

WC/Business Assets 30 30 60 100 150 160 Deduct: Score for debt/credit   0 -24 -36 -48 -48 Score: Net Business Assets   30 36 64 102 422

Total Score: HH & Bus. Wealth   130 163 247 315 610Progress  % Change of total wealth:     25% 52% 27% 94%

4

Poverty category & cut-off points

A person with total score of 15 is poorer than a person with score of 20, and vice versa

Poverty category

Score range

Approximate Income range

Very poor 0 – 29 ≤ $1/day

Poor 30 – 54 $1 – $2/day

Not so poor ≥54 ≥ $2/day

5

Experience so far – how it works

Collection of data from clients – the scorecard is part of routine loan application process;

Intake – a baseline data is gathered from all new clients upon entry to the program, at home with spouses (20 minutes)

Poverty scoring – LOs conduct assessment interview (scoring) on every new loan cycle (5 minutes), at group meeting place at end of current loan before taking the next loan; home visit of 5 clients per group (ave. group size = 15); random checking by branch manager, internal auditors,

Intensive training loan officer (LO) on the tool; but gaps in interviewing skills, mapping of house location

Data capture and analysis – data is entered on Access data base, report generated by Crystal Reports (C# Application)

Moving from SP Assessment to SP Management

It provided key information for decision to resolve tension between social and financial goals. BG’s average loan size is the lowest in Ethiopia, a

source of constant pressure from staff to increase loan size;

Board insists on small loan size to maintain focus on the poor & very poor as primary target group

The scorecard helps to guide SP management better segmentation of clients by poverty status,

gender, location (rural/urban), clients’ loan use pattern (IGA/MEs, agri/farming, consumption, housing improvement, etc.)

43%

39%

17%18%

25%

43%

9%

16%

32%

Very Poor Poor Not So PoorTotal Active Borrowers (n=9,318)Clients Dissatisfied with Very Small Loan SizeClient Suggesting Increase in Loan Size

Should we increase loan size? For whom?

Rural-urban segmentation was made possible by the scorecardLoan Cycle Rural Loan Urban Loan

1st $ 63 $ 832nd $ 125 $ 1253rd $ 146 $ 1674th $ 167 $ 2085th $ 188 $ 2506th $ 190 $ 2927th $ 250 $ 3338th $ 271 $ 3759th $ 292 $ 417 9

$73$88

$228

$77 $86$105

Very Poor Poor Not so poor

Client's Business Asset (US$) Loan Size ($US)

The loan size was fine-tuned to fit business size of clients’ segment

Benefits and Interests of the Scorecard System

It encourages accountability – it boosts MFI’s awareness of poverty mission, not just ad; social responsibility to clients as one important end akin to financial performance

It can be used as a benchmark to set SP goals, track progress over time – no of poor progressing to next level?

It helps to segment clients into poverty levels, business nature or type, etc… and offer tailored products

Incentives - eventually to set performance targets and compare poverty outreach among branches, staff, etc

BG’s winning of European Microfinance Award 2008 has helped BG to spearhead the agenda of client protection internally and in the Ethiopian MF sector.

1104/22/23

Challenges and the Way Forward

Locating rural clients’ home address for data auditing is a great challenge. Dispersion of rural HHs & poor infrastructure makes home visits very expensive.

Limited local capacity to develop the data base; the data base is rigid to generate various reports, thus limiting the advantages of existing data mining.

The ‘social ledger’ data processing is not integrated with the loan tracking system.

At least 3 rounds of scoring (~3 yrs) is needed to detect some pattern of change in clients’ livelihood.

1204/22/23

Poverty Indicator

Measurable Indicators  

Year of Scoring

1 2 3 4 5  Date of scoring as Month/Year:   m1/yy m2/yy m3/yy m4/yy m5/yy  Household wealth          

Household Wealth

 

o      # Oxen 18 0 1 3 3 2o      # Cows 16 1 1 1 3 2o      # Sheep/goats 2 0 1 4 1 1o      # Bed type – Metal/Wood 2/4 2 2 2 2 2o      # Tape recorder 2 0 1 1 1 1o      # TV 24 1 1 1 1 1

Total Score of HH wealth:   100 127 183 213 188

Growth in Business

WC/Business Assets 30 30 60 100 150 160 Deduct: Score for debt/credit   0 -24 -36 -48 -48 Score: Net Business Assets   30 36 64 102 422

Total Score: HH & Bus. Wealth   130 163 247 315 610Progress  % Change of total wealth:     25% 52% 27% 94%

13

Thank You!