Improving Income and Food Security through Cash Transfers in Southern Sudan: An Evaluation of the Save the Children’s Pilot Project in Northern Bahr el Ghazal State, southern Sudan “With the cash transfer, I know that if my baby is sick, I can take her to hospital. When she is old enough to go to school, she will go.” – Cash for Work recipient, Rupkou Boma Natalie I. Forcier September 2010
27
Embed
Save the Children - Cash Transfer project evaluation report
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Improving Income and Food Security through Cash
Transfers in Southern Sudan:
An Evaluation of the Save the Children’s Pilot Project in Northern Bahr el Ghazal State, southern Sudan
“With the cash transfer, I know that if my baby is sick, I can take her to hospital. When
she is old enough to go to school, she will go.” – Cash for Work recipient, Rupkou Boma
7 See Haddad, Hoddinot & Alderman, ‘Intra-Household Resource Allocation in Developing Countries: Models, Methods and
Policies’, 1997, and Khogali & Takhar, ‘Empowering Women Through Cash Relief in Humanitarian Contexts’, 2002.
Cash Transfer Project Evaluation, September 2010
15
The beneficiary households (n=1400) were distributed across seven bomas within Baach Payam,
with the total number of direct project beneficiaries estimated to be 10,854 individuals, which is
equivalent to approximately 23% of the population of the payam.8 Although the overall cost of
the cash transfer per beneficiary is low, it is still is worthwhile to explore exploiting potential
economies of scale wherein the number of beneficiaries could be increased without a significant
increase in stabilized overhead costs. One possible scenario would be to increase the number of
beneficiaries within the existing locations, which was also suggested and requested by project
staff and beneficiaries alike.
2.1.3 Effectiveness of Targeting
Save the Children created the initial targeting criteria, which was then augmented and
implemented by community leaders. The criteria ensured that households with the following
characteristics would be targeted to be a part of the project:
! Vulnerable returnees and their host communities with households having the following
characteristics:
- no cattle/livestock owned, or less than a defined small number (e.g. < 3 cattle, 5
goats);
- less than a defined small acreage of farmland (e.g. < 0.5 acres);
- very large families,
- no, or low food/cash crop harvest, animals not producing milk;
- no fixed employment or petty business;
- no food stocks held;
- no petty trading or no small business;
- households of widows, widowers, orphans, aged, disabled;
- no kinship support
- single woman headed households
Project staff, community leaders, beneficiaries, and non-
beneficiaries in all five locations all indicated that the
targeting process was transparent and fair and that there
was no tension between beneficiaries and non-
beneficiaries. The positive review of this crucial aspect
of the project can be attributed to two processes:
1. Community leaders were involved in the design
of beneficiary criteria and the selection of
individual beneficiaries for both cash-for-work
and unconditional cash transfers. Even though it
was understood that due to limits on the number
of beneficiaries not all vulnerable households 8 Based on baseline mapping exercise and South Sudan Relief and Rehabilitation Commission records
(SSRRC)
"In the future, the number of people in cash-for-work should increase. Cash-for-Work benefits everyone in the community." - CFW Recipient, Rupkou Boma
Cash Transfer Project Evaluation, September 2010
16
could participate, this still did not result in tension between beneficiaries and non-
beneficiaries. However, community leaders and participants in all five locations did
request that in future rounds of the project the number of beneficiaries be increased.
2. Vulnerable households who did not benefit from the cash transfer project were given
priority in the distribution of food cultivated from cash-for-work projects. Community
members felt this was the fairest use of the food products cultivated and greatly improved
the overall support of the project by the community and reduced tensions between
beneficiaries and non-beneficiaries.
Participants of FGDs in all four locations noted that there were households that were excluded
from the project that should have been included; both among cash-for-work and unconditional
cash transfer beneficiaries. However, none of those interviews felt that there were individuals
who had been included in the project that should have been excluded. Individuals in all four
locations indicated that the number of beneficiaries should be increased. If it is not possible to
increase the number of beneficiaries, those in Rupkou and Rumdier Bomas suggested that later
rounds target households that were not previously able to participate in the program to ensure
that as many as possible from the community are able to benefit from the project.
Table 5 summarizes the number and type of beneficiaries for each boma.
Table 5: Profile of beneficiaries by location
Cash for Work Unconditional Cash
Transfer
Boma Adults Children
Total
CFW Adults
Total
Children
Total
UCT
Total
Beneficiaries
Chuomator 872 1421 2293 232 441 673 2966
Rupkou 543 824 1367 135 256 391 1758
Mathiangdit 305 429 734 49 137 186 920
Wunatemei 435 628 1063 82 202 284 1347
Rumdier 550 732 1282 88 134 222 1504
Panabiei 320 467 787 83 183 266 1053
Angot 411 508 919 98 289 387 1306
Total 3436 5009 8445 767 1642 2409 10854
2.1.4 Effectiveness of Delivery
Distribution took place in Mobil and Warawaar Markets, with each location serving four and
three recipient bomas respectively. In line with the SSOP, community leaders and community
supervisors were informed of the distribution date 24 hours prior to the distribution. Despite the
short time window provided for beneficiaries to travel to the distribution sites, monthly PDM and
cash distribution reports indicate very high turnout by beneficiaries for collection of the
distribution (98% for all but the second distribution, which showed an average turnout of 96%).
The lowest turnout rates were consistently demonstrated by beneficiaries from Angot Boma, the
Cash Transfer Project Evaluation, September 2010
17
most distant cash distribution location. However, even this reduced turnout never fell below
83%, with an average turnout of 91%. This demonstrates: (1) the project developed effective
communication networks to convey information regarding distribution in a efficient manner to
mobilize beneficiaries to attend distributions; (2) distribution locations were feasibly accessible
for beneficiaries to access on a monthly basis without associated costs excessive enough to deter
or prevent collection of the transfer.
Save the Children staff also indicated that in several instances when unconditional cash transfer
recipients (specifically the elderly or disabled) were unable to collect the transfer from the
distribution points, relatives were sent with their beneficiary card to collect on their behalf. The
relationship between the individual presenting the card to collect the cash transfer and the
actually beneficiary was confirmed by community leaders present for the distribution to ensure
that fraud was not occurring. If this was not possible, community supervisors and assistant
project managers were able to deliver the transfer to the bomas in the following days.
All beneficiaries and community leaders interviewed indicated satisfaction with the timeliness
and location of the delivery and distribution of funds. Despite the distance that some
beneficiaries were required to travel to collect the funds, beneficiaries indicated that collecting
cash in the market allowed them to immediately purchase required items and thus reduce the
amount of money kept on their person while returning home.
Similarly, placing distribution points at local government authorities within Mobile and
Warawaar Markets allowed for police security to be present, which was appreciated by
beneficiaries. In addition, cultural attitudes towards the intervention also ensured security, as one
beneficiary noted, “No one would interfere with this project or those carrying money from the
market after distribution. To do so would be a sign of disrespect, and the community is greatly
appreciative of the project” (10/9/2010, Rumdier Boma). Of all the beneficiaries and
stakeholders interviewed, no one reported any security problems at any point during the cash
distribution or immediately afterwards. No evidence of fraud or corruption on the part of staff,
beneficiaries, or stakeholders was found.
2.1.5 Effectiveness of Disaster Risk Reduction Training and Nutrition
Promotion Components
As discussed previously, the project also included two training components covering disaster risk
reduction and nutrition education to be carried out in the beneficiary bomas. Interviews with
project staff indicated that the nutrition component was started late due to a delay in identifying a
qualified consultant to design the program. During the FGDs, the disaster risk reduction training
was not mentioned as a component of the program. When beneficiaries and community leaders
were asked directly about this component, it was shown that it was not clear that it was
intertwined with the cash transfer project. However, the only areas where disaster risk reduction
and cash-for-work components were specifically linked was in the lowland bomas (Wunatemie
and Chomator), where dykes were constructed by beneficiaries. No interviews or FGDswere
conducted in these areas.
Cash Transfer Project Evaluation, September 2010
18
Only one of the bomas (Panabei) mentioned the nutrition component during the focus group
discussions. Participants in this location cited it as a key benefit of the program and recited
several of the key messages of the program without prompting. Other locations responded
positively when asked directly about the component, but were not able to recall the specific
content messages. This is most likely due to the delay in the start of the component, as the
project had difficulty identifying a qualified consultant to develop the curriculum.
The nutrition and disaster risk reduction components, although relevant, appeared to lack
cohesion and integration within the overall cash transfer project, as their purpose was unclear to
community leaders and beneficiaries. However, staff did attempt to link the components with the
overall project. It is anticipated that as the program continues, the nutrition component in
particular, will be able to play a more prominent role within the overall project as intended.
2.2 Use of Transfers by Beneficiaries
Data was collected in the baseline assessment, monthly monitoring reports, and the evaluation
regarding beneficiary’s income, expenditure, asset acquisition, and savings. These data sets were
analyzed to identify similarities and differences among both beneficiaries to non-beneficiary
groups, as well as baseline and endline data. In particular, the data indicated the following:
• Beneficiaries were 30 percentage points more likely spend money on education than
non-beneficiaries. Over three-quarters, 78% (n=146) of beneficiaries reported that they
had spent money on education related expenses in the three months prior to the survey, as
compared with only 48% (n=20) of non-beneficiaries. Focus group discussions indicated
that one of the priority expenditures for beneficiaries was educational related expenses,
and beneficiaries in all four locations indicated that the only means by which they had
been able to afford to send children to school was because through using the cash
transfer. This was confirmed in quantitative data, which indicated that beneficiaries
(57%, n=193) were 16 percentage points more likely to have a school-aged child that was
currently attending school than non-beneficiaries (41%, n=33).
• Beneficiaries were 20 percentage points more likely to report expenditures on
medical services than non-beneficiaries. Beneficiaries in all five locations indicated
that medical services were a priority expenditure for the cash transfer. In particular,
mothers noted that they were now able to afford to take their children to the doctor when
they were ill. Children under five from beneficiary households were more than three
times as likely to receive medical advice when showing signs of pneumonia, such as
rapid breathing and coughing, (71%; n=25) than non-beneficiaries (18%; n=10).
• Beneficiaries were twice more likely to have invested in cultivation equipment
(49%; n=91) and household materials9 (42%; n=79) in the past three months than
non-beneficiaries. In addition, beneficiaries were 21 percentage points more likely
(89%; n=169) to report planting land in the previous season. The most commonly planted
crops included sorghum, pumpkins, sesame, and maize, without a significant difference
9 Household materials refer to any items used in the construction or renovation of traditional homes.
Cash Transfer Project Evaluation, September 2010
19
between beneficiary and non-beneficiary groups. Sorgum was by far the most frequently
planted crop, and FGDs in all five locations referenced a less productive than anticipated
sorghum harvest in 2010 due to striga lutea, a parasitic witchweed that stunts the growth
of the plant. The mean number of feddans cultivated in the previous season in 2009 by
respondents did not differ significantly between beneficiaries and non-beneficiary (2
feddans and 1.8 feddans, respectively).
• Beneficiaries were two and a half times more likely to have made payments towards
debt reduction, fines or taxes (24%; n=46) than non-beneficiaries (10%; n=4) in the
past month. Ability to reduce debt is significant both for the individual and the vendor,
as it unlocks future income for families to use towards services such as education and
health rather than repayment, and also relieves local vendors, relatives and families
members of debt owed.
• There was no significant difference in consumption patterns of alcohol and tobacco
between beneficiaries and non-beneficiaries. One of the most common criticisms of
cash transfer projects is that the funds can be used for negative or anti-social purposes,
specifically alcohol and tobacco. However, as the data indicates, beneficiaries are no
more or less likely to consume alcohol than those who did not receive the cash transfer.
Of those interviewed, 15% (n=28) of beneficiaries and 12% (n=5) of non-beneficiaries
indicated having purchased alcohol or
tobacco in the previous month.
2.3 Overall Project Impact
“Before the project started, I used to sit under a tree near the market and drink tea. Then the project started. I was able to save a little bit and buy a goat. One day, my goat became pregnant. Now I am able to sit under the tree and drink tea with milk.” -UCT recipient, Mathiangdit Boma
2.3.1 Income and Expenditures Throughout the duration of the project, monthly monitoring reports showed that the 80 SDG cash
transfer received from Save the Children constituted 35% of overall household income. This
percentage did not fluctuate significantly from month to month.
Cash Transfer Project Evaluation, September 2010
20
Figure 1: Beneficiary expenditures by month and type based on monthly monitoring
reports, 2009/2010
As Figure 1 indicates, the percentage of income spent on food varied from month to month
reflecting seasonal differences. When food purchases decreased, asset purchases increased
indicating that the cash transfer project was successful in enabling beneficiaries to build assets.
As one woman in Panabei Boma explained, “From the beginning, the first money was used on
food because it was the hunger season. After the hunger season ended we started buying
chickens and goats. As the hunger season approached again, we went back to buying food. Some
of us still have our chickens and goats, but if the project doesn’t begin soon we’ll be forced to
sell them.”
2.3.2 Asset Depletion and Coping Mechanisms
Less than half (42%; n=92) of respondents indicated that their household had experienced an
event that affected its usual ability to eat and/or buy foods of the quality, quantity and variety
they prefer. The five most common shocks experienced were the same among beneficiaries and
the non-beneficiary groups (Table 6):
• Increased price of food
• Drought/dry spell
• Insecurity or violence (in areas near Abyei border)
• Floods (in lowland areas)
• Sickness of household member
Cash Transfer Project Evaluation, September 2010
21
Table 6: Shocks experienced by households during life of the project
Shock Most Commonly Employed Coping Mechanism
Beneficiary Non-beneficiary
Increased price of food
(n=58)
Purchase food on credit
(37%; n=15)
Purchase food on credit
(50%; n=5)
Drought/dry spell
(n=57)
Nothing
(23%; n=10) or
Eat fewer meals per day
(23%; n=10)
Go an entire day without meals
(57%; n=5)
Floods
(n=47)
Reduce expenditures on health or
education
(22%; n=15)
Collect wild foods, hunt or harvest
immature crops
(33%; n=5)
Insecurity or violence
(n=32)
Eat fewer meals per day
(30%; n=6)
Eat fewer meals per day
(50%; n=4)
Sickness of household
member
(n=25)
Reduce expenditures on health or
education
(43%; n=7)
Distress sale/slaughter of livestock
(51%; n=4)
Although the shocks were the same for both groups sampled, the coping mechanisms employed
by beneficiaries as opposed to non-beneficiaries varied. In particular, the data shows that
beneficiaries were less likely to be forced to deplete assets or make significant nutritional
reductions than those sampled from the non-beneficiary group.
2.3.3 Education, Health and Nutrition
• No significant difference was observed in school attendance in the week preceding
the survey; however children from beneficiary households were less likely to miss
school to engage in work. Parents interviewed reported that a majority of children had
attended all days of school in the previous month, with the non-beneficiary group
reporting slightly higher rates of perfect attendance (63%; n=20) than beneficiaries (56%;
n=123). These rates were significantly lower among children in Rupkou Boma (37%;
n=21). Among beneficiaries the most common cause for missing school was illness
(79%; n=52), whereas among the non-beneficiary group the most common cause was
work (50%; n=4). The lower perfect attendance rates among children in Rupkou Boma
were all attributed to illness and all came from beneficiary households. In the FGDs in
Panabei Boma, beneficiaries noted that children were now more likely to attend school on
a regular basis because families no longer needed them to beg on the street or steal for
food. Children in Rumdier Boma noted that they were now able to attend school this
year, as last year they had to spend their days collecting wild foods or working as
laborers in the market.
• Livestock ownership among beneficiaries has more than doubled since the
beginning of the project. Both the baseline assessment and this evaluation study
collected information on cattle, chicken, goat and sheep ownership among project
beneficiaries. It is unclear to what extent this increase was caused by the cash transfer
project or other outside factors. The following table demonstrates the change in the
Cash Transfer Project Evaluation, September 2010
22
"One month you have some extra money, you
can buy a goat or chicken." UCT Recipient, Rupkou
percentage of beneficiaries indicating that they own livestock.
The greatest livestock ownership exhibited was the acquisition
of chickens and goats, which indicates a positive impact on
gender empowerment as goats and chickens are generally
female-acquired and owned livestock. Furthermore, goats and
sheep care were considered “stepping stones” for building
wealth in poor families (Southern Sudan Livelihood Profiles
2007). Focus group discussions revealed that goats and
chickens were also the primary store of value within the
communities and thus this acquisition can be viewed as a form
of savings as well as asset building.
Table 7: Absolute livestock ownership at baseline and evaluation
Type of Livestock Baseline Evaluation % Pts Increase
Cattle 13% 43% 30
Chickens 27% 79% 52
Goats 32% 79% 47
Sheep 6% 29% 23 *Absolute livestock was define as owning one or more of the specified livestock
• Beneficiaries were able to feed their children more frequently, even during the
hunger season, than non-beneficiaries. As Table 8 indicates, the difference between the
number of meals consumed by adults in beneficiary households versus non-beneficiary
households was not statistically significant. However, a significant difference can be seen
in the mean number of meals consumed by children during the hunger season – with the
average number of meals consumed by children of beneficiary households being 2.0,
compared to only 1.6 in non-beneficiary households. However, those interviewed
indicated that the ability to feed their children more regularly and healthier foods were
directly dependent on the availability of funds as a result of the cash transfer, noting
“when the project stops, the child malnutrition will be back” (Rupkou Boma).