Policy Options to Improve Educational Outcomes and Address Lifecycle Vulnerabilities of Social Cash Transfer Programme Households in Malawi. Presented in August 2020.

1. Policy Options to Improve
Educational Outcomes and Address
Lifecycle Vulnerabilities of SCTP
Households in Malawi
Virtual Dissemination
August 20, 2020

2. Education Study: Research Questions
• What are underlying causal factors that affect primary and secondary
school enrolment, attendance and drop-out rates for children in SCTP
households?
• How does the currently provided SCTP school bonus relate to and impact
SCTP children’s schooling (enrolment, attendance, retention, progression)?
• What are school attendance tracking options suitable for the unconditional
Malawi SCTP?
• How can the SCTP be modified to increase the educational impact of the
SCTP on enrolment, attendance and retention of SCTP children, particularly
of adolescent girls?

3. Life Cycle Vulnerability Study: Research
Questions
• Does the SCTP design consider and respond to vulnerabilities of all
SCTP household categories and household types?
• How does the SCTP design address the continuously changing
vulnerabilities of different categories of individuals living in SCTP
households?
• What are the policy and programme options for the SCTP to better
respond to the vulnerabilities of all SCTP household categories and
individuals?

4. Study Design: Mixed Methods Approach
• Qualitative approach
• Education: Focus Group Discussions with in- and out-of-school youth; In-Depth
Interviews (IDIs) with caregivers of school-age youth
• Life Cycle Vulnerability: IDIs with three vulnerable groups identified from previous
impact evaluation data and confirmed with participants at inception meeting
• Elderly caregivers with health issues; adolescents; pre-school children whose caregivers are
not recipients
• Quantitative approach
• Survey of ~1500 beneficiary households
• Detailed analysis of schooling outcomes and barriers
• Quantify outcomes of specific vulnerable groups

5. Study plan finalized at Inception Workshop
May 2019, Lilongwe

6. October 2018 –
March 2019
May 2019 June – August
2019
September
2019
September
2019 - now
Project
conception &
research
questions &
team
formation
Inception
workshop to
finalize study
objectives and
implementation
plan
Data
collection
and
analysis
Study reference
group workshop in
Salima to discuss
findings and
recommendations;
discuss additional
analysis
Review and
finalization of
report and policy
briefs
Research timeline

7. Mulanje (Q2)
Nkanda
Mthiramanja
Salima (Q2)
Ndindi
Maganga
Nkhata Bay
Fukamapiri
Mankhambira
(Quantitative only)
Study Sites

8. Happy Researchers
Maxton
Kaku
Courtney

9. Findings from the Education Study

10. School availability and quality are not favourable

11. School availability and quality are not favourable

12. Enrolment rates are high, regular attendance is moderate and so
effective enrolment is quite low (2018/2019 school year)

13. Regular Attendance does not differ by school term or sex
2016-2017 2017-2018 2018-2019
Age group Male Female Total Male Female Total Male Female Total
6-13 years 1st term 0.55 0.57 0.56 0.58 0.59 0.59 0.60 0.60 0.60
2nd term 0.55 0.58 0.56 0.57 0.56 0.57 0.59 0.59 0.59
3rd term 0.56 0.57 0.57 0.58 0.57 0.57 0.60 0.59 0.59
Academic year 0.51 0.54 0.53 0.54 0.53 0.53 0.55 0.55 0.55
14-17 years 1st term 0.58 0.57 0.58 0.57 0.61 0.59 0.60 0.65 0.62
2nd term 0.59 0.58 0.58 0.55 0.60 0.58 0.59 0.63 0.61
3rd term 0.58 0.59 0.58 0.58 0.61 0.59 0.58 0.64 0.61
Academic year 0.55 0.54 0.54 0.52 0.56 0.54 0.54 0.60 0.57
18-23 years 1st term 0.55 0.55 0.55 0.60 0.56 0.59 0.64 0.60 0.62
2nd term 0.55 0.53 0.55 0.58 0.59 0.58 0.62 0.60 0.61
3rd term 0.57 0.56 0.56 0.60 0.57 0.59 0.63 0.57 0.61
Academic year 0.51 0.53 0.51 0.55 0.53 0.54 0.60 0.55 0.58

14. Effective enrolment appears to be increasing for the 6-13 year age
group but flat for the 18-23 year age group
2016-2017 2017-2018 2018-2019
Age group Male Female Total Male Female Total Male Female Total
6 - 13 years Rate 0.45 0.47 0.46 0.50 0.49 0.49 0.55 0.54 0.54
N 1,361 1,369 2,730 1,245 1,274 2,519 1,128 1,172 2,300
14 - 17 years Rate 0.46 0.45 0.46 0.45 0.49 0.47 0.48 0.52 0.50
N 599 535 1,134 669 606 1,275 714 681 1,395
18 - 23 years Rate 0.34 0.26 0.31 0.33 0.26 0.30 0.32 0.24 0.28
N 305 234 539 455 370 825 599 483 1,082
Total Rate 0.44 0.44 0.44 0.45 0.45 0.45 0.47 0.47 0.47
N 2,265 2,138 4,403 2,369 2,250 4,619 2,441 2,336 4,777

15. Majority of students are too old for their grade
– resulting from late entry and repetitions

16. SCTP education bonus is small relative to
the estimated cost of schooling (9% and
12% for primary and secondary school
respectively)
Primary, 106,627
Secondary, 152,125
Primary, 9,600
Secondary, 18,000
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
Primary Secondary
Projected Education Cost Per Child SCTP Education Bonus Per Child
Average SCTP education bonus is less
than what households currently spend
on education
27960
27606
27400
27500
27600
27700
27800
27900
28000
Average expnediture on schooling Average education bonus

17. SCTP Amount affects Enrolment but not other
Schooling Outcomes

18. SCTP Amount and Education Expenditure (1)

19. SCTP Amount and Education Expenditure (2)

20. Non-monetary barriers to education
 Supply side
 Proximity to school increases both enrolment and progression rates
 School quality increases both enrolment and progression
 Violence from teachers affects males; violence from peers affects females
 Inadequate hygiene facilities for females during menstruation
 Sexual harassment from some school authorities
 Household and individual level factors
 Low material well being leads to lower participation in school
 Demand to spend more time on chores reduces effective enrolment

21. Non-monetary barriers to education

22. Qualitative Results: Financial barriers
 Mtukula Pakhomo helps pay for essential school supplies such as fees, clothes, shoes, soap for
bathing, and food. Some of these are more important for females than males
“You use [shoes] wisely. After classes when you are back home, you remove [them] and keep [them] so that it should last long.” -in-school
male , Mulanje
“We encourage [our caregivers] to get us soap for our clothes so that we can go to school”.
“Having no soap for clothes, I miss[ed] school”
 Proper clothing and hygiene more important for girls than boys
“We are different from boys, who are ok to go to school with torn clothes as compared to girls” –in-school
female, Mulanje
 Hunger is a barrier to learning
“You go to school without taking breakfast and coming from school, you find no lunch. You go
hungry, you come back [and] you stay hungry, so you conclude that, ‘I better quit school. Maybe
I can do some casual labor and buy food for myself’.” – in-school male, Mulanje

23. Qualitative Results: Socio-Cultural barriers
 Lack of parental/guardian support
“Some parents see that their daughter, one day, will be pregnant like the [others] in the
community, [so] they prefer not to support a girl knowing that they will lose money for
nothing, in terms of school fees.”
 Lack of role models in the society
“When you go to school, they say ‘Look at her, what is that? You think you will finish school?
You are no different from us, you will find us here’”
 Peer pressure
“…We want to enjoy the time we have to be in class [but] if we hear that ‘So and so sells
water at [the] trading centre’ then we want to be like him. ‘I should also go and do the
same’ [I say to myself]. When realizing [leaving school is a] disaster, we are already in it. You
start blaming one another, [saying], ‘You are the one who made me to do this!’ So, the
problem is that we entice others to follow you if you drop out from school”

24. Qualitative Results: Barriers within schools
 Over-punishment and mocking by teachers
“Instead of giving you punishment you deserve, they give you a punishment that you will fail to do it.“
‘mbuli iwe’ [unintelligent you are] ‘amagokutailira kwanu’ [they do not caution you at home]
 Sexual abuse
“[Girls] drop out from school because they go at school without any motive. They meet
with a teacher and [the teacher] ask[s] love to them, [girls] deny. [Girls] prefer to stay at
home. Asking them; ‘Why not going to school?’, they say ‘My teacher proposed me. So I
better stay away from him’ “ - In-school CG, Mulanje
 Teachers not motivated, low quality education
“Teachers are not teaching as they deserve. I think because they are not well paid. They just tell their
learners to say ‘teach one another’ and they go out and have a chat outside the classroom. We have
low quality education.” –Out-of-school CG, Mulanje

25. Proposed policy options
Policy options education bonus size and coverage
 Increase the overall value of the transfer and school bonus
 Early enrolment bonus
 Automatic secondary school fee waivers for SCTP children
 Direct incentives for school progression
 Incentives for girls’ education

26. Proposed policy options
Policy options on education bonus system design
 Labelled child bonus in lieu of educational bonus
 Condition the school bonus on minimum school attendance
 Strengthen linkages with Ministry of Education
 Linkages for young adult dropouts

27. Questions/Clarifications/Comments

28. Findings from the Lifecycle
Study

29. Unique demographic profile of SCTP households
More youth and elderly, fewer children under 5 compared to rural
ultra-poor
Few children under age 5
More youth
More elderly

30. 3 important subsets of beneficiary household
members were identified as potentially requiring
further support
 Adolescents and young people age 15-24 years, and their
caregivers;
 Elderly household heads with special needs or a chronic health
condition
 Children under age five years and their young caregivers

31. Heads of household in SCTP households are older, more likely to be
female, less likely to have ever attended school and more likely to
have a chronic condition compared to IHS4 ultra poor households
57.12
70
61
38
44.77
32
82
16
0
10
20
30
40
50
60
70
80
90
Age (years) Female Ever attended school Has a chronic condition
Study sample IHS4 comparison sample

32. Chronic conditions: SCTP population versus
rural ultra-poor by gender
Implication: Youth have more responsibilities, which affects their schooling outcomes

33. There is a penalty on household monthly expenditure in the
presence of one or more of these vulnerable groups
(not all SCTP households are the same)
11000
10264
11054
8379 8470
9398
0
2000
4000
6000
8000
10000
12000
Young mother Elderly with chronic condition/disability One or both
None One or more

34. What is the probability for a child/adolescent to do more than 28
hrs of Household chores in a week? (By Gender of the Child)

35. What is the probability for a child/adolescent to do Ganyu work?
(By Gender of the Child)

36. Elderly and Disabled People

37. Vulnerabilities
More money spent on medical
expenses
“The money we make after selling
the firewood, it keeps me going…Its
enough for me to buy pain killer. “
Limited work capacity
“If you are sickly you cannot
develop… You cannot do
anything if you are sickly.”
SCTP as the only source of income
“Oh no I do not source money on my own. if I have
the only channel that I make or receive money is
through this social cash transfer program.”
Lack of support
“I can say I am stressed up to say ‘I am alone
now; I am the one remaining. What should I do
then’ I am much affected. Elders are dying and I
am alone remaining now. I have no relations. I
am alone now without anyone to hold my
hand”

38. Vulnerabilities
Sole Caregivers
“What I can complain about them
it’s when they are sick. I go to
hospital with them alone. No one
else helps me.”
“It has not been easy to raise them. I
think of them how to make them live
and survive. This is the reason why I
developed BP.”
Unique Worries as Caregivers
“I think of them how I will support
them if they move to other levels in
their studies. If I die, what will they
do?”

39. Potential Solutions
Within Scope of SCTP
• Formalize the way the transfer is passed on if main
beneficiary dies
“If anything is when a beneficiary die. If you go and
report, then the household stops receiving the
transfers...People fight for the money when the real
beneficiary dies”.
• Ensuring that bonus is unconditional
“I do receive MWK 14 400. I was receiving MWK 16 000
but they have deducted my transfers…The school going
children did not attended school for a month. Then I
realize the payment has been deducted.”
Beyond SCTP
• Implementation of Pension Program
• Linkage and referral to supplementary
programs

40. Children age 0-5

41. 30% of children age 0-5 are not the child of the
head--many SCTP households have ‘sub-families’
that are not officially recognized
Age 20 years and below
(median age at first birth
15 years)

42. Vulnerabilities
Effects of low education/dropping out
“I do regret most of the time…Education
is good, there is a lot of programs that
comes in our communities that requires
someone who knows how to write.”
“There were a lot of problems, getting
married at a young age and having
more children at young age, that’s due
to dropping out.”
“As my parents had died a long time,
and I was left at a very young age, so
due to poverty and having no place to
stay, I found a man to marry me and we
indeed got married and that’s for my
first child.”
Barriers to re-entry
When I look at our household, I feel that
my parents can’t manage to send me
back to school because I would also be
imposing on them another responsibility
when the household is already lacking.
Who will the children find at home after
school, as I am their only parent
I am old now, imagine I dropped out
school in standard 2, so now is
impossible to go back to school

43. Vulnerabilities
Lack of hope
Instead of going to school or doing
something that can bring money
in your life you just stay at home,
all your dreams dies, so I wasn’t
happy.
Social Isolation
The challenge I face, people
laughs at me, they say that I gave
birth when I was too young […] I
just stay at home; I don’t even
answer them.
People and even friends talk about
you most of the time [...] they can
call you names like ‘you are a
prostitute’

44. SCTP Response
Receiving part of the transfer
“I have benefited a lot, as I get what I
lacked like soap and lotion…[my
mother] gives everyone the money to
buy for themselves”
Insufficient amount with addition of
child
“When we started receiving social cash
transfer money, there have been
improvements but not entirely, the
money is not enough according to the
number of people in our family.”

45. Potential Solutions
Within Scope of SCTP
• Better explanation of incentive
“if I can find someone to support me
and pay my school fees I can go back
to school.”
• Nutrition bonus for child 0-5
• Separate transfer for young
mothers (sub-family)
“When they receive, they give me at
least K2000 to go and buy what I
want.”
Beyond SCTP
• Young Mothers Groups
• Adult School
• Intervention from groups like
YONECO

46. • MNSSP II calls for a social protection system that responds to lifecycle
vulnerabilities
• SCTP is a narrowly targeted program that only explicitly addresses one
recognized vulnerability in MNSSP II (labor-constraints)
• Elderly, chronically ill, disabled and youth are indirectly reached; children 0-5
are under-served
• Will the SCTP remain a small program with a narrow target group, or evolve
into the flagship social protection pillar for Malawi, addressing lifecycle
vulnerability directly – this is the ‘big’ decision
• As countries develop their social protection systems, they typically move from narrow
programs to broad-based programs that reach a wider constituency
Strategic decision facing GoM regarding
future of SCTP and relationship to MNSSP II

47. What if SCTP slowly transformed into a broad-based categorical
program? This figure shows the number of ultra-poor rural
households reached under alternative categorical targeting
approach

48. Can Malawi afford a categorical programme?
Annual cost of programme under alternative targeting approach
0
10000
20000
30000
40000
50000
60000
Current SCTP 65+ 0-2 years 65+ and 0-2
years
0-5 years 65+ and 0-5
years
Annual Transfer Costs (MK millions)
Old transfer level New transfer level

49. Policy options
 Directly target lifecycle vulnerability – the ‘big’ decision
 Support specific vulnerabilities within the current targeting
approach via linkages and complementary services
 Nutrition bonus for children age 5 years and under
 Support to mother-child sub-families within SCTP
households

50. Team:
Ashu Handa
Frank Otchere
Kaku Attah Damoah
Maxton Tsoka
@TransferProjct
transfer.cpc.unc.edu
ZIKOMO.

51. Education Study Qualitative Design
Salima Mulanje
In-school Males (2 groups) Males (2 groups)
Females (2 groups) Females (2 groups)
Out-of school Males (2 groups) Males (2 groups)
Females (2 groups) Females (2 groups)
Caregivers In-school (1 group) In-school (1 group)
Out-of-school (1 group) Out-of-school (1 group)
Total FGDs 10 10
155 individual participants

52. Life Cycle Vulnerability Study: Qualitative
Design
Salima Mulanje Total
Elderly heads with chronic conditions or disability 5 IDIs 5 IDIs 10 IDIs
Caregivers of adolescent/youth 5 IDIs 5 IDIs 10 IDIs
Caregivers of young children 5 IDIs 5 IDIs 10 IDIs
Adolescents/young adults (combined with Education) 8 FGDs 8 FGDs 16 FGDs

53. Quantitative Design
• A total of 1520 beneficiary households were sampled for both Education
and Lifecycle studies.
• Given the objectives of the two studies, we identified beneficiary
households with children of school going-age as a common denominator
for the two studies.
• After identifying households that meet the common denominator we
computed the proportional distribution of these households by Traditional
Authority and Village Cluster in each District.
• For Mulanje and Nkhata Bay the sampled HHs were drawn from the MIS
administrative data
• For Salima the sampled were drawn from the Impact Evaluation dataset.

54. Distribution of Sampled Households and Final Data
DISTRICT TA Number of Sampled
HH
Number of HH in Final
Data
Response Rate
(%)
Mulanje Mthiramanja 196 194 98.98
Nkanda 317 316 99.68
Nkhata Bay Fukamapiri 235 227 96.60
Mankhambira 271 270 99.63
Salima Maganga 222 221 99.55
Ndindi 279 277 99.28
Total 1,520 1,505 99.01

55. 100
200
300
400
500
600
700
Index
Trend in Total Enrolment from 1980 – 2017
Index (1980=100)
Introduction of free
Primary School
Source: Authors Elaboration based on data from World Development Indicators.
The Big Push in 1994/1995 and the Introduction of MSCTP

56. .4.6.8
1
6 7 8 9 10 11 12 13 14 15 16 17
Age
Baseline Midline
Endline
Control
.4.6.8
1
6 7 8 9 10 11 12 13 14 15 16 17
Age
Baseline Midline
Endline
Treatment
School Enrolment Impacts (6-17 year olds)
Impacts at:
-Midline: 11.4 pp***
-Endline: 8.9 pp**
No difference between boys and girls

57. Enrolment by Age-group, Gender, and Academic year
2016-2017 2017-2018 2018-2019
Age group Male Female Total Male Female Total Male Female Total
6 - 13 years Rate 0.86 0.87 0.86 0.90 0.90 0.90 0.94 0.94 0.94
N 1,384 1,384 2,768 1,281 1,302 2,583 1,187 1,222 2,409
14 - 17 years Rate 0.84 0.83 0.83 0.85 0.86 0.85 0.87 0.87 0.87
N 607 542 1,149 676 611 1,287 721 685 1,406
18 - 23 years Rate 0.66 0.49 0.59 0.60 0.49 0.55 0.54 0.42 0.48
N 306 240 546 459 378 837 603 493 1,096
Total Rate 0.83 0.82 0.82 0.83 0.82 0.83 0.82 0.81 0.82
N 2,297 2,166 4,463 2,416 2,291 4,707 2,511 2,400 4,911

58. Effective Enrolment by Age-group, Gender, and Academic year
2016-2017 2017-2018 2018-2019
Age group Male Female Total Male Female Total Male Female Total
6 - 13 years Rate 0.45 0.47 0.46 0.50 0.49 0.49 0.55 0.54 0.54
N 1,361 1,369 2,730 1,245 1,274 2,519 1,128 1,172 2,300
14 - 17 years Rate 0.46 0.45 0.46 0.45 0.49 0.47 0.48 0.52 0.50
N 599 535 1,134 669 606 1,275 714 681 1,395
18 - 23 years Rate 0.34 0.26 0.31 0.33 0.26 0.30 0.32 0.24 0.28
N 305 234 539 455 370 825 599 483 1,082
Total Rate 0.44 0.44 0.44 0.45 0.45 0.45 0.47 0.47 0.47
N 2,265 2,138 4,403 2,369 2,250 4,619 2,441 2,336 4,777