A blurred picture, lagged effects or hidden realities?

1. Keetie Roelen
IDS Members seminar
4 November 2015
Poor Children living in Rich Households:
A Blurred Picture, Lagged Effects or
Hidden Realities?

2. Rationale
early investment matters

3. Rationale
mismatch of monetary and multidimensional poverty

4. Rationale
Vietnam Congo

5. This study: aims and objectives
Investigate mismatch and overlap patterns
Repeated cross-sections
Longitudinal pattern
Explain mismatch patterns
Use mixed methods approach
Research in Ethiopia and Vietnam, complemented by
research from Burundi
exploring mismatch patterns in child poverty outcomes
- Measurement error
- Lagged effect
- Opportunity costs
- Household factors
- Infrastructure and services
- Awareness and attitudes
- Aspirations

6. Outline seminar
Data and methods
• Quantitative data
and measures
• Qualitative data
and methods Explaining mismatch:
• Measurement error
• Lagged effects
• Opportunity costs
• Household factors
• Infrastructure and services
• Social protection
• Coaching and support
• Aspirations
• Attitudes and behaviour
Policy implications:
• Measurement
• Targeting
• Programme support
• Whose responsibility?
Child poverty outcomes
• Respondent perspectives
• Cross-sectional findings
• Longitudinal findings

7. Data and methods
quantitative data and measures
Multidimensional poverty measure
Ethiopia Vietnam
Indicators Domains Indicators
not attending school (6-18) Education net enrolment (5-15)
working on farm (6-18) primary completion rate (12-15)
working on domestic chores
(age 6-18)
Health
visit to health facility (2-4)
Shelter living in house with electricity
(0-15)
living in proper house (0-15)
Water and
Sanitation
living in dwelling with improved
toilet (0-15)
drinking from improved water
source (0-15)
Child work child work (6-15)
Social
inclusion
having caregiver unable to work
(0-15)
quantitative 1999 2004 2009 panel
Ethiopia (ERHS) 5054 3709 4937 1497
2004 2006 2008 panel
Vietnam (VHLSS) 12154 10696 9960 1068
monetary poverty measure
Ethiopia Vietnam
real per capita consumption real per capita expenditures

8. Data and methods
qualitative data and methods
qualitative adults children Total
Ethiopia 88 61 159
Vietnam 145 78 223

9. Data and methods
qualitative data and methods
adults children Total
91 40 131

10. Child poverty outcomes
respondent perspectives
Are household wealth and child wellbeing always the same?
“They are related and always the same, because rich households usually
feed their children well, purchase clothes frequently, send their
children to school and provide health care more than the poor
families.” [14 year old boy, Harresaw, Ethiopia]
“No. If the family is rich but they do not live in peace and do not take
care of their children, their health, study and entertainment, child
wellbeing is not good.” [15 year old girl, Dong Thap, Vietnam]
“Parents in poor families care about and educate their children more
encouraging them to aim at higher especially at school so that they
have a better life. Parents in wealthy families do not have time to
educate their children.” [Caregiver, Cibitoke, Burundi]

11. Child poverty outcomes
cross-sectional trends

12. Child poverty outcomes
longitudinal patterns
2006
N AB A B C Total
2004 AB 567 49.6 14.5 26.1 9.9 100
A 393 3.6 30.5 2.8 63.1 100
B 407 13.8 8.9 39.3 38.1 100
C 1126 0.1 5.9 3.5 90.6 100
2008
N AB A B C Total
2006 AB 352 53.4 10.8 26.7 9.1 100
A 304 13.8 41.5 2.6 42.1 100
B 358 18.4 7.0 35.5 39.1 100
C 1479 1.4 5.9 4.3 88.4 100
Transition matrices poverty groups Vietnam, 2004-08

13. Explaining mismatch
measurement error
Community criteria for household wealth and child
wellbeing in Vietnam
Identification of households in
Ethiopia

14. Explaining mismatch
lagged effects
2006
N AB A B C Total
2004 AB 567 49.6 14.5 26.1 9.9 100
A 393 3.6 30.5 2.8 63.1 100
B 407 13.8 8.9 39.3 38.1 100
C 1126 0.1 5.9 3.5 90.6 100
2008
N AB A B C Total
2006 AB 352 53.4 10.8 26.7 9.1 100
A 304 13.8 41.5 2.6 42.1 100
B 358 18.4 7.0 35.5 39.1 100
C 1479 1.4 5.9 4.3 88.4 100
Transition matrices poverty groups Vietnam, 2004-08

15. Explaining mismatch
opportunity costs
Livestock ownership and family work across consumption deciles for children aged 10-15
in rural Ethiopia
810121416
averagehoursfamilywork(perweek)
23456
0 2 4 6 8 10
real per capita consumption (deciles)
livestock (TLU) hours family work (per week)
>> Where is the tipping point?
“Parents from poor households give more responsibilities to their
children from a young age, which help them to grow up with a
better education compared to their peers from rich families.”
[Male caregiver, Cibitoke, Burundi]

16. Explaining mismatch
household factors
Ethiopia Vietnam
quantitative analysis
household size: AB↑, B↑, A↓ single hh head: AB↑, B↑, A↑
no education hh head: AB↑, A↑ no education hh head: AB↑
primary education+ hh head: A↑ unemployed hh head: AB↑, A↑
living in Mekong River Delta: AB↓,
B↓, A↑
qualitative analysis
female hh head: B↑ attitudes and awareness: B↑, A↑
attitudes and awareness: B↑, A↑

17. Explaining mismatch
infrastructure and services

18. Explaining mismatch
social protection

19. Explaining mismatch
coaching and support
Concern Worldwide ‘Terintambwe’ programme in Burundi

20. Explaining mismatch
aspirations

21. Explaining mismatch
aspirations and attitudes
Sara, 16 years old, living with her father, in grade 8
“I can say my wellbeing is good and bad. It is good because
I am in school. My wellbeing is bad because I am working at
home when I return from school.”
Her father says: “I don’t send my children to work for other
households but I believe children should work at home in
household production.”
“If I pass the national examination, I want to continue my
education in the town of Atsbi. But my father wants me to join the Dera high
school in order to support him. I want to be an engineer in order construct road
to my community in particular and my country in general.”

22. Policy implications: measurement
Child poverty measurement requires a comprehensive use of
monetary and multidimensional approaches.
Child poverty measurement requires its own measure
Longitudinal analysis is important for gaining insight into
movements in and out of child poverty.
The use of both quantitative and qualitative data allows for
more in-depth analysis of child poverty and its causes and
solutions.

23. Policy implications: targeting
Targeting remains heavily focused on establishing
approximations of monetary poor.
How to expand targeting practice to:
>> include all vulnerable children?
>> move from a household to individual perspective?

24. Policy implications: programme interventions
Livelihood strengthening and income generation can work for
children but:
>> it is not enough, and
>> it may have adverse consequences.
Need for more critical reflection on role of children in asset
accumulation strategies and income generating activities.
How to effectively incorporate behaviour change elements?

25. Policy implications: whose responsibility?
>> How to avoid that a focus on knowledge, awareness and
individual behaviour does not become an unconstructive
and unfair blame game?
Source: Kuenstler 2015

26. Thank you!
k.roelen@ids.ac.uk
@KeetieRoelen