Sajjad Zohir, Discussion of Nationwide Surveys in India and Bangladesh on Client Movement Above the US$1.25 a Day Threshold and on Tools MFIs Are Using to Measure that Movement
G.R. Chintala, NABARD, Bangladesh, Partnerships that Build Bridges to New Fro...
Sajjad Zohir, Discussion of Nationwide Surveys in India and Bangladesh on Client Movement Above the US$1.25 a Day Threshold and on Tools MFIs Are Using to Measure that Movement
1. Client Movements across a threshold of $1.25 a day
and proxy inferences used to track such movements
Findings from a nation-wide survey in Bangladesh
Sajjad Zohir
Research Director, Economic Research Group
www.ergonline.org
Presented at the Global Microcredit Summit 2011
Valladolid, Spain
Sajjad/Valladolid/14-17 Nov '11 1
2. Background and Outline
• A study in 2009 trying to capture the number of MFI clients
moving above the threshold (poverty line)
- Monitoring on a regular basis; versus
- Stock-taking at an interval.
• The process with expert panel involving academia and
practitioners; and choice of ‘respectable’ methods to convince
the critics
- Lead academic/researchers in the field and representatives from major
MFIs
- External expert to develop poverty scorecards
- Local research agency to undertake the design, survey, analysis and report
writing
• Current presentation
- Outline general methods and leave the details for discussion
- Important findings from Bangladesh study
- Are we asking the right questions?
Sajjad/Valladolid/14-17 Nov '11 2
3. Brief on Methods - Concepts
• Choice of a threshold: US $ 1.25 PPP per person per
day
• Who do we measure? Clients are individuals, while
measures on poverty status refer to households. We
addressed ‘graduation’ at household level.
• But habitats are not stable. With mobility, how does
one capture the ‘population’? How to account for
urban-rural dynamics?
• What do we measure with? Proxy inference in the
guise of Poverty Scorecards (poverty likelihood
measures tagged to various score groups) developed
from HES/HIES unit level data and using
bootstrapping technique.
Sajjad/Valladolid/14-17 Nov '11 3
4. Brief on Methods - Operational
• If a member (or more) of a household had ever been a client of
an MCI during 1990-2008, that household was considered as
‘ever borrower’ household and was treated as an element of
the statistical population for the survey.
• Multi-stage sampling, methodically retraced, allowed one to
blow up sample results to population estimates
• Poverty score cards allow one to estimate the net number of
people crossing the threshold, not meant to track individual’s
progress.
• Three components of ever borrowers:
- Current residents in non-metropolitan areas. Scorecard-based findings were fine-
tuned with findings from life trajectory study
- Those who had migrated out of the non-metropolitan areas, but had borrowed from
MCIs during the period under study (1990-2008)
- MC clients (new) in the metropolitan areas were left out of current exercise
• Problems in finding the right scorecard – time and space?
Sajjad/Valladolid/14-17 Nov '11 4
6. Study Findings
• The survey finds two-third of the current non-metropolitan households to
be ever borrowers. Of the first time entrants, on an average, 62 % were
below the threshold defined by the $1.25 PPP. On the net, about 9.41% of
the ever-borrowers currently residing in the non-metropolitan areas were
found to have crossed the threshold.
• It is estimated that the number of people who had migrated out during the
1990-2008 period is equivalent to 4.73% of the current population in the
non-metropolitan areas. More than 55% of these households (current
residents in metropolitan areas) took microcredit before migration. Urban
surveys revealed that one-fourth of those households crossed the threshold.
• Movement above poverty is not unidirectional and poor oscillate between
below and above poverty. Almost 25% of those (62%) below the threshold
at the time of entry into microcredit programs had crossed the threshold,
while almost one-fifth of those (38%) above threshold slid below.
• Movements across threshold was influenced by: (i) time of entry - early
entrants were more successful, (ii) location - connectivity and proximity to
urban centers provided greater opportunities. Generally, extent of client
movements across poverty threshold had strong correlation with overall
macro economic environment.
Sajjad/Valladolid/14-17 Nov '11 6
7. Factors behind positive changes in life trajectory
• Increase in earning members;
• Increase in income generating assets (cows, van, rickshaws, boat);
• Good business (mostly fish cultivation);
• Good harvest/agriculture/ increased land cultivation;
• Increase in income (job/diversified/change/additional job taken);
• Lack of ‘shocks’ or events that involve a one-off expenditure;
• Migration to Dhaka;
• Dowry taken for male household members;
• Migration abroad;
• Separation of respondent from household;
• Help from in-laws (for male household members) and family;
• Government aid
Sajjad/Valladolid/14-17 Nov '11 7
8. Factors behind negative changes in life trajectory
• Treatment costs (illness, childbirth complications and then accidents);
• Natural disasters (flood /storm/ heavy rain/ /river erosion/drought);
• Wedding (including dowry) costs;
• Loss in business; and/or bad harvest;
• Separation of household (son leaving and establishing own household);
• Increase in dependant members;
• Difficulties with repayment;
• Litigation costs;
• Theft;
• Death of earning member; Lack of work;
• Inflation;
• Death of cow; Loss in fish cultivation/ due to flood/storm;
• Expenses to send son abroad; and fraudulence
Sajjad/Valladolid/14-17 Nov '11 8
9. After thoughts
• Poverty and microcredit: was it all about poverty
reduction? Is there a need to be defensive?
• Access to credit through innovation in service
delivery – why confine to impact assessment?
• Agency building and harnessing their potentials for
poverty reduction
• Changing politics of resource controls
- Renewed interest to avail the traditional routes for
trans-boundary flows
- Traditional banking and the new technology
• Strategic thinking for survival and beyond – did the
MFIs miss the boat?
Sajjad/Valladolid/14-17 Nov '11 9
10. Poverty Likelihood – individual vs aggregate
Entry Average poverty 2009 % of hhs, % of hhs, Net %
Cohort likelihood change in change in with
base year plh >= 0 plh > 0 improvements
(1) (2) (3) (4) (3)-[100-(4)]
1992 60.67 49.64 63.7 63.4 27.1
(11.03)
1996 73.27 53.86 71.5 71.5 43.0
(19.41)
2000 55.48 57.48 44.4 44.4 -11.2
(-2.0)
2005 58.93 58.12 69.2 46.6 15.8
(0.71)
Sajjad/Valladolid/14-17 Nov '11 10
11. Poverty Likelihood (%), variations across time & space
Entry cohort Entry year beg 2009 Entry year beg 2009
South/Southwest Central/North
1992 52.75 48.63 61.37 44.47
1996 67.23 50.05 75.27 53.74
2000 49.87 50.73 55.30 58.10
2005 51.22 50.12 61.16 58.70
East Northwest
1992 62.48 51.12 62.52 52.94
1996 80.74 56.48 70.81 55.09
2000 63.73 61.52 52.98 59.10
2005 66.18 64.29 57.77 59.21
Sajjad/Valladolid/14-17 Nov '11 11
12. Change in enrollment rates, 6-10 years age
Entry cohort entry year early 2009 entry year early 2009
South/Southwest Central/North
1992 87.50 86.21 71.95 80.77
1996 84.38 90.14 78.13 83.18
2000 77.72 88.18 78.48 87.76
2005 81.11 89.19 77.99 85.32
Division-level 81.04 88.76 77.35 85.48
East Northwest
1992 63.64 81.48 57.63 82.61
1996 68.29 88.51 59.57 80.31
2000 80.68 86.57 72.48 78.57
2005 73.12 80.46 65.52 69.95
Division-level 73.97 83.68 65.56 75.24
Sajjad/Valladolid/14-17 Nov '11 12
16. Indicator Value 91/2 95/6 2000 2005
1. How many household members were 20- A. Five or more 0 0 0 0
years-old or younger in <year>?
B. Four 0 0 0 6
C. Three 10 0 9 6
D. Two 10 9 14 12
E. One 21 17 21 18
F. None 29 27 34 31
2. What was the highest educational A. No class passed 0 0 0 0
attainment by any household members
B. Class 1-5, but cannot write letters 0 – – –
in <year>?
C. Class 1-5, but can write letters 5 – – –
D. Class 1 – 0 0 0
E. Class 2 – 0 0 0
F. Class 3 – 0 0 0
G. Class 4 – 0 0 0
H. Class 5 – 0 4 4
I. Class 6 7 8 6 4
J. Class 7 7 8 6 4
K. Class 8 7 8 6 4
L. Class 9 7 12 6 4
M. SSC but not BA 12 – – –
N. SSC/equivalent – 12 9 4
O. HSC/equivalent – 17 14 8
P. Bachelor’s degree or above 14 17 14 8
Sajjad/Valladolid/14-17 Nov '11 16
17. 3. What was the total operating A. No land; or less than 1.5 acres 0 0 0 0
land of the household in
<year>? B. 1.5 acres or more, but less than 2
(Total operating land = Cultivable 0 0 0 4
acres
agricultural land owned
C. 2 acres or more, but less than 4
+ Dwelling-house/homestead 8 4 6 4
acres
land owned
+ Cultivable agric. land rented in D. 4 acres or more, but less than 5
8 10 6 4
/share-cropped in/mortgaged acres
in
– Cultivable agric. land rented
out /share-cropped E. 5 acres or more 13 10 6 4
out/mortgaged out)
4. Did you own any cows in A. No 0 0 0 –
<year>?
B. Yes 3 5 4 –
5. Did you own any ducks in A. No 0 – – 0
<year>?
B. Yes 5 – – 2
Sajjad/Valladolid/14-17 Nov '11 17
18. 6. What type of latrine was used in the A. Open field or unknown 0 0 0 0
house in <year>?
B. Temporary 5 – – –
C. Pacca (latrine with brick and cement) 15 – – –
D. Hang latrine – 5 – –
E. Pit latrine without water seal – 5 – –
F. Latrine with septic tank – 10 – –
G. Kacha latrine (temporary) – – 4 3
H. Kacha latrine (permanent) – – 7 5
I. Pacca latrine (pit) – – 9 5
J. Pacca latrine (water seal) – – 9 5
K. Sanitary – – 9 5
7. What was the material of the walls A. Others 0 0 0 0
of the dwelling house of head of
B. Bamboo/hay/straw (or leaf, or hemp) 7 0 0 0
the household in <year>?
C. Mud/unburned brick (kada/kacha
8 3 0 0
brick)
D. C.I. sheet/brick/cement/timber (or
8 5 4 4
wood)
E. Brick/Cement 8 8 4 8
8. Did your household have an A. No – 0 0 0
electricity connection in <year>?
B. Yes – 10 4 4
Sajjad/Valladolid/14-17 Nov '11 18
19. 9. What was the material of the A. Others 0 0 – 0
roof of the dwelling house of B. Bamboo/hay/straw (or hemp) 0 0 – 0
head of the household in <year>?
C. C.I. sheet/tali/timber (or wood) 3 4 – –
D. Tile/ wood – – – 0
E. C.I. sheet/timber (or wood) 3 6 – 3
F. Cement 3 13 – 10
10. What was the source of A. Ponds/ river water 0 – – –
drinking water in <year>? B. Dug-wells/idara (draw-well) 0 – – –
C. Tube wells 0 – – –
D. Piped water 5 – – –
11. How many bedrooms did the A. One 0 – – 0
house have in <year>? B. Two 0 – – 2
C. Three 5 – – 3
D. Four 5 – – 5
E. Five or more 5 – – 9
Sajjad/Valladolid/14-17 Nov '11 19
20. 12. Did the household own a A. No – – 0 0
radio or a two-in-one cassette
player in <year>? B. Yes – – 7 5
13. Did the household own a TV A. No – – 0 0
in <year>? B. Yes – – 6 3
14. Did the household own a A. No – – 0 0
clock in <year>? B. Yes – – 5 2
15. Did the household own a A. No – – 0 0
wristwatch in <year>? B. Yes – – 3 3
16. Did any household member A. No – – 0 0
work for a daily wage in <year>? B. Yes – – 4 6
Poverty scorecards for National Microfinance Survey of Bangladesh
Source: Microfinance Risk Management, L.L.C., http://www.microfinance.com
Sajjad/Valladolid/14-17 Nov '11 20