What are the impacts of cash transfer schemes?

1
What are the impacts of cash transfer schemes?

• COUNTRIES
• Ethiopia
• Lesotho
• Malawi
• Zambia
• South Africa

• IMPACTS
• Maternal / child health development
• Vulnerable populations (OVC, PLWHA, elderly)
  
• Nutrition
• Food security
• Economic empowerment
• Asset accumulation

2
Ethiopia

• John Hoddinott
• IFPRI
• j.hoddinott_at_cgiar.org

3
Impact of Ethiopias Productive Safety Nets
Programme

• Against the background of rising food prices and
  widespread drought, participation in the public
  works component of the PSNP between 2006 and 2008
  had modest effects. It
• Improves food security by 0.40 months
• Increased growth in livestock holdings by 0.28
  Tropical Livestock Units (TLU).
• Improved the resilience of households as measured
  by their ability to raise funds in an emergency
  and
• Relative to non-beneficiaries, beneficiary
  households perceive that their welfare has
  improved.
• These aggregate results generated by the use of
  matching methods - mask important differences
  within the sample of beneficiary households.

4
Impact of the PSNP, contd

• Program impacts on asset accumulation are greater
  when higher levels of transfers are received and
  when participants have access to both the PSNP
  and agricultural productivity packages such as
  improved seeds.
• On average, households receiving high levels of
  transfers had a 14.3 percent higher growth rate
  in the value of their livestock holdings.
• Access only to public works transfers, or only to
  specific agricultural packages of services such
  as irrigation or seeds, had no effect on
  agricultural productivity as measured by wheat or
  maize yields. By contrast, households where both
  public works transfers and these services were
  received saw large increasesover 200 kg/hain
  wheat and maize yields.

5
Impact of the PSNP, contd

• Payment for at least 10 days work per month in
  the three months prior to the 2008 survey
  improved food security, even though food prices
  were rising dramatically at the time. These
  households experienced a 19.2-percent higher
  growth rate in caloric acquisition and higher
  growth in livestock holdings.
• The PSNP does indeed act as a safety net.
  Restricting the treatment group to households
  that received at least 10 days work per month in
  the three months prior to the 2008 survey and
  reported being affected by drought, program
  participation increases the growth rate of
  caloric acquisition by 30 percent.

6
Impact of the PSNP, contd

• Receipt of public works transfers under the PSNP
  crowds out private transfers, but by a very small
  amount
• There is no meaningful evidence that
  participation in Public Works employment has a
  disincentive effect on adult male or female labor
  employed in nonfarm own-business activities, wage
  employment, or work on the family farm.
• When payments are near the level intended in the
  program design, school attendance rates increase
  substantially for boys (19 percentage points),
  and for girls age 11-16 years (15 percentage
  points).
• PSNP public works transfers plus agricultural
  productivity packages increase child labor by
  girls by (4.5 hours per week).
• There was no evidence that receipt of PSNP
  transfers improves child anthropometric status.

7
Lesotho

• David Croome
• dr.croome_at_nul.ls

8
LESOTHO PENSION CHILD HEALTH - 1

• Health of babies relatively good but significant
  growth problems from malnutrition as children get
  old
• Pension seen as household income, allocated by
  pensioner
• 20 used for health, mainly access to local
  doctors and clinics.
• Use for childrens health not specified
• Pension cash used for transport to clinics and
  for medicines

9
LESOTHO PENSION CHILD HEALTH - 2

• When pension started, clinics required user fees,
  especially in church-owned facilities (50 of all
  provision)
• 2009, clinic fees abolished, but still cost for
  transport etc. More use being made of facilities

10
LESOTHO PENSION CHILD HEALTH - 3

• Children benefiting from pension cash providing
  more food in the household.
• General situation Effect of pension probably
  quite small. Need more active community health
  service to take health care to the children in
  school (Free primary education 90 kids in
  school)

11
LESOTHO PENSION VULNERABLE GROUPS - 1

• Orphans and Vulnerable children (OVCs)
• AIDS related vast increase in OVC number
• 200,000 10 of population
• 60 of pensioners claim to be left with 1 or more
  OVCs to look after. ( n.b, OVC dumping, pensioner
  has a little regular cash and cannot run away)
• Real problems of caring. Old people unaware of
  modern needs for care. Esp. for HIV infected
  children, needs from schooling.
• ? Additional pension for OVC carers.

12
LESOTHO PENSION VULNERABLE GROUPS - 2

• Women
• Over 75 pensioners are women. In Lesotho a
  large proportion (gt 80) are literate
• Strong evidence that pensioner and households
  agree that pension belongs to the pensioner, who
  must the decide allocation within household.
• Substantial evidence that position of old women
  is enhanced, spending allocations accepted (lt5
  of occasions where pensioner subject to violence)

13
LESOTHO PENSION VULNERABLE GROUPS - 3

• Case History Me Marabeki
• Widow looking after 4 OVC grandchildren in her
  small hut in local village. Little cash income
  from her own children.
• Did not claim pension because she misunderstood
  documentation required. Childrens situation seen
  by teacher. Arranges for her to get pension
• But, chooses to spend money on getting rid of her
  mourning clothes, worn for death of daughter and
  fund family celebrations

14
LESOTHO PENSION NUTRITION - 1

• 55 of Basotho households classified as
  chronically hungry In mountain areas may be
  70. No cash incomes, subsistence farming
  opportunities poor.
• Pension makes significant difference, chronic
  hunger cut by up to 50, esp. in v. poor areas
• Not much evidence of pension invested in
  improving yields from subsistence farming. C 3
  used for seed, fertilizers, more livestock

15
LESOTHO PENSION NUTRITION - 2

• 60-70 of pension cash used for food for
  household. Pensioner get 40 of this.
• Cash grant used to buy shop foods, esp. meat,
  dairy products, sweets , pasta, rice, biscuits
• Evidence of some extra expenditure on alcohol and
  tobacco
• Contributions to church and charities

16
LESOTHO PENSION FOOD SECURITY

• Little evidence from surveys/ but, pension seen
  as a vital safety net for food purchases in
  times of drought and failure of subsistence
  farming
• Pension could probably supply half the amount of
  food needed for household in month, if no other
  resources available. Note availability of free
  primary school meals
• Food expenditure smoothing, traders will make
  short term loans for food on the security of
  pension

17
LESOTHO PENSION ECONOMIC EMPOWERMENT 1

• Expenditure smoothing
• Local job creation, but small scale and may be
  replacing previous unpaid services
• May help most those that already are above the
  poverty /chronic hunger line

18
LESOTHO PENSION ECONOMIC EMPOWERMENT- 2

• Some , mostly anecdotal evidence that some women
  are released from role of care givers to the
  elderly relations. This now provided by local
  non-relatives paid for from pension. Released
  women can get jobs in textile factories

19
LESOTHO PENSION ASSET ACCUMULATION

• Little evidence of this from surveys. 98 of
  pension is spent, almost all on immediate
  consumption. But
• I) Surveys done soon after pension first grated.
  Asset accumulation needs longer term planning
• II) Surveys done in areas with few of the 20 in
  general population considered well off ( Lesotho
  has very skewed income distribution, low A prop
  to save, poor financial structures to help
  saving)
• SHOULD MORE ATTENTION BE GIVEN TO HELPING
  PENSIONERS SAVE AND INVEST ?

20
LESOTHO PENSION WIDER ECONOMIC IMPACTS

• Total flow of new cash into the area probably
  too small to have much general impact. Note that
  national, expenditure multiplier very small
  because 70 of consumption goods are imported
• Some immediate benefits for local traders on
  pension days, but traders often not from the
  permanent community.
• Lesotho has a poor record of using dextra
  funding for development purposes. Poor
  implementation, alleged political interference.
  Main value of pension is at micro level of
  household.

21
Malawi

• Candace Miller
• Boston University
• candace_at_bu.edu

22
Anthropometry (children lt 5)
C I Difference in differences (percentage points) p-value
Stunting (height for age z score lt-2) Baseline 52 51
0-36 months n134 Endline 59 48 -12.0 0.33
Underweight (weight for age z score lt-2) Baseline 38 36
0-36 months n139 Endline 29 14 -13.2 0.21
Key C Comparison/control household I
Intervention household
Miller, Tsoka, Reichert. (2009). The Malawi
Social Cash Transfer and the impact of 14 per
month on child health and growth. In review.
23
Height gain and recent illnesses
Height gain in centimeters
Recent illnesses
Key C Comparison/control household I
Intervention household
Miller, Tsoka, Reichert. (2009). The Malawi
Social Cash Transfer and the impact of 14 per
month on child health and growth. In review.
24
Education and child work
Age of child / youth Age of child / youth Boys Boys Boys Boys Girls Girls Girls Girls
C I Double Difference ( points work) Double Difference ( points work) C I Double Difference ( points work) Double Difference ( points work)
Income Generating Activities Baseline 12 11 16 16
Endline 18 6 -11 15 6 -10
25
Food security and nutrition (16,16)

• Focus Groups with children (n169 children 16
  groups)
• We are not starving anymore because they buy us
  food everyday (F, 9)
• We are happy nowin the past we were not eating
  in the afternoon and sometimes in the evening.
  Now we are able to eat 3 times a day. We are able
  to eat good food like meat and fish (F,10)
• Now we dont frequently get ill since we are
  eating different types of food to make our bodies
  strong (M,15)

26
Education Increased attendance/performance,
school fees paid, purchased uniforms (16/16)

• We are also benefiting by getting good education
  because we have uniforms and we are going to
  school everyday (F,13)
• We have more chance because we have everything
  in life, our parents receive money from the cash
  transfer we have food, school uniforms, exercise
  books, pens, and clothes. While our friends dont
  have that(F,13)
• We are also scoring good grades because we are
  very attentive in class because we eat good food,
  we have uniforms and books and pencils due to
  scheme money (F,12)

27
PLWHA (in depth interviews n24)
Percent describing positive impacts for themselves Percent describing positive impacts for themselves
Personal health improved 10 of 24
Transportation costs paid for 9 of 24
Better Nutrition / food security 17 of 24
Employment 9 of 24
Buying livestock 15 of 24
Purchasing other basic necessities 15 of 24
Percent describing positive impacts on family
Adequate housing 13 of 24
Food security 16 of 24
Education costs met 18 of 24
Family health improved 6 of 24
Purchasing other basic necessities 15 of 24
28
Complementary healthcare

• I am now able to receive the ARV at hospital
  I am able to buy medicine or use the money for
  transport to receive the ARV tablets. (F,30)
• The frequency of falling sick has dropped now
  since receiving the transfers because I have
  something for food and painkillers (F,55)
• So we use the money that we receive for transport
  once every 2 months to go and collect these ARV
  drugs from hospital (F,38)

29
Economic support / empowerment Can work or
buy labor

• I can now have my garden cultivated because the
  money I receive from the scheme enables me to pay
  for labornowadays, I am able to have at least
  half of my field cultivated. (F,33)
• We use this money as pay wages to those who are
  farming on our field. (M,48)
• Since 2004, I was bed ridden and could not work
  but now I can do a little. (F,30)
• I also grow sunflower for making cooking oil. (F,
  28)
• I am able to buy chickens and managed to pay for
  casual labour. (F, 44)

30
Livestock/crops

• I managed to buy a goat with this moneyI had to
  save the money for 3 months... I also bought 3
  chickens. (F, 42)
• I bought one goat. I also bought 3 chickens but
  all these died. I bought one bag of fertilizer
  (F, 54)
• I was also able to buy goats and build kraal for
  livestock. Able to keep buy fertilizer for this
  coming farming season. Managed to pay for ganyu
  people (F, 43)
• I afford fertilizer last season and yield one
  ox-cart .... I have also managed to buy a goat
  in two installments. (F, 37)
• I bought one pig which had piglets, but now there
  is one piglet, others died. I saved for
  fertilizer. Salt and milling fees were a problem,
  but now they are available. (M, 33)

31
Food security and diversity
HH food consumption is less than enough
HH members consume at least 2 meals daily
Households consuming meat / fish past week
Food Diversity Score
Miller, C., Tsoka, M, and Reichert, K. (2009).
The impact of the Social Cash Transfer Scheme on
food security in Malawi. In review.
32
Double difference intervention comparison
groups Weekly food expenditures
This is the difference in the amount of MK per
week added into the local economy for food due to
the SCTS for various items
33
Asset ownership
34
Asset ownership
SCTS households Comparison (non-SCTS households)
Axe
Sickle
Goats
Chickens
35
Percent of households that purchased items
between baseline and endline
SCTS households Comparison (non-SCTS households)
36
Larger Economic Impacts

• About 90-95 of cash transfer money is fed back
  into local economy through purchase of
• Food
• Household items
• Services
• Many businesses benefit through increased sales
  smoothing seasonal cycles in turn, they build
  capital
• Cash transfers recipients indirectly support
  family members outside of the household through
  labor, gifts, loans
• Households that receive indirect inputs are able
  to purchase food, household items, healthcare etc.

37
Zambia

• Gelson Tembo
• University of Zambia
• tembogel_at_zamnet.zm tembogel_at_yahoo.com

38
Impact of Social Cash Transfers on Household
Welfare, Investment and Education in Zambia
Ministry of Community Development and Social
Services
39
Introduction

• 5 Social Cash Transfer (SCT) pilots in Zambia
• Kalomo launched in May 2004, to be used as a
  model for the proposed national scale-up
• Unconditional SCT scheme targeting the 10 most
  labour-constrained ultra-poor
• The scheme was set up to meet the following
  objectives
• Reduce extreme poverty amongst the poorest 10 in
  the district
• Generate information on feasibility, costs and
  benefits and all positive and negative impacts of
  SCT as one component of a comprehensive SP
  programme
• Community-based targeting

39
40
SCT Pilots in Zambia
District TA Agency Special Features
Kalomo GTZ PWAS, capacity building, scaling up
Monze GTZ the above, plus soft conditionality
Kazungula CARE transfer level
Chipata CARE urban transfers, school allowance
Katete CARE old-age pensions (age 60 plus)
TOTAL
Source adapted and updated from Zambia, 2007,
The Pilot Social Cash Transfer Scheme, Zambia,
Summary Report, 5th Edition, Lusaka MCDSS/GTZ,
May (Table 5, p.14)
40
41
Objectives of the Impact Evaluation

• identify factors that explain household
  participation in the SCT programmes among
  eligible households
• determine the impact of the SCT interventions on
• welfare
• investment
• education
• determine the impact of asset wealth on the
  effectiveness of the SCT programmes

41
42
Impact Evaluation
District Number of households Sample (bene controls) Methodology
Kalomo 3,300 886 Retrospective (MASDAR/RHVP)
Monze 3,300 2,748 Baseline (MASDAR/RHVP)
Kazungula 627 200 Retrospective (CARE)
Chipata 1,400 200 Retrospective (CARE)
Katete 1,000 - None
TOTAL 9,627
42
43
Methodology

• Propensity score matching (PSM) through
  odds-weighted regression analysis was used to
  estimate impact
• Propensity score satisfied a battery of balancing
  and common support tests
• Impact was estimated both as an aggregate measure
  and disaggregated by asset wealth
• A household was categorized as asset poor if it
  fell in the bottom two quintiles of the
  distribution of an asset wealth index,
• Wealth index constructed using principal
  components analysis (PCA)

43
44
Characteristics of the sample households
Variable Chipata Kalomo Kazungula
Number of meals children per day 2.4 2.4 1.5
Household income (000 ZMK) 979.0 402.9 366.6
Age of household head (years) 55.4 58.9 61.9
Widow-headed households () 65.0 63.0 55.0
Male-headed households () 28.0 36.0 31.0
Orphanhood () 78.0 47.0 44.0
Effectively active members () 19.3 25.1 15.2
44
45
Determinants of Participation in the SCTs

• Largely consistent with postulated targeting
  criteria
• Widowhood (Kalomo/Chipata)
• Age of HH head (Kalomo/Chipata)
• Wealth status (Kalomo)
• Household composition (Kazungula)
• But significant exceptions
• HH head gt61 years were 39 less likely
  (Kazungula)
• Male-headed (31-45 years) were more likely
  (Kalomo)
• Married HH head more likely than
  divorced/separated (Kazungula)
• HHs closest to main roads, schools, clinics more
  likely (all)

45
46
Impact Estimates
46
47
Impact Estimates Welfare

• Impact on consumption expenditure unambiguously
  positive and statistically significant in all
  three districts, regardless of the wealth status
• Consumption effect was greatest on non-food items
• In Chipata, effect of SCTs on non-food
  consumption expenditure was 9.6 percent greater
  among asset poor households than it was among
  their asset non-poor counterparts

47
48
Impact Estimates Investment

• Impact of participation on value of small
  livestock was unambiguously positive and
  significant in the two rural schemes of Kalomo
  and Kazungula
• Impact on investment in micro-enterprises was
  positive and significant in the urban-based
  scheme of Chipata
• Program effects on value of livestock were
  greatest among asset poor households (Kalomo
  only)
• Evidence of threshold effects on asset
  accumulation
• Impact on asset accumulation positive and
  significant only among relatively non-poor
  households
• Asset-poor households may require higher transfer
  levels
• No discernible effect on cultivated land area in
  all three districts

48
49
Impact Estimates Education

• Impact on educational outcomes was mixed
• Enrolment rates improved only among boys in
  Kalomo
• Attendance rates improved for both male and
  female children in Chipata (soft condition of
  school attendance linked to payment of
  educational premium), regardless of asset wealth
  category
• Also improved for asset poor households in Kalomo
• In Kazungula, attendance rates improved only
  among girls in asset non-poor households

49
50
Summary and Conclusions

• SCT impact
• Unambiguously positive for household welfare
• More pronounced for non-food items
• Asset wealth matters only in urban scheme
• Investment effects significant but varied by
  design and setting of scheme
• Small livestock in rural schemes
• Micro-enterprises in urban scheme with
  educational premiums
• Threshold effects with respect to asset
  accumulation
• Educational effects were mixed
• Enrolment rates improved only among boys in
  Kalomo
• Attendance improved unambiguously in scheme with
  attendance soft condition
• Attendance effects more heterogeneous in rural
  schemes
• In conclusion, SCTs have scored but could be
  improved
• Some outcomes require more time, design
  adjustments
• Need to recognize heterogeneity among eligibles -
  No one size could fit all!

50