This report is the endline activity evaluation of the World Food Programme’s (WFP) Local and Regional Food Aid Procurement Programme (LRP) in Kenya. The programme is funded by the United States Department of Agriculture (USDA) and is implemented in three arid counties in north-west Kenya: Baringo, Turkana and West Pokot. The evaluation is commissioned by the WFP Kenya Country Office and follows a baseline assessment conducted in April 2018 by the same evaluation team. The main objective of the final evaluation is to assess the performance and results achieved through the LRP in the three targeted counties over the project period from September 2017-March 2020. The evaluation serves the dual and mutually reinforcing objectives of accountability and learning.
The LRP was implemented from October 2017 to March 2020 using a grant of USD 1 million. The LRP was specifically designed to support the implementation of the Government’s Home-Grown School Meals Programme (HGSMP) by supporting farmers, farmer organizations (FOs) and local traders to produce and provide sufficient food for the ongoing school feeding activities in the three targeted counties. USDA also gave approval for WFP to locally purchase sorghum and cowpeas from FOs for use in the school feeding programme. There have not been any budget revisions or changes to the programme design since it was approved in September 2017, and there are no sub-recipients under this agreement. No other donors have contributed funds to the LRP per se, however some of the FOs supported under the LRP have also benefitted from support from other donors. The LRP is implemented directly by WFP Kenya, in close collaboration with the Ministries of Education (MoE), Agriculture and Irrigation (MoA) and Health (MoH), at county and sub-county levels.
The key objectives of the LRP are as follows:
-
Improve effectiveness of food assistance by improving cost-effectiveness and improving timeliness.
-
Increase the capacity of suppliers and school meals procurement committees to effectively and efficiently procure local commodities for school’s meals, promoting sustainability of school feeding;
-
Strengthen local and regional food market systems, improving access to culturally acceptable commodities and connecting them to Government of Kenya HGSMP; and
-
Improve nutrition of students by increasing access to and use of various high quality, nutritious and culturally appropriate foods in school’s meals.
To achieve these objectives, the LRP has five main activities: 1. Assessment of local food systems, 2. Capacity building for national and county institutions, 3. Capacity strengthening for local traders and FOs, 4. Develop school meal menus using local and nutritious produce, and 5. Procure locally produced, drought-tolerant crops. The main users of this evaluation report are WFP Kenya and USDA, both of whom have a responsibility to ensure that the evaluation proceeds as per the programme agreement between the two parties. There are also several other internal and external stakeholders with an interest in the evaluation findings including the MoE, MoA and MoH and other WFP offices.
Context: WFP and the MoE have jointly implemented a school meals programme in Kenya since 1980, targeting the most food-insecure counties with the lowest enrolment and completion rates and high gender disparities. Initially, school meals used an in-kind modality with food commodities provided directly by international donors including the USDA/McGovern-Dole. In 2009, the Government of Kenya started the national HGSMP to provide meals to children at school, using a cash transfer modality, providing cash to schools to enable them to buy their food directly from local traders and farmers. The HGSMP stimulated local agricultural production through purchase of food from smallholder farmers and local traders. The arid counties in Kenya’s north were the last to transition to cash-transfers, to give adequate time to assess the local markets and ensure they would be able to cope with the demand of school feeding. All the targeted LRP schools in Baringo and West Pokot were handed over to the HGSMP in September 2017. WFP officially handed over all remaining WFP-supported schools, including the LRP schools in Turkana to the government’s HGSMP in June 2018. This was on the understanding that the government would provide cash to schools to procure their own food for school meals.
Methodology: This evaluation uses the OECD-DAC international evaluation criteria of relevance, effectiveness, efficiency, sustainability and impact to evaluate the implementation of the LRP project. The evaluation team has also ensured that gender equality and the empowerment of women (GEEW) was mainstreamed through the evaluation approach and methodology by ensuring that whenever possible, men and women, boys and girls from different stakeholder groups, participated in the evaluation, and data has been gender disaggregated when possible. Where possible, the evaluation has also tried to assess whether the project had different impacts on male and female farmers and traders. The evaluation used mixed methods and collected both quantitative and qualitative primary data to answer each of the evaluation questionsincluding – 1. Was the project designed to reach the right people with the right type of assistance?
Is the project aligned with national government’s relevant policies and strategies? 3. To what degree have the interventions resulted in the expected results? 4. Did assistance reach the right beneficiaries at the right time? 5. Were there any unintended outcomes, either positive or negative? 6. What internal and external factors affected the project’s results? 7. To what extent is it likely that the benefits of the project will continue after the end of the project? And 8. What are the key factors that affect the likelihood of sustainability of the results of the project?
The evaluation also utilized WFP Kenya’s own project monitoring data, to complete the Performance Monitoring Plan (PMP) required by USDA. The evaluation incudes three quantitative surveys: FOs, local traders and schools, as well as qualitative field interviews and focus group discussions, to assess the effect of the program against the LRP objectives. The evaluation methodology has three identified limitations:
Firstly, that two schools from the original baseline sample were not included at endline.
These two schools were replaced with schools with 99% matching characteristics to the original schools, so there should not be any effect on the group comparison of results from baseline. Secondly, that the school survey was partially conducted during the Term 1, 2020 half term holidays. This meant the survey teams had to organize times with each school to conduct their visit, to ensure that appropriate school personnel would be available to answer the survey questions. The mitigation measures were successful, and personnel from each school were located to participate in the survey. Lastly, although there is some gender disaggregated data available from WFP, the LRP Results framework does not include gender indicators. Furthermore, the beneficiary units are largely genderless (schools and farmer organizations). This has limited the genderrelated findings of the evaluation.
Key findings: Evaluation question 1: Was the LRP relevant? The LRP was designed in collaboration with MoE, MoA and MoH and the evaluation found the design to be relevant to support HGSMP transition in the three targeted counties. The planned modality of cash-transfers for providing school meals was the preferred option of all evaluation key informants, as it not only supports local farmers and trader, but aligns well with the government’s existing HGSMP. Overall, the endline evaluation found that the LRP complements the HGSMP, and aligns well with key Government of Kenya policies, strategies and framework, as well as WFP’s own polices and with the direction of other United Nations actors in Kenya.
Evaluation question 2: Was the LRP effective? WFP and partners implemented most of the LRP project activities as planned. However, because of the 2017/18 drought, the MoE decided not to transition the targeted LRP schools onto cash-transfers as planned, but to keep them on in-kind assistance until food prices reduced. This has had a significant effect on the LRP results and some of the intended school-related objectives have not been achieved. Out of the 191 schools surveyed, only 18 non-LRP schools in West Pokot reported receiving cash-transfers for HGSMP as planned. None of the other schools received any resources (food or cash) from the MoE for school meals for Term 3, 20195 As a result, only around a quarter of schools were able to provide lunch every day that term. Although many schools tried to obtain food from other sources to supplement any left-over food from Term 2, many schools (n=75) were unable to provide any school meals during Term 3, 2019.
Through the LRP, WFP conducted several formal training exercises and workshops, in collaboration with the relevant government ministries. While the evaluation received positive feedback on all the training work, most participants did not have a chance to put the learning into practice due to schools not requiring local procurement as planned. The exception has been the procurement training for FOs and traders. As the HGSMP procurement process follows the government’s procurement process, traders have been able to use their learning to bid for other government tenders, particularly for food for secondary school and boarding school meals. Similarly, the training for FOs has contributed to more FOs monitoring product quality requirements. The market forums and other support to FOs has also effectively resulted in changes to FO production including a greater percentage of FOs aggregating and marketing members food commodities, and FOs in West Pokot reported selling a greater percentage of their production. This despite the in-kind provision by MoE reducing the potential market opportunities for FOs and acting as a disincentive to produce the promoted crops. The collaborative implementation approach taken by WFP with MoE, with MoA, MoH and MoTIC has also been an effective capacity strengthening approach, providing government personnel with added confidence to provide training on their own in future.
Evaluation question 3: Was the LRP efficient? One of the objectives of the LRP was to improve the timeliness of procurement of school meals commodities. This was done by providing schools with cash transfers, enabling them to do local procurement of commodities, hopefully resulting in more timely delivery of commodities to the schools. However, since the MoE did not provide any cash resources to LRP schools for Term 3, 2019, none of the LRP schools undertook a procurement process. The target of 90 percent of LRP schools completing procurement before the start of term has therefore not been achieved.
Further, none of the LRP schools received any in-kind resources from the MoE for Term 3, 2019, so the target of 90 percent of LRP schools having their food delivered before the start of term has also not been achieved.
Evaluation question 4: What was the impact of the LRP? The LRP was specifically designed to support the transition to cash-based transfers in the LRP schools. The impact of the LRP was therefore considerably reduced by the MoE’s decision to provide in-kind commodities to schools instead of providing cash-transfers to school as planned. As a result, the intended objective of improving the effectiveness of assistance through improved timeliness of procurement and delivery, improved cost-effectiveness has not been achieved. The objective to increase access to, and use of various high quality, nutritious and culturally appropriate foods in school’s meals of the school meals by using locally produced crops has been achieved only in Turkana Country and not in Baringo or West Pokot. This is because in Turkana, WFP directly procured and delivered the commodities to LRP schools.
The objective to increase the capacity of suppliers and school meals procurement committees to effectively and efficiently procure local commodities for school’s meals, has been partially achieved. FOs, traders and school personnel all reported attending training and gaining knowledge on HGSMP. School personnel also gained increased awareness of the importance of dietary diversity and how the school meal menus can be altered to increase the diversity of commodities. However, most of these groups were unable to put their learning into practice as no school-based procurement took place.
The final LRP objective was to strengthen local and regional food market systems, improving access to culturally acceptable commodities and connecting them to Government of Kenya HGSMP. Although the FOs and traders were unable to link into the HGSMP as planned, the LRP activities have resulted in several positive outcomesfor FOs and traders. These include increased awareness of FOs and traders of the HGSMP in general, and increased knowledge for FOs and traders on the required HGSMP procurement process. As a result of the LRP market forums, some traders were also able to supply commodities to other schools – secondary and boarding schools – that also implement school meals. Qualitative interviews also identified some negative outcomes for FOs and traders, as both groups were left with larger than usual volumes of commodities after planning to sell stock to the HGSMP schools. Some traders also allowed schools to buy
commodities on credit, pending future payment of the cash-transfers, however with uncertain plans for
reverting to cash, it is unclear when schools will be able to pay off those debts.
Finally, the evaluation commends WFP for the ongoing effort to include women in the programme and
impact the lives of women. Women have been actively encouraged to participate in all aspects of the
programme, and this active recruitment has succeeded, with women making up half the traders sampled
(46.3%) and more than half the FO membership.
Evaluation question 5: Are the results of the LRP sustainable? As a programme designed to support the
HGSMP, the LRP was never intended to be a sustainable programme in its own right. However, several of
the LRP activities have contributed to sustainable results. The decision by the MoE to implement an in-kind
modality reduced the effectiveness of the LRP, but it in no way reduced the sustainability of the HGSMP
itself. It is, however, difficult to clearly state whether some of the LRP results are likely to be sustained or
not as the sustainability of some results depend on the MoE’s future plans regarding providing cash
transfers. Regardless of the HGSMP modality, the LRP has succeeded in strengthening relationships
between traders and FOs and contributed to improving the policy landscape promoting smallholder
procurement. The least sustainable aspect of the LRP is likely to be the use of drought tolerant crops in the
school menus. Currently, those commodities are more expensive than maize and beans, so unless schools
are provided with a cash transfer rate than enables procurement of these commodities, they are unlikely
to purchase those commodities regardless of the nutritional or economic benefits to the community
Overall conclusions: The evaluation found that while all the planned LRP activities and outputs have been
completed, and the project has resulted in several positive outcomes, the LRP has not been able to achieve
the school-based objectives of improving the timeliness or cost-effectiveness of food assistance. This is a
result of MoE providing LRP schools with in-kind commodities instead of the cash-transfers that were
planned, and which the LRP was specifically designed to support. The LRP has however succeeded in
increasing the capacity of traders, FOs and school meals procurement committees on the procurement
processes of HGSMP, although none of these groups have been able to put their learning into practice.
Similarly, although capacity of school personnel on nutrition and diversity of the school menus has been
strengthened though training and the menu development workshops, this training and the resulting menus
have not been put to use in Baringo and West Pokot. In Turkana County, schools used the new menus to
incorporate sorghum and cowpeas into the school meals. The expected boost to the local and regional food
market systems that the HGSMP would have provided, has not resulted, although other market activities
have helped FOs and traders sell greater volumes. The evaluation found it was not possible for WFP to have
foreseen the sudden change in position on cash-transfers by the MoE especially after successful transition
to cash transfers in other arid areas prior to handover to government. Based on the findings and conclusions
of this evaluation, the evaluation team made the following recommendations:
• Recommendation 1: WFP and the MoE school feeding technical officers should continue to advocate
to the MoE senior leadership on the importance of school feeding, timeliness of delivery, timeliness of
cash disbursement and importance of providing daily school meals. This advocacy role can also be done
at county level in support of the draft ECDE framework.
• Recommendation 2: WFP and MoE should develop a position paper outlining the key decisions over
the last decade that have led to the largely cash-based modality of HGSMP that is seen today. The paper
should also include the evidence that underpins those decisions including recent market assessments
in the arid areas. The paper should then be used for ongoing advocacy with the MoE as described in
Recommendation 1.
• Recommendation 3: WFP and the MoE school feeding technical officers should consider ways to
support improvements in the MoE’s in-kind procurement and delivery system. This is especially
important if the MoE intends to continue providing in-kind support to the arid and semi-arid areas
which are among the locations most in need of the food and nutrition security improvements that the
HGSMP can provide.
• Recommendation 4: WFP should continue to work with county/sub-county governments to support
farmers, FOs and market system improvement in general. This is especially important for work involving
the MoA, as their authority has been devolved to the counties.
• Recommendation 5: WFP should continue to work together with MoH at national and county level, on
school health and nutrition related work, especially on activities that improve the nutritional status of
school-aged children such as the provision of nutritious, locally produced food commodities.
• Recommendation 6: WFP should look for funding to continue to the work started through the LRP
supporting farmers and FOs in Baringo and West Pokot as the LRP was the first opportunity for support,
and without the opportunity to put their learning into practice during the LRP, any gains made are likely
to regress without some reinforcement.
• Recommendation 7: WFP should continue to work with the MoA and other relevant ministries on the
promotion of small holder procurement. This should include continuing the development of county and
national level strategies and other means to increase opportunities for small holder engagement in
public procurement.
• Recommendation 8: WFP should continue to support MoE’s efforts to conduct regular programme
monitoring to ensure that HGSMP processes and procedures are being adequately followed.
• Recommendation 9: WFP and partners should continue to support the active inclusion of women in all
WFP projects and ensure that gender related indicators are present in future WFP projects, as
appropriate.
