The Story of Cash Transfers & Impact Evaluation in Sub-Saharan Africa, presented at 8th Annual SPIAC B Meeting

1. FROM EVIDENCE TO ACTION:
The Story of Cash Transfers & Impact Evaluation in Sub-Saharan Africa
Ms. Jenn Yablonski, UNICEF
Dr. Sudhanshu Handa, University of North
Carolina
On behalf of all the editors & authors
8th SPIAC-B Meeting
September 22, 2016

2. I. Introduction to the Transfer Project
II. From Evidence to Action:
Addressing myths, findings and
impacts
Outline

4. SDG 1.3: Implement nationally appropriate social protection systems &
measures for all, including floors, and by 2030 achieve substantial coverage of
the poor & the vulnerable.
Wide range of
social & economic
outcomes
Universal coverage & access
to
social protection

5. Key features of the African
‘Model’ of cash transfers
• Programs tend to be unconditional (or with ‘soft’
conditions)
• Targeting tends to be based on poverty & vulnerability
 OVC, labor-constrained, elderly
• Important community involvement in targeting process
• Payments tend to be manual, ‘pulling’ participants to
pay-points
 Opportunity to deliver complementary services

6. Transfer Project countries
Ethiopia, Ghana, Kenya, Lesotho, Malawi,
Madagascar, South Africa, Tanzania,
Zambia and Zimbabwe

7. Transfer Project approach
Stage I.
Design of Impact
Evaluation
Stage II.
Implementation &
Analysis
Stage III.
Use of Results &
Dissemination
• Focus on supporting impact evaluations of national
programmes & research-policy interface
• Close relationship between all national stakeholders
• Impact evaluations as part of broader evidence/learning
agendas & policy processes at national & regional level

8. Transfer Project approach
• Innovative research components:
 Mixed methods: quantitative, qualitative & local
economy impacts simulation
 Youth transitions to adulthood & HIV risk (UNICEF &
UNC)
 Productive impacts, local economy effects FAO (PtoP)

9. Country Quantitative Qualitative Lewie Other analysis
Ethiopia Non-experimental X X Targeting, payment process
Ghana Non-experimental X X Transfer payments
Kenya Experimental X X Operational effectiveness
Lesotho Experimental X X
Rapid appraisal, targeting, costing
& fiscal sustainability
Malawi
(incl.
Mchinji
pilot)
Experimental X
Not on
Mchinji pilot
Targeting, operational
effectiveness, transfer payments
South
Africa
Non-experimental X No Take up rate, targeting
Zambia
(CG &
MCTG)
Experimental
Not on
MCTG
Not on
MCTG
Impact comparisons across
programme, targeting
Methods used by the Transfer Project

10. Contribution to strengthening the evidence-
based case for promoting social protection
as a poverty reduction instrument
Impact of Transfer Project:
Global level
Generation of evidence on the broad range impacts of social cash
transfers
• Poverty impacts: child & household level
• Social impacts: education, access to health, nutrition-sensitive
indicators, food security
• Addressing economic & social determinants of HIV risk:
adolescent wellbeing
• Building the economic case: economic & productive impacts at
household level; Impacts to beneficiaries & to local economy

11. Social cash transfers can work in low-
income contexts, including SSA; can be
affordable; are a worthwhile investment
Impact of Transfer Project:
Regional level
• Strong evidence base on impact of cash transfers now available in
SSA
• Context-specific design and implementation (home grown models,
community participation, unconditional transfers, etc)
• Strengthen evidence base to feed to important regional processes
(AU commitments, etc)
• Contribution to changing the discourse: SP as an investment, not
a cost

12. Results from impact evaluations have
influenced design of programs and
contributed to strategic policy decisions
Impact of Transfer Project:
Country level
• Adjustment to program design & implementation (targeting,
transfer size)
• Moving from cash to cash+ (specifically in terms of nutrition,
agriculture & HIV/AIDS)- cash is important, but not sufficient
• Contribute to build & strengthen the case for scale-up &
expansion:
 Impact evaluations instrumental in strengthening reputation of
social cash transfer programs, & confidence with which
policymakers decide scale up
 Economic & productive impacts: addressing concerns regarding

13. • Ghana – Gov triples transfer size after
baseline simulations show level too low
• Kenya – Increase transfer size based on
4-year follow-up; Gov able to respond to criticisms w/ rigorous data
• Lesotho – Scale up after secondary analysis/ large impacts seen on the ground
• Malawi – Ghana & Zambia lessons on predictability ensured payments not
skipped
• Zimbabwe – Revised tageting system after positive comparison to more mature
programs
• Zambia – Massive scale up: Gov contribution jumped from $4m to $35m (2014)
Research informing policy:
Examples of scale-up

14. The highlights
Domain of impact Evidence
Food security
Alcohol & tobacco
Subjective well-being
Productive activity
Secondary school enrollment
Spending on school inputs (uniforms, shoes, clothes)
Health, reduced morbidity
Health, seeking care
Spending on health
Nutritional status
Increased fertility

15. Common criticisms or doubts
that we hear on the ground
• Cash will be wasted: Will be spent on alcohol and other
bads’
• Its just a ‘hand-out’, not used for productive activities,
cannot contribute to development
• Causes dependency, laziness
• Leads to inflation, disrupts local economy
• For child focused grants, increases fertility

16. Wasted? Across-the-board impacts on food security
Ethiopia
SCTP
Ghana
LEAP
Kenya
CT-
OVC
Lesotho
CGP
Malawi
SCTP
Zambia
MCTG
Zambia
CGP
ZIM
HSCT
Spending on food & quantities consumed
Per capita food expenditures        
Per capita expenditure, food items      
Kilocalories per capita  
Frequency & diversity of food consumption
Number of meals per day   
Dietary diversity/Nutrient rich food      
Food consumption behaviours
Coping strategies adults/children    
Food insecurity access scale   
Green check marks represent significant impact, black are
insignificant and empty is indicator not collected

17. Wasted? No evidence cash is ‘wasted’ on alcohol & tobacco
• Alcohol/tobacco represent 1% of budget share
• Across 7 countries, no positive impacts found on
alcohol/tobacco
• Data comes from detailed consumption modules covering
over 250 individual items
• In Lesotho negative impacts on alcohol consumption
(possible decrease through decrease in poverty-related
stress?)
• Alternative measurement approaches yield same result:
• “Has alcohol consumption increased in this community over the last
year?”
• “Is alcohol consumption a problem in your community?”

18. Claim: Its just a ‘hand-out’, not used for productive
activities, cannot contribute to development

19. School enrollment impacts (secondary age children): Same
range as those from CCTs in Latin America
8
3
7
8
15
8
9
12
6
9
6
10
0
2
4
6
8
10
12
14
16
18
20
Primary enrollment already high, impacts at secondary level. Ethiopia is all children age 6-16.
Bars represent percentage point impacts

20. Significant increase in share of households who spend on
school-age children’s uniforms, shoes and other clothing
11
26
30
23
32
11
5
0
5
10
15
20
25
30
35
Ghana (LEAP) Lesotho (CGP) Malawi (SCTP) Zambia (MCTG) Zambia (CGP) Zim (HSCT)
small hh
Zim (HSCT) large
hh
Solid bars represent significant impact, shaded not significant. Lesotho includes shoes and school uniforms
only, Ghana is schooling expenditures for ages 13-17. Other countries are shoes, change of clothes, blanket
ages 5-17.
Percentage point increase

21. Grade 3 math test – Serenje District, Zambia
More kids in school but school quality still a challenge

22. Productive impacts positive but vary by transfer size,
other operational features
Zambia Ethiopia Malawi ZIM Lesotho Kenya Ghana
Agricultural inputs +++ +++ +++ NS ++ NS +++
Agricultural tools +++ +++ +++ +++ NS NS NS
Agricultural
production
+++ +++ +++ NS ++ NS NS
Livestock
ownership
+++ +++ +++ +++ ++ Small
hhlds
NS
Non farm enterprise +++ NS +++ +++ NS +FHH NS
Stronger impact Mixed impact Less impact
NS=not significant
+++=positive, significant
---=negative, significant

23. Claim: Leads to laziness
We find reduction in casual wage labor, shift to on farm
and more productive activities
Zambia Kenya Malawi Ethiopia ZIM Lesotho Ghana
Agricultural/casual wage
labor
- - - - - - - - - --- --- -- NS
Family farm +++ +++ NS +++ NS NS +++
Non farm business
(NFE)
+++ +++ ++ NS NS NS NS
Non agricultural wage
labor
+++ NS +++ NS NS NS
Shift from casual wage labor to family business
consistently reported in qualitative fieldwork

24. Claim: Leads to laziness
“I used to be a slave to ganyu (labour) but now I’m a bit free.”
-elderly beneficiary, Malawi
0
.01.02.03.04
Density
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
age

25. Claim: Lead to inflation, disrupts local economy
• In six countries, tested for inflation in intervention versus
control communities using basket of goods
• No inflationary effects found
• Why not?
• Beneficiaries small share of community, typically 15-20 percent
• Poorest households, low purchasing power, don’t buy enough to affect
market prices
• Sufficient supply response to meet demand

26. In fact, cash transfers lead to positive multiplier effects in local
economy!!
0
0.5
1
1.5
2
2.5
3
Kenya
(Nyanza)
Ethiopia
(Abi_adi)
ZIM Zambia Kenya
(Garissa)
Lesotho Ghana Ethiopia
(Hintalo)
Multiplier: Amount generated in local economy by every $1
transferred

27. Claim: Cash transfers cannot contribute to development
Multiplier effects of cash transfers in Zambia & Malawi
Zambia (ZMK) Malawi (MK)
MCP CGP SCTP
Annual value of transfer (A) 720 720 26,169
Savings 33 41 381
Loan repayment 3 2 916
Consumption 1021 767 41,520
Livestock & productive assets 138 66 124
Non agricultural assets 163
Total spending (consumption + spending) (B) 1195 876 44,282
Estimated multiplier (B/A) 1.66 1.22 1.69
Impacts are based on econometric results and averaged across all follow-up surveys.
Estimates for productive tools and livestock derived by multiplying average increase
(numbers) by market price. Only statistically significant impacts are considered.

28. Claim: In child focused programs, increases fertility
Evidence suggests the opposite, if anything
• Zambia Child Grant Programme
 No impacts on total fertility or whether currently pregnant
 Some indication of improved birth outcomes (fewer pregnancy
complications)
 Kenya Cash Transfer for Orphans & Vulnerable Children
Reduction in early pregnancy among young women age 15-24 by
6 pp
No increase in number of children living in household
• South Africa Child Support Grant
 Reduction in early pregnancy by 11 percentage points

29. Where is evidence the weakest in terms of impact?
Young child health and morbidity
Regular impacts on morbidity, but less consistency on care seeking
Ghana
LEAP
Kenya
CT-OVC
Lesotho
CGP
Malawi
SCTP
Zambia
CGP
Zimbabwe
HSCT
Proportion of children who suffered
from an illness/Frequency of illnesses      
Preventive care   
Curative care    
Enrollment into the National Health
Insurance Scheme 
Vitamin A supplementation 
Supply of services typically much lower than for education sector.
More consistent impacts on health expenditure (increases)
Green check marks represent positive protective impacts, black are insignificant and red is risk factor
impact. Empty is indicator not collected

30. Where is evidence the weakest in terms of impact?
No impacts on young child nutritional status (anthropometry)
• Evidence based on Kenya CT-OVC, South Africa CSG, Zambia CGP,
Malawi SCTP, Zimbabwe HSCT
However, Zambia CGP 13pp increase in IYCF 6-24 months
• Some heterogeneous impacts
If mother has higher education (Zambia CGP and South Africa CSG) or if
protected water source in home (Zambia CGP)
• Possible explanations…
Determinants of nutrition complex - involve care, sanitation, water, disease
environment & food
Weak health infrastructure in deep rural areas
Few children 0-59 months in typical OVC or labor-constrained household

31. Meanwhile, emerging evidence that transfers enable safe-
transition of adolescents into adulthood:
Impacts on sexual debut among youth
36%
27%
17%
11%
44%
32%
28%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
Kenya (N=1,443) Malawi (N=1684) Zimbabwe (N=787) South Africa, girls (N =
440)
Treat Control
-6 pp impact**
-7 pp impact**
-13 pp
impact***
Kenya and Zimbabwe impacts driven by girls, Malawi driven by boys. Zambia no impacts.
-11 pp
impact***

32. How to make cash work better? Impacts depend on transfer size
‘Rule of thumb’ of 20 percent of consumption
0
5
10
15
20
25
30
35
40
Ghana
2010
Kenya
CT-OVC
(big)
Burkina TASAF
2012
Kenya
CT-OVC
RSA
CSG
Malawi
2014
Lesotho
CGP
(2010)
Ghana
2015
Kenya
CT-OVC
(small)
Zim
(HSCT)
Zambia
CGP
Zambia
MCP
Malawi
2007
Widespread impact
Selective impact
%orpercapitaconsumption

33. How to make cash work better? Transfers must be predictable and
regular!
Regular and predictable transfers facilitate planning,
consumption smoothing and investment
0
1
#ofpayments
Zambia CGP
0
1
2
3
4
5
6
#ofpayments
Ghana LEAP
Regular and predictable
Lumpy and irregular

34. From Evidence to Action:
Key messages
• Social cash transfers can be transformative for children, families &
communities
 Wide range of impacts across many social & economic domains — but
depends on implementation & other factors (cash transfers are not a
magic bullet)
• Impact evaluations have helped build credibility of social protection
sector in SSA
• Evidence debunks myths, e.g. cash transfers do not create dependency
• Transfer Project utilizes an innovative approach: the ‘how’ matters
• Evaluations & learning not only to ‘assess’ results, but to inform national policy &
progressively strengthen program design & implementation

35. THANK YOU!
Website: www.cpc.unc.edu/projects/transfer
Facebook: https://www.facebook.com/TransferProject
Twitter: @TransferProjct

36. 0
.01.02.03.04
Density
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
age 0
.01.02.03.04
Density
0 20 40 60 80 100
age
0
.01.02.03.04
Density
0 20 40 60 80
age at baseline
Zambia SCT (Monze Evaluation)
0
.02.04.06
Density
0 20 40 60 80 100
Age in Wave 1
Kenya CT-OVC
Malawi SCT Zimbabwe HSCT
Unique demographic structure of recipient households
in OVC and labor-constrained models (missing prime-ages)

37. How much do programs pay? Benefit structure and level in
selected programs (US$)
# members Ghana
LEAP
Malawi
SCT
MOZ
PSA
Zimbabwe
HSCT
Kenya
CT-OVC
Zambia
SCT
1 person 8 2.83 7 10 15 flat 12 flat
2 9.50 3.66 9 15
3 11 4.83 11 20
4+ 13.25 6.17 13.50 25
Beneficiary consumption
pp per day
0.62 0.34 0.50 0.85 0.70 0.30
ZAM: $24 if disabled member
MLW: 0.83 and 1.67 top-up per child in primary and secondary school
respectively