unite for
children
The role of cash transfers in sub-Saharan Africa in
poverty alleviation:
Evidence from the Transfer Project
Richard de GrootUNICEF Office of Research—Innocenti
On Behalf of the Transfer Project Team
UNU-WIDER Conference
5-7 July 2017: Maputo, Mozambique
2
Source: Cirillo & Tebaldi 2016 (Social Protection in Africa: Inventory of Non-Contributory Programmes): www.ipc-
undp.org/pub/eng/Social_Protection_in_Africa.pdf
Rise of social protection in Africa:Non-contributory Gov’t programming triples over last 15 years
3
▪ Programs tend to be unconditional (or with ‘soft’ conditions)
▪ Targeting is based on poverty and vulnerability (OVC, labor-
constraints, elderly)
▪ Important community involvement in targeting process
▪ Payments tend to be manual (‘pulling’ beneficiaries to pay-
points)
▪ Opportunity to deliver complementary services
Key features of the African ‘Model’
4
Transfer Project: Partners & motivation
▪ Created 2009 as an Institutional Partnership between FAO,
UNICEF, Save the Children, University of North Carolina at
Chapel Hill
▪ Originally 6 countries in SSA, but expanded given high demand
▪Working in close collaboration with national counterparts,
including national governments and research institutions
▪ Objectives:
1. Provide evidence on the effectiveness of cash transfers in
achieving impacts for children and households
2. Inform the development and design of cash transfer policy and
programs
3. Promote learning across the continent on the design and
implementation of cash transfer evaluations and research
5
The Transfer Project & From Evidence to Action
Ethiopia, Ghana,
Kenya, Lesotho,
Malawi, Madagascar,
South Africa,
Tanzania, Zambia and
Zimbabwe
6
Total consumption pc
Food security scale (HFIAS)
Overall asset index
Relative poverty index
Incomes & Revenues index
Finance & Debt index
Material needs index (5-17)
Schooling index (11-17)
Anthropometric index (0-59m)
-.2 0 .2 .4 .6 .8Effect size in SDs of control group
36-month results at a glance
Broad Impacts from two Zambian programsMCP
CGP
Source: Handa et al. (2016). Can Unconditional Cash Transfers
Lead to Sustainable Poverty Reduction? Working Paper.
7
Reductions on poverty measures
10 10
1.4
12
11
10
00.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
Ethiopia(24 months)
Kenya(24 months)
Lesotho(24 months)
Malawi(36months)
Zambia MCTG(36 months)
Zambia CGP(48 months)
Zimbabwe(12 months)
Increase inhouseholdincome (%)
Reduction inpoverty headcount (pp)
Reduction in poverty gap (pp)
Solid bars represent significant impact, shaded insignificant.
Impacts are measured in percentage points, unless otherwise specified
8
Across-the-board impacts on food securityEthiopia
SCTP
Ghana
LEAP
Kenya
CT-OVC
Lesotho
CGP
Malawi
SCTP
Zambia
MCTG
Zambia
CGP
Zim
HSCT
Spending on food & quantities consumed
Pc food expenditures (overall) ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Pc 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
Source: Hjelm 2016 (The impact of cash transfers on food security): https://transfer.cpc.unc.edu/wp-
content/uploads/2015/09/The-Impact-of-Cash-Transfers-on-Food-Security.pdf
9
No evidence cash transfers increase
beneficiary spending on ‘undesirable’ goods
(e.g. alcohol & tobacco)
▪ Alcohol & tobacco represent <2% of budget share across 7
evaluations (data comes from detailed consumption modules
with over 120 food items)
▪ No positive impacts on alcohol & tobacco spending (consistent
with meta-analysis by Evans & Popova (2017) on cash
transfers & temptation goods).
▪ In Lesotho, impacts are negative (decreases in spending)
▪ Alternative measures in 4 evaluations yield same result:
▪ “Has alcohol consumption increased in this community over the
last year?”
▪“Is alcohol consumption a problem in your community?”
Source: Handa et al. 2017 (Myth-Busting? Confronting Six Common Perceptions about Unconditional Cash
Transfers as a Poverty Reduction Strategy in Africa): https://www.unicef-irc.org/publications/899/
10
School enrollment impacts (secondary age children):
Same range as those from CCTs in Latin America
8
3
78
15
89
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 - Solid bars represent significant impact, shaded not significant.
11
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) largehh
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.
Bars represent percentage point impacts - Solid bars represent significant impact, shaded not significant.
12
Resilience
©FAO/Ivan Grifi
13
Productive investmentsEthiopia
SCTP
Ghana
LEAP
Kenya
CT-OVC
Lesotho
CGP
Malawi
SCTP
Zambia
CGP
Zambia
MCTG
Zim
HSCT
Tropical Livestock Units (TLU) ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Any livestock ownership ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Any agricultural asset ownership ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Expenditure on crop inputs ✓ ✓ ✓ ✓
Value of harvest (local units) ✓ ✓ ✓ ✓ ✓ ✓
Green check marks represent significant impact, red adverse impact,
black are insignificant and empty is indicator not collected
▪ Overall impacts mask impacts on specific livestock/activities
▪ Households substitute out of casual paid labor (off farm) to on farm/small
businesses – no evidence that households systematically decrease work
participation
▪ Positive impacts on savings, networks, decreases in credit constraints vary by
country
Source: Handa et al. 2017 (Myth-Busting? Confronting Six Common Perceptions about Unconditional Cash
Transfers as a Poverty Reduction Strategy in Africa): https://www.unicef-irc.org/publications/899/
14
Community-level impacts
• In 5 evaluations, tested for inflation using a basket of 10
commonly purchased goods
• No inflationary impacts found (*exception price of beef in
Lesotho)
• Why? Supply expanded to meet demand, beneficiaries
relatively small proportion of population
• Local Economy Wide Impact Evaluation (LEWIE) models
collect information on non-beneficiaries and local
businesses/markets at baseline to simulate community-level
impacts of transfers
15
LEWIE estimates
Source: Taylor E, Thome K, Filipski M. “Local Economy-Wide Impact Evaluations of Social Cash Transfer
Programmes.” In “From Evidence to Action.” Eds. Davis B, Handa S, Hypher N, Winder Rossi N, Winters P,
Yablonski J. 2016. Oxford: Oxford University Press.
1.35
2.52 2.50
1.81
1.34
2.23
1.27
1.79 1.73
0
0.5
1
1.5
2
2.5
3
EthiopiaSCTPP(Abi-Adi)
EthiopiaSCTPP(Hintalo)
GhanaLEAP
KenyaCT-OVC(Garissa)
KenyaCT-OVC(Nyanza)
LesothoCGP
MalawiSCTP
ZambiaCGP
ZimbabweHSCT
16
Conclusions and what’s next?
▪ Large-scale government unconditional cash transfers have
strong, positive impacts on:
▪ Poverty, food security and expenditures
▪ Human capital
▪ Resiliency-related outcomes (assets, productive investment)
▪ Beyond beneficiaries: Local economies
▪ Design matters: Amount of transfer, regularity or payments
▪ Cash is important, but not sufficient: Supply side limitations
(health and education)
▪ Next frontier: “cash plus” programming and evaluation to
examine synergies
▪ Policy uptake important for understanding how to maximize
impact on poverty and human capital for poor populations
17
Tack
Asante
Zikomo
Grazie
Obrigado!
Ghana LEAP 1000
(© Michelle Mills)
18
Transfer Project is a multi-organizational initiative of the United Nations Children’s Fund (UNICEF) the
UN Food and Agriculture Organization (FAO), Save the Children-United Kingdom (SC-UK), and the
University of North Carolina at Chapel Hill (UNC-CH) in collaboration with national governments, and
other national and international researchers.
Current core funding for the Transfer Project comes from the Swedish International Development
Cooperation Agency (Sida) to UNICEF Office of Research, as well as from staff time provided by
UNICEF, FAO, SC-UK and UNC-CH. Evaluation design, implementations and analysis are all funded in
country by government and development partners. Top-up funds for extra survey rounds have been
provided by: 3IE - International Initiative for Impact Evaluation (Ghana, Malawi, Zimbabwe); DFID - UK
Department of International Development (Ghana, Lesotho, Ethiopia, Malawi, Kenya, Zambia,
Zimbabwe); EU - European Union (Lesotho, Malawi, Zimbabwe); Irish Aid (Malawi, Zambia); KfW
Development Bank (Malawi); NIH - The United States National Institute of Health (Kenya); Sida
(Zimbabwe); and the SDC - Swiss Development Cooperation (Zimbabwe); USAID – United States
Agency for International Development (Ghana, Malawi); US Department of Labor (Malawi, Zambia). The
body of research here has benefited from the intellectual input of a large number of individuals. For full
research teams by country, see: https://transfer.cpc.unc.edu/
Acknowledgements
19
• Transfer Project website: www.cpc.unc.edu/projects/transfer
• Briefs:
http://www.cpc.unc.edu/projects/transfer/publications/briefs
• Facebook: https://www.facebook.com/TransferProject
• Twitter: @TransferProjct
For more information
©FAO/Ivan Grifi
20
▪ A number of fledgling government programs and growing practice in SSA on cash transfers (2008)
▪ Some with plans for scaling up
▪ Most with models that were different from the well-known Latin American programs
▪ Little evidence from SSA
▪ A few programmes rolling out quantitative evaluations
▪ Others with evaluations but not rigorous methodology
▪ limited documentation and sharing on lessons, experience and impact evaluation
▪ Transfer Project: Responding to high demand for evidence to:
1) answer policy and program questions and
2) to influence and inform scale-up
In the beginning…
21
Overview of Transfer Project Evaluations
Country
(program)
Targeting
(in addition to
poverty)
Sample
size
(HH)
Methodology LEWIE YouthYears of data
collection
Ghana (LEAP)Elderly, disabled or
OVC1,614 Longitudinal PSM X 2010, 2012, 2016
Ghana (LEAP
1000)Pregnant women,
child<22,500 RDD 2015, 2017
Ethiopia (SCTP) Labour-constrained 3,351 Longitudinal PSM X 2012, 2013, 2014
Kenya (CT-OVC) OVC 1,913 RCT X X 2007, 2009, 2011
Lesotho (CGP) OVC 1,486 RCT X 2011, 2013
Malawi (SCTP) Labour-constrained 3,500 RCT X X 2011, 2013, 2015
South Africa (CSG) Child <18 2,964 Longitudinal PSM X 2010, 2011
Tanzania (PSSN) Food poor 801 RCT X 2015, 2017
Zambia (CGP) Child 0-5 2,519 RCTX 2010, 2012, 2013,
2014, 2017
Zambia (MCTG)Female, elderly,
disabled, OVC3,078 RCT X 2011, 2013, 2014
Zimbabwe (HSCT)Food poor, labour-
constrained3,063
Longitudinal
matched case-
controlX X 2013, 2014, 2017
22
Scaled up cash transfers are affordable in SSA
0%
5%
10%
15%
20%
Co
ng
o,
De
mo
cra
tic R
ep
ub
licZ
imb
ab
we
Bu
run
di
Lib
eri
aE
ritr
ea
Nig
er
Ma
law
iC
en
tra
l A
fric
an
Re
pu
blic
Ma
da
ga
sca
rM
ali
To
go
Gu
ine
aS
ou
th S
ud
an
Mo
za
mb
iqu
eG
uin
ea
-Bis
sa
uC
om
oro
sE
thio
pia
Sie
rra
Le
on
eB
urk
ina
Fa
so
Ug
an
da
Rw
an
da
Be
nin
Ta
nza
nia
, U
nite
d R
ep
ub
lic o
fZ
am
bia
Cô
te d
'Ivo
ire
Ke
nya
Th
e G
am
bia
Se
ne
ga
lM
au
rita
nia
Sa
o T
om
e a
nd
Pri
ncip
eL
eso
tho
Ca
me
roo
nC
ha
dS
ud
an
Djib
ou
tiN
ige
ria
Gh
an
aC
ap
e V
erd
eC
on
go
Bra
zza
ville
Sw
azila
nd
An
go
laN
am
ibia
So
uth
Afr
ica
Ma
uri
tiu
sB
ots
wa
na
Ga
bo
nS
eych
elle
sE
qu
ato
ria
l G
uin
ea
SO
CIA
L C
AS
H T
RA
NS
FE
R E
XP
EN
DIT
UR
E E
ST
IMA
TE
S
In % of general government total expenditure In % of GDP
Simulations show average cost 1.1% of GDP
or 4.4% of total Government spending
23
Age pyramids: Labor constrained
populations
0-56-1011-15
16-2021-25
26-3031-35
36-4041-45
46-5051-55
56-6061-65
66-7071-75
76-8081+
0-56-1011-15
16-2021-25
26-3031-35
36-4041-45
46-5051-55
56-6061-65
66-7071-75
76-8081+
15 10 5 0 5 10 15 15 10 5 0 5 10 15 15 10 5 0 5 10 15 15 10 5 0 5 10 15
15 10 5 0 5 10 15 15 10 5 0 5 10 15 15 10 5 0 5 10 15 15 10 5 0 5 10 15
Ethiopia SCTPP Ghana LEAP Kenya CT-OVC Lesotho CGP
Malawi SCT Zambia CGP Zambia MCTG Zimbabwe HSCT
males females
Population (%)
24
General approach to modeling
• Probit or ordinary least squares (OLS) multivariate
regressions
• Baseline balance/ successful randomization in all countries
• For “once occurring outcomes” use endline cross section and
drop those who had already reported outcome at baseline
• For outcomes changing over time (mental health, education,
aspirations), use difference-in-difference models
• Control for baseline individual, household, community
characteristics & cluster standard errors
• Weight for probability of appearing in sample (among all
eligible youth in any given household)
25
Responding to
the critics
with evidence