The World Bankdocuments.worldbank.org/curated/en/...MET(s) Mobile Extension Team(s) ... Senior Global Practice Director: Juergen Voegele Practice Manager: Dina Umali-Deininger ...

Document of The World Bank

Report No: ICR00004143

IMPLEMENTATION COMPLETION AND RESULTS REPORT

ON A

GRANT

IN THE AMOUNT OF US$ 5.5 MILLION

TO THE

REPUBLIC OF KENYA

FOR THE

KENYA ADAPTATION TO CLIMATE CHANGE IN ARID AND SEMI-ARID LANDS (KACCAL) PROJECT

March 16, 2018

Agriculture Global Practice Africa Region

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

CURRENCY EQUIVALENTS

Exchange Rate Effective March 30, 2010

Currency Unit = Kenya Shilling (KES) 77 KES = US$1

US$ 1.00 = SDR 0.658

FISCAL YEAR

July 1 – June 30

ABBREVIATIONS AND ACRONYMS

ALRMP Arid Lands Resource Management Project ASAL Arid and Semi-Arid Lands ASDS Agricultural Sector Development strategy ASPC Agricultural Sector Programs Steering Committee CAPs Community Action Plans CASPSC County Agriculture Sector Programs Steering Committee CCU Climate Change Unit CDD Community-driven development CIDP County Integrated Development Plan CIG Common Interest Group CMPs County Management Plans CPIA Country Policy and Institutional Assessment CPS Country Partnership Strategy CRA Community Resource Assessment CRPs Climate Risk Profiles CSU County Service Unit CTT County Technical Team CWG Common Working Group DA Designated Account DCU District Coordination Units DDC District Development Committee DSG District Steering Group EA Environmental Assessment ECAAT Eastern and Central Africa Agricultural Transformation Project EFA Economic and Financial Analysis EMF Environmental Management Framework ENSO El Niño Southern Oscillation FEWSNET Famine Early Warning System-Network FM Financial Management FYM Farm yard manure GA Grant Agreement GDP Gross Domestic Product GEF Global Environment Facility GEO Global Environmental Objective GoK Government of Kenya

GRM Grievance Redress Mechanism GRP Grass Roots Partners ICIs Inter-community Investments ICRR Implementation Completion Results Report IDA International Development Association IEG Independent Evaluation Group IFR Interim Financial Report INT Institutional Integrity IP Indigenous People IPMF Integrated Pest Management Framework IPPC Intergovernmental Panel on Climate Change IPPF Indigenous Peoples Planning Framework IPP Indigenous Peoples Plan IRMPF Institutional Risk Management Policy Framework IRR Internal rate of return ISR Implementation Support Report KACCAL Kenya Climate Change Adaptation in the Arid and Semi-Arid Lands KAPAP Kenya Agricultural Productivity and Agri-business Project KAPSLM Kenya Agricultural Productivity and Sustainable Land Management KCSAP Kenya Climate Smart Agriculture Project KENAO Kenya National Audit Office KES Kenyan shillings KS KAPAP Secretariat M&E Monitoring and evaluation MET(s) Mobile Extension Team(s) MIS Management information system MoALF Ministry of Agriculture, Livestock and Fisheries MoU Memorandum of Understanding MS Moderately Satisfactory MTC Ministerial Tender Committee MTR Mid-Term Review MU Moderately Unsatisfactory NARIG National Agricultural Rural Inclusive Growth Project NCCRS National Climate Change Response Strategy NPV Net present value NRM Natural resource management PAD Project Appraisal Document PCU Project Coordination Unit PDO Project Development Project PSDASALs Policy for Sustainable Development of ASALs RPLRP Regional Pastoral Livelihoods Resilience Project RF Results Framework SA Social Analysis SCCF Special Climate Change Fund SP Service Provider TNA Training Needs Assessment

TOC Theory of Change TTL Task Team Leader UNCCD United Nations Convention to Combat Desertification UNFCCC United Nations Framework Convention on Climate Change VMGs Vulnerable and Marginalized Groups WKCDD&FM Western Kenya Community Driven Development and Flood Mitigation

Senior Global Practice Director: Juergen Voegele Practice Manager: Dina Umali-Deininger

Project Team Leader: Joseph Oryokot ICR Author: Irene Bomani

REPUBLIC OF KENYA

Kenya Adaptation to Climate Change in Arid and Semi-Arid Lands Project

CONTENTS

Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph

1. Project Context, Development Objectives and Design ........................................................... 1 2. Key Factors Affecting Implementation and Outcomes ......................................................... 12 3. Assessment of Outcomes ...................................................................................................... 18 4. Assessment of Risk to Development Outcome ..................................................................... 28 5. Assessment of Bank and Borrower Performance .................................................................. 29 6. Lessons Learned .................................................................................................................... 31 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners ....................... 32 Annex 1. Project Costs and Financing ...................................................................................... 33 Annex 2. Outputs by Component .............................................................................................. 34 Annex 3. Economic and Financial Analysis ............................................................................. 42 Annex 4. Bank Lending and Implementation Support/Supervision Processes ......................... 48 Annex 5. Beneficiary Survey Results ....................................................................................... 50 Annex 6. Stakeholder Workshop Report and Results ............................................................... 52 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ................................. 53 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ................................... 57 Annex 9. List of Supporting Documents................................................................................... 58

MAPs

The World Bank (P091979)

vi

A. BASIC INFORMATION

Country: Kenya Project Name:

Kenya: Adaptation to Climate Change in Arid and Semi-Arid Lands (KACCAL)

Project ID: P091979 L/C/TF Number(s): TF-96908

ICR Date: 03/16/2018 ICR Type: Core ICR

Financing Instrument: SIL Borrower: MINISTRY OF FINANCE GOVERNMENT OF KENYA

Original GEF grant amount USD 5.50M Disbursed Amount: USD 5.50M

Revised Amount: USD 5.50M

Environmental Category: B Global Focal Area: C

Implementing Agencies: Ministry of Agriculture, Livestock and Fisheries

Cofinanciers and Other External Partners:

B. KEY DATES Process Date Process Original Date Revised / Actual Date(s)

Concept Review: 10/23/2006 Effectiveness: 11/21/2012 11/15/2012

Appraisal: 09/25/2009 Restructuring(s):

11/29/2012 03/24/2016 10/25/2016 04/18/2017

Approval: 06/10/2010 Mid-term Review:

Closing: 12/31/2014 06/30/2017

C. RATINGS SUMMARY C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory

Risk to Global Environment Outcome Moderate

Bank Performance: Moderately Unsatisfactory

Borrower Performance: Moderately Unsatisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR)


vii

Bank Ratings Borrower Ratings

Quality at Entry: Moderately Unsatisfactory Government: Moderately Unsatisfactory

Quality of Supervision: Moderately Satisfactory Implementing Agency/Agencies: Moderately Satisfactory

Overall Bank Performance: Moderately Unsatisfactory Overall Borrower

Performance: Moderately Unsatisfactory

C.3 Quality at Entry and Implementation Performance Indicators Implementation

Performance Indicators QAG Assessments (if any) Rating

Potential Problem Project at any time (Yes/No): No Quality at Entry (QEA): None

Problem Project at any time (Yes/No): Yes Quality of Supervision

(QSA): None

GEO rating before Closing/Inactive status Moderately Satisfactory

D. SECTOR AND THEME CODES Original Actual

Sector Code (as % of total Bank financing) Agriculture, Fishing and Forestry

Other Agriculture, Fishing and Forestry 40 40

Animal production 20 20

Irrigation and Drainage 20 20

Public Administration

Public administration - Agriculture, fishing and forestry

20 20

Theme Code (as % of total Bank financing) Environment and Natural Resource Management

Climate change 60 60

Mitigation 60 60

Water Resource Management 20 20

Water Institutions, Policies and Reform 20 20

Urban and Rural Development

Rural Development 20 20

Land Administration and Management 20 20

E. BANK STAFF


viii

Positions At ICR At Approval

Vice President: Makhtar Diop Obiageli Katryn Ezekwesili

Country Director: Diarietou Gaye Johannes C.M. Zutt

Practice Manager/Manager: Dina Umali-Deininger Karen Mcconnell Brooks

Project Team Leader: Joseph Oryokot Christine E. Cornelius

ICR Team Leader: Irene Bomani

ICR Primary Author: Irene Bomani

F. RESULTS FRAMEWORK ANALYSIS Global Environment Objectives (GEO) and Key Indicators (as approved) The Project's global environment objective (GEO) is to improve the ability of participating districts and communities in the arid and semi-arid lands to plan and implement climate change adaptation measures. Revised Global Environment Objectives (as approved by original approving authority) and Key Indicators and reasons/justifications The GEO was never revised.

(b) Intermediate Outcome Indicator(s)

Indicator Baseline Value

Original Target Values (from

approval documents)

Formally Revised Target Values

Actual Value Achieved at Completion or Target Years

Indicator 1: Climate risk profiles developed and used for district management plans. Value quantitative or Qualitative)

0

4 15 15

Date achieved 09/25/2009 06/11/2010 10/31/2016 06/30/2017 Comments (incl. % achievement)

"district" was replaced by "county" at project mid-term in 2015 to reflect Kenya's devolution of some of the central government functions to counties. The target was adjusted to 15 and achievement was 100%.

Indicator 2: Climate scenarios developed and adjusted to regional and provincial levels. Value quantitative or Qualitative)

0

1 0

Date achieved 09/25/2009 06/11/2010 06/30/2017 Comments (incl. % achievement)

At mid-term, the indicator was dropped & replaced by "Methodology & tool for screening agricultural investment programs for climate risk developed". The target was fully met, however the tool was developed at project end, and will be used under other projects.

Indicator 3: Mobile extension teams trained/accredited in community climate risk management. Value 0 4 80 497


ix

quantitative or Qualitative)


The indicator was replaced by "Public and private advisory agents trained in community climate risk management" in June 2012. The end target of 80 was exceeded by almost 6 times.

Indicator 4: Percentage of Arid Lands Resource Management Project (ALRMP), sub-projects in the participating districts screened for improving response to climate risk.

Value quantitative or Qualitative)

0

100% 100%


Reference to the "ALRMP" was replaced by "Kenya Agriculture & Agribusiness Project" (KAPAP) after KACCAL was delinked from the former and linked to the latter. Target achievement was 100%.

Indicator 5: Percentage of public and private sector investments rated satisfactory or better by beneficiaries.


0

75% 80% 81%


The original target was adjusted up from 75% to 80%. The end target was exceeded.

Indicator 6: Community Action Plans (CAPs) with concrete climate risk management activities reflected in the budget.


0

32 80 82


Reference to "budget" was replaced by "County Integrated Development Plan (CIDP)" to reflect devolution of government functions. The original end target was revised from 32 to 80. The end target was exceeded.

Indicator 7: Community adaptation micro-projects developed and implemented. Value quantitative or Qualitative)

0

80 156


The final target was exceeded by over 6 times.

Indicator 8: Number of direct project beneficiaries Value 0 40,000 10,000 37,977


x

quantitative or Qualitative)


The original target was reduced in June 2012 under the first restructuring to align with the reduced project timeframe. The revised end target was exceeded three times over.

Indicator 9: Percentage of direct project female beneficiaries. Value quantitative or Qualitative)

0

50 69


The end target was exceeded.

G. RATINGS OF PROJECT PERFORMANCE IN ISRs

No. Date ISR Archived GEO IP Actual Disbursements

(USD millions)

1 03/28/2012 Unsatisfactory Unsatisfactory 0.00

2 08/11/2012 Moderately Satisfactory Moderately Satisfactory

0.00


0.00

4 11/01/2013 Moderately Satisfactory Moderately Unsatisfactory

0.80


0.80


0.93

7 06/25/2015 Moderately Unsatisfactory Moderately Unsatisfactory

1.06


2.47


5.21


5.21


5.21

H. RESTRUCTURING (IF ANY)


xi

I. DISBURSEMENT PROFILE

1

1. Project Context, Development Objectives and Design

1.1 Context at Appraisal 2. At project appraisal, global attention to climate change adaptation was growing, especially following the 2006 Nicholas Stern Review on the Economics of Climate Change and the findings of the Fourth Intergovernmental Panel on Climate Change (IPCC)1. The Stern Review, considered one of the most influential reports on climate change, concluded among others, that climate change could have severe impacts on growth and development; climate change action is required in all countries; and delay in addressing it would be dangerous and very costly. The IPCC fourth assessment report, deemed the largest and most detailed summary of the climate change situation, also raised grave concerns and warnings.

3. Kenya was among the countries considered most at risk from climate change due to increasingly unpredictable weather events (droughts and flooding2). These had been growing in intensity and frequency and were having an extremely damaging impact on the agriculture sector performance, Kenya’s economic backbone and key development driver3. A lack of investments in the agriculture sector adapted to climate risk had been identified and was a becoming a growing challenge for the Government of Kenya’s (GoK) poverty reduction initiatives and economic growth strategies. Among those disproportionally affected by climate change risks and significant adaptation deficit were the 30% of Kenya’s predominantly pastoral\agro-pastoral population, singularly reliant on weather dependent agricultural livelihoods, and living on the severely underdeveloped, and economically and politically marginalized Arid and Semi-Arid Lands (ASALs).

4. The ASALs’ socio-economic well-being was considered critical to Kenya’s long-term sustainable development, poverty reduction and food security and featured prominently in government policies and strategies. Covering more than 80% of the country’s land mass, the Project Appraisal Document (PAD) notes the ASALs at the time supported 75% of the country’s livestock production, agriculture was contributing almost 30% to the national Gross Domestic Product (GDP), of which a quarter of agricultural GDP was from the livestock sector. Over 90% of the wild game that support the tourist industry is in the ASALs, which at the time was bringing in an estimated 50 billion Kenyan shillings (KES) annually. In its National Policy for the Sustainable Development of the Arid and Semi-Arid Lands of Kenya (PSDASAL), the GoK specifically states that “Kenya will not achieve sustained growth in the national economy as long as the ASALs and their enormous resources are not factored into effective national planning and development”. Other policies and strategies articulated the GoK’s growing commitment to climate change adaptation with specific focus on the ASALs, including the: (i) National Climate Change Response Strategy (NCCRS); (ii) the Kenya Strategy for Revitalizing Agriculture (2005), which emphasized the importance of reducing risk and vulnerability for groups whose livelihoods rely natural resources; and (iii) Kenya’s “Vision 2030” launched in 2008 which explicitly called for “increased investments in the ASALs, enhanced disaster preparedness in all disaster-prone areas and capacity strengthening to adapt to climate variability and change”.

1 Recognized as the primary international, intergovernmental forum for negotiating the global response to climate change. 2 The PAD cites the 1997-1998 El Niño cycle which was followed by severe droughts in 2001–02, 2006, and 2009 (page 2). 3 “Flood damage during the 1997–98 El Niño cycle was estimated to have equaled about 11 percent of annual GDP. Droughts between 1998 and 2002 reportedly reduced GDP by 16 percent in the two subsequent years (PAD page 1).

2

5. The Kenya Adaptation to Climate Change in Arid and Semi-Arid Lands (KACCAL) project concept was formulated in response to the critical adaptation deficit and the need for alternative adapted livelihood options in the ASALs. Funded by the Global Environment Fund (GEF) Special Climate Change Fund (SCCF), KACCAL was designed as a four-year US$5.5 million pilot initiative to complement the Arid Lands Resource Management Project II4 (ALRMP), whose objective was “reduced livelihood vulnerability, enhanced food security, and improved access to basic services in 28 drought-prone arid and semi-arid districts in Kenya”. At the time, the ALRMP II was the latest in a series of development operations5, which spanned a nearly 20-year partnership between the International Development Association (IDA6) and the GoK in implementing programs for rural development in the ASALs. KACCAL’s concept represented an evolution of IDA’s support from predominantly drought emergency assistance, to targeted adaptive interventions for climate change adaptation in the ASALs. It would build climate adaptation capacity at national and county level to facilitate better planning and integration of climate change into the national development perspective. It would pilot new approaches for climate change adaptation measures to provide options for alternative adapted livelihood options, which could be replicated and scaled-up within the ASALs and in other parts of the country. Lastly, it would provide lessons and experiences to inform the design of future agriculture climate adaptation operations.

Rationale for Bank assistance

6. The rationale was sound. KACCAL’s focus on filling the adaptive capacity deficit made it a logical follow-up and important addition to the Bank’s evolving portfolio of operations aimed at “climate-smart” development in Kenya. IDA had been gaining a wide range of experiences in addressing climate change related challenges globally and was well positioned to support the GoK. Through KACCAL, GEF incremental support was to contribute to mainstreaming climate risk management into development plans and investment programs and align with IDA’s strategies for scaling up climate change mitigation initiatives. KACCAL’s objective was fully consistent with IDA’s Country Assistance Strategy 2004-2007; the Strategic Framework on Climate Change and Development formulated to guide the scaling up of the integration of climate change planning; the Africa Region’s climate change strategy which focused on capacity building for mainstreaming climate change into development; the fifteenth replenishment of the International Development Association (IDA 15) under which scaling up of climate change initiatives was a key element; and the GEF-4 long-term objective of supporting pilot and demonstration projects for climate change adaptation.

Theory of Change (results chain)

7. The Project design did not include a detailed results chain. A simplified theory of change (TOC) demonstrating the underlying assumptions at appraisal is recreated and is captured below in graphic 1. The TOC’s point of departure is the significant adaptive capacity deficit in the ASALs for planning and implementing climate adaptation measures. In summary, the logic assumes that increasing awareness on climate risk and management, generating and disseminating knowledge and providing tools, and institutional and technical capacity building on climate

4 ALRMP I (Credit 2797) for US$21 million in 1996, ALRMP II (Credit 3795) for US$60 million in 2003, and a supplemental credit for US$60 million in 2006. 5 the two-phase 10-year $140 million the Kenya Agricultural Productivity and Agri-business Program (KAPAK); the Kenya Agricultural Productivity and Sustainable Land Management Project (KAPSLM for US$10 million) Western Kenya Community Driven Development and Flood Mitigation Project (WKCDD FMP) for US$86 million and the Natural Resource Management (NRM) Project. 6 Interchangeably referred to as the World Bank

3

adaptation for government and communities; and bringing together communities, and local and national government officials to actively work together to identify, prioritize and implement adaptation measures, would collectively contribute to achieving the main pillars of the project’s development objective, namely:

Improved climate change integration into national development planning. By providing targeted institutional and technical capacity building on climate risk management and adaptation, and making available climate knowledge products to relevant national level government officials, the project would contribute to improved capacity for integrating climate action into ASAL development strategies and programs.

Improved county level capacity for climate resilient planning. Through technical assistance (TA) for assessing climate risks in ASALs investments, and climate adaptation capacity building for county officials, the project would strengthen capacity to integrate climate risk management into county planning processes.

Improved capacity to implement climate adaptation measures. By financing feasibility studies and civil works equipment for county level investments, and climate related training for community members and public and private service providers, the project would help improve implementation of adaptation measures.

Livelihoods adapted to climate and non-climate factors. Raising community awareness of climate risks and climate forecasts, providing extensive capacity building, and demonstrating adaptation technologies and practices, was expected to lead to increased community awareness on climate risks, improved land and water management, and uptake of climate resilient technologies and agro-pastoral practices.

4

Information available for ASAL strategies

dialogues at the national and district

levels

1.1 Developing climate - related knowledge

products: TA, training, software, data work,

partnerships for advisory services

1.2 Capacity building ALRMP/ ASAL on climate change: training, study tours, TA

Capacity for integrating climate

action into ASAL development strategies

and programs at national level improved

2.1 TA for risk in ALRMP investments,

equipment and software for climate knowledge products

Capacity to integrate climate risk management

into district planning process

improved

2.2 Feasibility studies for

community level public and private

investments

Civil works and equipment for

district level investments

2.2 Training for community/

private investors

3.1 Raising awareness of targeted

communities about climate risks, climate

forecasts: training

Awareness and advocacy on

climate risks within communities

improved

3.2 Matching grants for community micro -

projects

Land and water

managem ent

improved

Climate change resilient

agropastoral practices adopted

Livelihoods adapted to climate and non - climate factors

Climate - resilient planning and implementation capacity improved at county level

Climate change integration in national

development plans improved

5

1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved)

8. As stated in the PAD and Grant Agreement (GA), the Project Development Objective (PDO) was “to improve the ability of participating districts and communities in the Arid and Semi-Arid Lands to plan and implement climate change adaptation measures”. The expected outcomes were: (a) improved climate change integration into national development planning; (b) improved capacity to plan climate change measures; (c) improved capacity to implement climate adaptation measures; and (d) livelihoods adapted to climate and non-climate factors. Achievement of the PDO was to be measured through 2 PDO and 8 intermediate indicators as detailed below in table 1.

Table 1 Original Results Framework (from approved PAD, pp 41)

Project Development Objective (PDO) Project Outcome Indicators The PDO is to improve the ability of participating districts and communities in the arid and semi-arid lands to plan and implement climate change adaptation measures.

Number of District management plans with concrete climate risk management activities reflected in the budget. Percentage of community adaptation projects rated satisfactory or better by participating communities (communities assess whether outcomes have been achieved)

Intermediate Outcomes Intermediate Outcome Indicators

Increased understanding among national and regional stakeholders of issues related to climate change.

Improved availability of climate risk information at national and regional level.

Climate risk profiles developed and used for district management plans.

Climate scenarios developed and adjusted to regional and provincial levels.

Increased understanding among local stakeholders of climate-related issues.

Improved availability of climate risk information at district and local level.

Mobile extension teams trained/accredited in community climate risk management.

Percentage of Arid Lands Resource Management Project, sub-projects in the Participating Districts screened for improving response to climate risk.

Public and private sector investments rated satisfactory or better by beneficiaries (%) (beneficiaries assess whether outcomes have been achieved).

Enhanced ability of communities to plan, manage, and implement climate-related activities.

Community Action Plans with concrete climate risk management activities reflected in the budget.

Community adaptation projects developed and implemented.

Number of direct beneficiaries (of which % are female). Source: PAD

6

1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 9. There was no change to the PDO. However, per recommendation of the mid-term review (MTR) conducted in June 2015, one PDO and 3 intermediate results indicators, and 3 targets were justifiably modified, and 2 intermediate results indicators were dropped as summarized below and in the Results Framework (RF) to reflect changes in project scope and institutional arrangements (see table 2). 10. The PDO indicator “number of districts management plans with concrete climate risk management activities reflected in the budget” was changed to “the number of county management plans with concrete climate risk management activities reflected in the County Integrated Development Plans (CIDPs)”. 11. Five intermediate indicators were modified as summarized below and the targets of 3 indicators were adjusted as shown in table 2. “Climate risk profiles (CRPs) developed and used for district management plans” was changed to “climate risk profiles developed and used for county management plans”.

“Climate scenarios developed and adjusted to regional and provincial levels” was replaced by “Methodology and tool for screening agricultural investment programs for climate risk developed”.

“Mobile Extension Teams (METs) trained/accredited in community climate risk management” was replaced by “Public and private advisory agents trained in community climate risk management”.

“Percentage of Arid Lands Resource Management Project, sub-projects in the Participating Districts screened for improving response to climate risk” was modified to “Percentage of Kenya Agricultural Productivity and Agri-business Project, sub-projects in the participating districts screened for improving response to climate risk”.

“Community action plans with concrete climate risk management activities reflected in the budget was changed to “community action plans with concrete climate risk management activities reflected in the County Integrated Development Plan”.

Table 2: Revised Results Framework

PDO: To improve the ability of participating districts and communities in the arid and semi-arid lands to plan and implement climate change adaptation measures

PDO Level

Original Indicator

Original Target

No Change/Dropped/Revised Indicator

New Indicator

Revised Target

Achieved Target

1. Number of District management plans with concrete climate risk management activities reflected in the budget

4 Number of county management plans with concrete climate risk management activities reflected in the County Integrated Development Plan

N/A 15 15

7

2. Percentage of Community adaptation projects rated satisfactory or better by participating communities

80% No change N/A No change

81%

Inter-mediate

Level

1. Climate risk profiles developed and used for district management plans (number)

4 Climate risk profiles developed and used for county management plans”

N/A 15 15

2. Climate scenarios developed and adjusted to regional and provincial levels (number)

1 Dropped Methodology & tool for screening agricultural investment programs for climate risk

No change

1

3. Mobile Extension Teams (METs) trained/accredited in community climate risk management (number)

4 Dropped Public and private advisory agents trained in community climate risk management

80 4977

4. Percentage of Arid Lands Resource Management Project, sub-projects in the Participating Districts screened for improving response to climate risk

100% Percentage of Kenya Agricultural Productivity and Agri-business Project, sub-projects in the participating counties screened for improving response to climate risk

N/A No change

100%

5. Public and private sector investments rated satisfactory or better by beneficiaries (%)

75% No change N/A No change

81

7 Per data from the GoK’s end of project Impact Evaluation Study. This is an update to the figure of 358 in the last ISR dated June 2017.

8

6. Community

Action Plans with concrete climate risk management activities reflected in the budget (number)

80 Community action plans with concrete climate risk management activities reflected in the CIDP.

N/A No change

82

7. Community adaptation micro-projects developed and implemented (number)

80 No change N/A No change

156

8. Number of direct beneficiaries (of which 50% are female)

40,000 No change N/A 10,000 37,9778

1.4 Main Beneficiaries 12. KACCAL’s primary target beneficiaries were agro-pastoralist and pastoralist communities, including vulnerable and marginalized groups (VMGs) and Indigenous People (IPs) (VMGs/IPs9), in Garissa, Tana River, West Pokot and Kilifi.

1.5 Original Components (as approved) 13. KACCAL had three components. (i) Component 1. Climate Information Products, Policy, and Advocacy (US$1.46 million). The component was designed to strengthen capacity and institutional coordination among national agencies managing disasters, to better assess and respond to climate risks. It had two sub-components: Sub-component 1.1. Development of climate-related knowledge products to inform climate risk management strategies in the ASALs; and Sub-component 1.2. Integration of climate action into ASAL development strategies and programs. Among others, the main activities comprised development of county management plans (CMPs), county climate risk profiles (CRPs), community action plans (CAPs), and tool kit for screening agricultural investment programs for climate risk. (ii) Component 2. Climate Risk Management at the County Level (US$1.37 million). The component aimed to integrate climate risk management into county planning processes and programs through 2 subcomponents: Sub-component 2.1. Capacity building to integrate climate risk management into county planning processes and Sub-component 2.2. Support for “climate-smart” public and private investments to implement selected public and private sector interventions identified in county plans. Key activities included the construction of 14 mega intercommunity investments (ICIs)10 designed to increase access to better quality water for

8 The last ISR in the system was done in March 2017 and does not reflect the final number of direct beneficiaries. 9 VMGs included unemployed youth, Indigenous Peoples (IP), elderly women and men, widows/orphans, the differently-abled, recovering substance abusers, and people living with HIV/AIDS. 10 i.e. 13 water pans (5 in West Pokot, 3 each in Garissa and Kilifi, and 2 in Tana River) and 1 camel milk value addition processing plant in Bura, Tana River County. Project funds covered the purchase of demonstration materials, excavation for the ICIs and equipment for the milk processing plant.

9

domestic and livestock use, and to provide a centralized outlet for value addition for camel milk production. Each water pan included construction of a de-silting trap, water pump (solar operated or electricity operated), water tank, watering points for livestock and domestic use, fencing to keep away both livestock and intruders, capacity building to manage the ICIs, pit latrines and bathrooms. The milk value addition investment included a building, equipment for milk value addition and processing, and capacity building to operate and manage the facility. (iii) Component 3. Community-driven Initiatives for Climate Resilience (US$ 2.67 million) was designed to help beneficiary communities adopt climate change adaptation strategies and investments to reduce climate related vulnerabilities and strengthen their resilience to climate risk primarily through adoption of adaptation technologies. It had two subcomponents. Subcomponent 3.1: Support for community capacity building; and Subcomponent 3.2: Support for community-based micro-projects. Priority activities included among others, micro-projects11 to disseminate adaptation information through trainings and demonstrations of both traditional and new innovative approaches12 to climate change adaptation. Demonstrations covered various enterprises in three major value chains13 selected by the beneficiaries: (i) Dairy; (ii) Fruit/Nuts/Vegetables; (iii) Meat; and (iv) Natural Resource Management (NRM). Beneficiary farmers were then expected to use personal funds to uptake their preferred technologies and\or practices on their individual farms.

1.6 Revised Components 14. The project maintained the 3 components and overall design. However, component costs were adjusted during the first and second project restructurings. At the MTR, changes were made under the sub-components, to improve project performance and accelerate progress towards achievement of the development objective. 15. Component 1: Climate information products, policy, and advocacy. The number of county CRPs, an intermediate results indicator was increased from 4 to 15 and is reflected in the revised RF. The 11 additional CRPs would cover counties where a new mega operation, the US$250 million Kenya Climate Smart Agriculture Project (KCSAP), under preparation at the time, would be implemented. Knowledge products were revamped to avoid replication of other similar programs initiated and/or completed at the time. It was determined that preparation of climate change scenarios for Kenya, also an intermediate results indicator, could not be completed by project closing and was dropped. The RF was adjusted accordingly. Improvement of the Early Warning System was shifted to the KCSAP. Preparation of a methodology and tool for screening agricultural investment programs for climate risk was added as a new activity and included as an intermediate results indicator. The RF was corrected accordingly. Under the first restructuring, the component cost was adjusted from US$1.46 to US$1.50 million. Under the second restructuring the cost went up to US$2.25 million primarily because of the additional CRPs.

11 Project funding covered the cost of extension service provision by service providers to farmers and the purchase of demonstration materials. 12 establishment of conservation of pasture and fodder, water harvesting technologies, efficient use of harvested water, superior crop varieties, improved livestock breeds, integrated pest and disease management and control, drought resistant crops, value addition, integrated soil fertility management and improved farm structures. 13 Key determining factor for the selected value chains was the contribution of these value chains to the livelihoods of farmers and economic development of the county.

10

16. Component 2: Climate risk management at the district level. To align with the remaining implementation timeframe, the project reduced the planned number of inter-community investments (ICIs), increased the geographical coverage, and agreed to construct a total of 14 ICIs. Thirteen 13 water pans would address the severe water scarcity and access challenges for both domestic and livestock use in the project areas, factors identified as the source of serious conflicts and a critical constraint to livelihoods in the ASALs. One value addition facility for camel milk would support the dairy industry in Tana River county. In addition, the Project revised the main recipients of the project’s institutional capacity building support to reflect new implementation arrangements, and participating institutions and stakeholders14. This resulted in the addition of a new intermediate results indicator, “Public and private advisory agents trained in community climate risk management” and target of “80”. The original component cost went from US$1.37 million to US$1.90 million under the first restructuring and then to US$2.00 million under the second restructuring. 17. Component 3: Community-driven initiatives for climate resilience. The initial scale of the community adaptation micro-projects, an intermediate results indicator, was reduced and resulted in a change in the overall component cost. However, no changes were made to the end target. In addition, after findings from the World Bank forensic audit of the ALRMP II and other operations with similar CDD approaches (Western Kenya Community Development Project), stronger fiduciary and accountability measures to minimize fiduciary risks were introduced. The original component cost went from US$2.67 million to US$2.10 million under the first restructuring and then to US$1.25 million under the second restructuring.

1.7 Other significant changes 18. The project was affected by four major events, which resulted in unexpected challenges and serious delays during preparation and implementation of, at a minimum, 1 year per event:

(i) political instability following the highly contested 2007 presidential elections, significantly delayed KACCAL’s design and project launch phase (discussed further in section 2).

(ii) the launch in January 2009 of an Institutional Integrity (INT) forensic audit of the ALRMP II, KACCAL’s parent project and its preliminary findings, led to the informal suspension of the ALRMP II in July 2010, before KACCAL’s effectiveness. The ALRMP II closed on December 21, 2010. With effectiveness suspended and no IDA parent project (a GEF requirement), KACCAL came to a complete standstill until the INT investigation was completed in June 2011. To move KACCAL forward, required a new parent IDA project. This led to the first project restructuring detailed below.

(iii) enactment of the 2010 Kenyan Constitution introduced significant institutional changes (e.g. elimination of the Ministry of State for the Development of Northern Kenya and Other Arid Lands, the ministry responsible for KACCAL), and a decentralization proclamation, which among others required the creation of a new tier of government at county level and the devolution of resources and some functions to counties.

14 (KAPAP PCU, MoALF departments including its Climate Change Unit [CCU]), other agriculture sector ministries and departments, and the Agricultural Sector Programs Steering Committee (ASPC);

11

(iv) the 2013 presidential election marked the official launch of the devolution process and brought with it a host of challenges and further delays as new county governments began the job of navigating the new institutional maze.

19. Restructuring. The project had two full level 2 restructurings and 2 short no-cost extensions as detailed below. The changes were necessary first to move the project forward after the stalled effectiveness in 2010; and second to significantly improve and accelerate performance, and allow more time to ensure completion of key activities and achievement of the PDO, following long delays in project start-up and implementation interruptions. Budget reallocations were essential to accommodate adjustments to key deliverables, among them the additional 11 CRPs for the KCSAP (noted earlier in paragraph 15) and methodologies and tools for screening investments for climate risk. Both were critical for the KCSAP. The CRPs were expected to inform preparation of the KCSAP and serve as the basis for determining the financing for its community investments. Having them available in advance helped fast track the launch of the now approved KCSAP. The changes had no impact on the original results chain. June 29, 2012: The first restructuring enabled project effectiveness and launch. It

comprised: (a) identification of a new IDA parent project, the Kenya Agricultural Productivity and Agri-business Project (KAPAP-P109683); (b) a change in the project’s geographic location to align with KAPAP; (c) new institutional arrangements, including the transfer of KACCAL from the then Ministry of State for the Development of Northern Kenya and Other Arid Lands to the Ministry of Agriculture, Livestock and Fisheries (MoALF); (d) reassignment of the coordination and implementation management to the Kenya Agricultural Productivity Program (KAPP) Secretariat (KS)15; (e) changes to financial management and procurement arrangements; (f) reallocation of grant proceeds; and (g) an extension of the closing date from December 2014 to October 31, 2016.

March 24, 2016: The second restructuring put into effect the recommendations of the MTR (discussed further in section 2) conducted in June 2015. The aim was to turn project performance around and accelerate implementation and fund use. Key modifications included changes to the RF and sub-component activities (as discussed above in section 1), reallocation of funds, and further adjustments to the financial and procurement management arrangements.

October 2016: The third restructuring was a six months no-cost extension of the closing

date from October 30, 2016 to April 28, 2017. The extension allowed completion of the methodologies and tools for screening investments for climate risk and the 11 county CRPs that had been significantly delayed by an extremely slow flow of funds from the National Treasury to the project account. The extension also allowed the continuation of technology demonstration activities.

April 2017: Kenya declared drought a national disaster in February 2017 and appealed

for local and international help. In response the Bank approved a two months no-cost extension of the closing date from April 28th, 2017 to June 30th, 2017. The extension was granted to allow completion of the water pan in West Pokot, one of the hardest hit counties by the drought, expand the water pans to increase their capacity, procure and

15 a project coordination unit responsible for the implementation of the Kenya Agricultural Productivity and Agribusiness Project (KAPAP) and KAPSLM.

12

distribute seeds of food staples and planting material for post-drought planting and install additional water harvesting facilities (e.g. boreholes).

2. Key Factors Affecting Implementation and Outcomes

2.1 Project Preparation, Design and Quality at Entry 20. Project design and preparation took place during a period of major political and institutional changes in Kenya, which significantly extended both the design and preparation phase. The project concept was approved in October 2006. The project was appraised in September 2009 and approved in June 2010. This is a three-year gap between the concept and appraisal stage, and an eight-month interval between appraisal and approval. 21. KACCAL’s design clearly reflected both IDA’s and the GoK’s sectoral development strategies and priorities for climate change adaptation. As a pilot project accompanying the ALRMP II, KACCAL’s design drew on ALRMP II’s background analysis, including the sector and policy context, poverty background, and specific ASAL development challenges. The rationale for the project was sound and well-articulated. The development objective was clear, realistic and achievable. The components and proposed activities were aligned with the project objective and would contribute to its achievement. Selection of the project’s three categories of beneficiaries (ASAL communities; county and national government institutions; and private sector stakeholders) was appropriate. The design team considered the option of a stand-alone larger operation. It ultimately opted for piloting on a smaller scale, specialized climate adaptation activities using existing implementation mechanisms (e.g. private and public partnerships), to fast track service delivery to farmers. The team recognized the importance of beneficiary participation to project success, effectiveness and sustainability, and selected the already well established CDD approach to promote relatively simple adoption of adaptation measures, which if successful would be scaled up in the region. Lessons learned from implementation of the ALRMP I and II, and other similar projects in Kenya, and best practices from other IDA operations implemented in South Asia and Latin America were incorporated. Critical risks were identified and mitigation measures put in place. Collectively these elements provided a sound basis for the KACCAL design. 22. The design team’s decision to use the ALRMP II’s existing implementation infrastructure16 to avoid duplicity and reduce operational costs was a reasonable cost saving measure despite associated unexpected challenges. The unexpected suspension and eventual closure of the ALRMP II, and the subsequent linking of KACCAL to KAPAP significantly impacted KACCAL’s human resource needs and other logistical requirements at project launch, a very critical point. Unlike the ALRMP II, KAPAP was initially not technically staffed to manage a climate change focused operation. A protracted recruitment process to acquire the necessary climate change expertise (discussed later in section 2.2), was among the key contributing factors to serious project implementation delays. Establishing a new stand-alone implementing agency would not have been an allowable or better alternative. First, per GEF requirements the grant had to be attached to an IDA operation. Second, notwithstanding the associated costs and steep learning curve, a new implementing agency would have faced the same climate change expertise limitations and recruitment challenges.

16 institutional, fiduciary, monitoring and evaluation (M&E), environmental and social safeguards arrangements and human resources.

13

23. The project design was overly ambitious in proposing a 4-year project lifespan for a project geared towards enhancing “resilience”. There should have been greater awareness of the inherent “teething” challenges, which take valuable time away from actual work on the ground, especially at project start. Greater consideration for a longer timeframe for actual implementation of activities on the ground would have been ideal. Assessment of risk and mitigation measures 24. Overall, the identified key project risks and mitigation measures reflected the project’s main concern areas. For the most part, many of the risks were minimized with the proposed mitigation measures. However, the team’s evaluation of the impacts of a few of the risks (fiduciary risks) were underestimated, and one (frequent institutional changes) was unforeseen and not included in the risk matrix. 25. Given the known persistent competition for resources and spillover from associated conflicts in neighboring counties, the risk from “continued and growing conflict, especially in arid districts” rated as “moderate”, was under estimated. The proposed conflict management measures and interventions intended to decrease vulnerability and therefore conflict, were appropriate mitigation measures. However, in view of the reported incidences of clashes between project and non-project community members, especially during the 2016\2017 drought, suggests the measures had not been adequately institutionalized at the county level. Given the team’s awareness of the high probability of the recurrence and deepening of conflict, especially in the face of relentless weather variability, deemed a given for the ASALs, a rating of “substantial” would have been a more realistic rating. 26. The financial management (FM) risk was rightly rated “substantial” due to strong concerns about Kenya’s overall weak governance environment, weak judiciary, corruption and the 2008 post-election crisis identified in the Country Policy and Institutional Assessment (CPIA). The internal controls established as a risk management measure for ALRMP II were to also be applied to KACCAL. However, they were not effective in mitigating some of the risks under ALRMP II, which led to its downfall. The silver lining for KACCAL in this instance was that there was enough time between the closure of the ALRMP II and effectiveness of KACCAL, to identify and implement additional measures to strengthen KACCAL’s internal controls. This is reported to have ultimately helped KACCAL avoid the ALRMP II FM pitfalls. 27. The team underestimated the impact of the fiduciary and accountability issues flagged during the implementation of the parent ALRMP II. The team did not anticipate the risk of a potential suspension of the ARLMP II and the related impacts. The ensuing issues such as the protracted ALRMP INT process, the long deferment of the KACCAL effectiveness and launch pending completion of the INT investigation, and delays to link KACCAL to another IDA project, were major contributors to the KACCAL implementation delays and challenges. They would have likely elevated the overall project risk from “moderate” to “substantial”. 28. The frequent institutional changes, political turmoil and constitutional reform process on the side of the GoK were also unforeseen but beyond the control of the World Bank. They would have likely elevated the coordination risk to “high”.

2.2 Implementation 29. Implementation was affected by negative (unforeseen and some beyond the control of the World Bank team) and positive factors.

14

30. Frequent GoK institutional and political changes adversely impacted implementation. There was a two-and-a-half-year gap between project approval and effectiveness, during which time Kenya was recovering from the aftermath of the 2007 elections. Furthermore, the project was in limbo because of the ongoing forensic audit of the ALRMP II, whose preliminary findings had led to the ALRMP suspension. The project became effective in November 2012, however its launch then coincided with implementation of the devolution process, i.e. introduction of new institutions and implementing arrangements (new ministry, new implementing agency and establishment of county level implementation arrangements). Implementation was further delayed by the long KACCAL\KAPAP alignment process and recruitment delays for climate change experts both at the national and county levels. The recruitment process took almost 2 years because of a shortage of qualified climate change experts. Expected to be on board by December 31st, 2012, the specialists were recruited in January and September 2014, respectively. This resulted in the project being flagged as a “problem project” for several supervision cycles.

31. Significant delays in flow of funds. Complicated financial flows and heavy bureaucracy at the National Treasury resulted in a persistently slow and debilitating flow of project funds from the designated account (DA), which it managed, to the project account managed by MoALF. The delays ranged between 3-8 months17. Despite continual follow up by the World Bank (management and implementing team), the problem continued throughout implementation. The resultant non-alignment between fund receipt, and annual workplans and budgets contributed to extensive implementation and procurement delays, and disrupted implementation sequencing of critical project activities. For instance, planting materials for demonstrations on tree planting, tree seedlings and grasses were distributed late and off season, which unfortunately coincided, with the 2016 drought. Key outputs such as the CRPs, a precursor to preparation of the CIDPs and Community Action Plans (CAPs), and the methodology and tool kit for screening investments, were prepared out of sync and completed close to project closure. As a result, they could not be used for KACCAL as initially planned. However, as intended, they are already being used under the now ongoing KCSAP and Regional Pastoral Livelihoods Resilience Project (RPLRP) to guide identification of climate smart investments, key value chain commodities and the most problematic climate hazard priorities. They are also expected to be used for the same purpose, under the proposed Eastern and Central Africa Agriculture Transformation Project (ECAAT) currently under preparation, as well as future operations. 32. Lengthy procurement processes. Protracted procurement processes compounded by the flow of funds delays impacted the timely delivery of some key project outputs. Recommendations to identify and remove procurement bottlenecks were shared with the GoK’s Ministerial Tender Committee (MTC), but the issue, beyond the control of IDA, remained an ongoing challenge till project completion. 33. Security challenges. Clashes between counties\communities due to land and water use disputes reportedly occurred in varying degrees throughout implementation. Incidences of cattle rustling, produce theft, and water and pasture related conflicts between project and non-project areas are reported to have been exacerbated by the 2016\17 severe drought. In some of the project areas, security issues, including those related to the 2017 elections (both pre- and post) delayed the collection of M&E data, delivery of capacity building services, and support to project beneficiaries. On several occasions, IDA support missions were suspended in certain areas. For instance, the ICR final field visits were not possible due to the post-election security issues.

17 KACCAL End of Project Impact Evaluation Study, June 2017.

15

34. Expansive project areas and distances. As reported by the KAPAP Secretariat, the vastness of the target counties, the considerable distance among them, weak infrastructure and a poor communication infrastructure made overall coordination more challenging in some areas. 35. Compressed implementation timeframe. Because of the long delays during preparation and project launch, actual implementation on the ground was compressed into two and a half instead of the original 4 years. 36. Prolonged Drought affected uptake of some of the adaptation technologies and other activities. Drought in some project areas hindered farmer uptake of certain technologies such as drip irrigation, planting of fodder trees, tree seedlings, and grass. Training attendance in some of the areas hardest hit by the drought was lower as attendees focused on other activities they deemed more urgent. 37. The MTR was a turning point for KACCAL and led to improved project performance. At mid-term, KACCAL was rated “moderately unsatisfactory”, fund disbursement from the World Bank to the GoK’s DA was at 20% and IDA management was in favor of closing the project. Initially planned for November 201418, the MTR was eventually conducted in June 2015, given the project start-up delays. The IDA implementing team provided a compelling argument to restructure the project. The MTR provided critical guidance and set the stage for a remarkable turnaround in the project. With the appropriate counter-part teams in place on the ground, implementation of an action plan based on the MTR recommendations and close support from the IDA implementing team, project performance turned around. Within a year after the MTR, the performance rating went from “moderately unsatisfactory” to “moderately satisfactory”. By end December 2016, the remaining grant funds had been fully disbursed to the GoK’s. 38. Multiple factors helped improve project performance. The presence of the IDA task team leadership and core fiduciary staff in the country office allowed for real-time support and quicker resolution of problems because of their physical proximity to the Gok counterparts and beneficiaries. On the GoK side, the enabling elements included use of: (i) an existing project implementation and oversight infrastructure19; (ii) reliable and successful implementation mechanisms such as the already tested CDD approach and contracted service delivery model, albeit not without some obstacles20; (iii) existence of implementation guidance documents such as the Community Grants Manual (CGM) and Project Implementation Plan (PIP), which only required retro-fitting to the KACCAL project; and (iv) dedicated staff, who albeit overloaded given the additional responsibilities from coordinating KAPAP and KAPSLM, managed, despite the challenges, to effectively work well with the county stakeholders to ultimately achieve the expected outcomes. On the beneficiary side, the active involvement of the target communities in different project aspects (e.g. management, supervision, monitoring of the ICIs, and in-kind and monetary contributions) also facilitated implementation and laid the groundwork for future sustainability of the investments.

18 based on the November 2012 effectiveness 19 County Technical Teams (CTT), community or micro-catchment oversight committees, Service Provider consortiums (from public and private sector). 20 Lack of experienced Grass Roots Partners (GRPs) to provide climate change and risk adaptation advisories in some of the project areas. Instances of poor service delivery to some beneficiaries, including failure by Principle Partners to ensure timely payment of GRPs.

16

2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 39. The overall quality of M&E is rated “modest”. 40. Design. KACCAL used the KAPAP M&E system which drew on data from three levels; national, county and community. The design had two positive features. For data collection and reporting, the system relied on a combination of beneficiary participation to record and collect data on their activities at community level, consolidation of county level data by county M&E officers, and final overall data consolidation at the national level by the KS M&E officer. It also allowed for gender disaggregation of data. KACCAL’s results framework was straightforward with a manageable number of PDO and intermediate results indicators (2 and 8 respectively), with baselines and realistic targets. However most of the indicators were output oriented, therefore, there was no way to measure benefits through the indicators. In addition, at the PDO level, as one of only two indicators, the indicator measuring beneficiary “satisfaction levels” for the community adaptation projects was subjective. An indicator to reflect adaptation rates of technologies or practices on individual farms would have incorporated an aspect of objectivity. In addition, including an indicator or two to measure benefits and\or impacts would have provided a more robust RF. 41. Implementation and Utilization. By project end, data for all the results indicators was fully recorded in the RF. All indicator targets were met or surpassed. Implementation and use of the M&E system had some deficiencies. There were delays with collecting and moving data from the beneficiaries to the CSUs, and then to the KS, which caused delays in updating of the MIS. At times, the delays also impacted the performance and delivery of services by the SPs, as their payments were dependent on timely data collection and MIS entries. Another key concern noted in project documentation was that the M&E data was not used to inform decision making and or to facilitate corrective measures to improve project performance. 42. There were several oversights when changes to the RF were made at mid-term. First, the new indicator “Public and private advisory agents trained in community climate risk management” reflected the expanded definition of recipients of the capacity building initiatives. A target of “80” was set. However, based on the final achieved target of “497”, the target was grossly underestimated. Second, the overall amount of funds allocated for the adaptation micro-projects was reduced, as was the size of individual micro projects. However, the original target of “80” for the indicator “Community adaptation micro-projects developed and implemented” was never adjusted. Therefore, the “156” micro-projects implemented by project end, exceeded the target almost twice. Third, the original direct beneficiary target of “40,000” was reduced to “10,000” given the remaining timeframe at MTR. While the adjustment was prudent, the reduction may have been much lower than necessary. Finally, there was a missed opportunity to include indicators to measure benefits and\or impact. 43. An end of project impact evaluation was conducted in June 2017. It identifies both the project’s short-comings and strengths, and concludes with an overall satisfactory rating of the project and its achievements. The assessment’s weakness is that it does not provide control data. The ICR fills in this gap using statistics from Kenya’s 2017 Economic Review of Agriculture and the MoALF’s 2018 State Department of Livestock Statistics.

2.4 Safeguards and Fiduciary Compliance 44. Environmental compliance. KACCAL was classified as a ‘Category B’ project and originally triggered four safeguards policies; OP 4.01 (Environmental Assessment), OP 4.04

17

(Natural Habitats), OP 4.09 (Pest Management), and OP 4.10 (Indigenous People). The relinking of KACCAL to KAPAP however resulted in the un-triggering of OP 4.04 on Natural Habitats, as the KACCAL activities would be implemented only on existing farmland. Furthermore, the project would not fund activities in protected areas. 45. Compliance with IDA’s environmental and social safeguards requirements was 100%. All the required environmental and social safeguards instruments (Environmental and Social Management Framework (ESMF), the Integrated Pest Management Framework (IPFM), and Indigenous People’s Planning Framework (IPPF) were prepared, disclosed and applied during implementation. All 156 micro-projects were screened in line with the ESMF. Environmental Impact Assessment Studies (ESIAs) were conducted for the 14 ICIs. A broad range of stakeholders including the CTTs, SPs, county government leadership and beneficiaries were trained in environmental and social safeguards before the implementation of the sub projects. 46. Social safeguards compliance. KACCAL complied with the requirements of OP 4.10 Indigenous Peoples Planning Framework (IPPF), Indigenous Peoples Plans (IPP) and applicable country requirements related to the vulnerable and marginalized groups (VMGs/IPs) (widows, youth and IPs). An in depth social assessment helped identify the VMGs/IPs for inclusion into project activities. As a result, ten Common Interest Groups (CIGs) covering three IPs were identified and supported by the project.

Table 3: Project supported CIGs of Indigenous People

County IP Number of CIGs

Value Chain IP ICIs

West Pokot Sengwer 2 Irish Potatoes Sengwer - Kilifi Watha 2 Apiculture Watha Bibithole water

pan Tana River

Wailuana 3 Apiculture Wailuana -

Wailuana 3 Mango Wailuana - Source: PAD 47. Targeted support to the IPs included assistance for the development of funding proposals, awareness training on social inclusivity and peaceful coexistence, and capacity building on peace building, conflict resolution, constitution-writing, group dynamics, and leadership skills. End of project documentation does however highlight the need for further training in all these issues, in the Tana River area. 48. A grievance redress mechanism (GRM) was established and functional throughout implementation. A complaint lodged by the Sengwer Cultural Center representing an IP group not targeted by the project was submitted to the Bank’s grievance mechanism but was quickly resolved21 to the satisfaction of the complainants. A resourceful element of the GRM was the use of county level implementation committees as channels for conflict resolution and amicable settlement of disagreements. Each CWG had a complaint/conflict handling subcommittee. The complaints were documented at the CSU level in a complaints register by the M&E officer, and

21 dissenting sentiments by non-project targeted groups is allowed

18

reported on in a monthly progress report. While it reportedly took a bit of time and concerted effort from the safeguards team to help communities fully understand and internalize the mechanism, leadership and governance, and gender and resource equity aspects, records of the settled disputes indicate the efforts paid off. After project closure, management of the mechanism was handed over to County governments for continuity and to ensure its institutionalization into the county structures. Fiduciary Compliance

49. Financial Management (FM) performance was rated “moderately satisfactory”. Recommended mitigation measures for weaknesses identified by the FM review at appraisal were implemented. Additional measures to strengthen internal controls were also applied following the results of the forensic audit findings referred to earlier in paragraph 1.3. KACCAL fully complied with the financial reporting requirements. Quarterly Interim Financial Reports (IFRs) and annual audit reports prepared by the Kenya National Audit Office (KENAO) were submitted to the World Bank within the stipulated timelines. There are no records of qualified audit reports or fraud in the official files. One critical drawback was the impact of the GoK’s internal bureaucracy on the flow of funds arrangements. Despite what were considered relatively simple flow of funds procedures, the woefully slow flow of funds from the Treasury to the Project’s Designated Account (DA) remained a significant challenge throughout implementation and was one of the most egregious contributors to implementation delays. 50. Procurement performance was rated “Moderately Satisfactory” throughout most of the implementation period, with significant delays in the procurement process at critical times in implementation as the main reason for the rating. At the national level, major bottlenecks were predominantly within the Ministerial Tender Committee (MTC). At county level, procurement of goods and services for the micro-projects carried out by the beneficiary communities was affected considerably by the slow flow funds.

2.5 Post-completion Operation/Next Phase 51. There is no plan for a second phase of the KACCAL project but the GoK’s commitment to climate change adaptation and demand for adaptation programs remains strong. There will be continued attention to climate change adaptation in Kenya’s agriculture sector through other IDA supported operations, among them the ongoing KCSAP, the National Agricultural Rural Inclusive Growth Project (NARIG), RPLRP and the proposed ECAAT project. These operations can, on a larger scale help consolidate and expand the KACCAL impacts. Ongoing and future use of some of KACCAL’s key outputs (climate change knowledge products, climate planning and budgeting tools, climate risk assessment tool), lessons learned and successful approaches stand to have a significant impact under Kenya’s climate smart agriculture portfolio.

3. Assessment of Outcomes

3.1 Relevance of Objectives, Design and Implementation Overall Rating: High

52. KACCAL’s objective remains relevant to the GoK’s development priorities as expressed in some of its key strategies (National Policy for Sustainable Development of ASALs [PSDASALs] and Vision 2030 among others). It is also consistent with World Bank’s Kenya Country Partnership Strategy (CPS) for FY14–18 that was current when project implementation

19

was completed. The design and implementation approach remain relevant to address ongoing climate variability impacts and related issues of planning and implementation of adaptation measures. Continuous climate adaptation related capacity building at national and county level, use of climate profiles, climate knowledge products, and practical climate planning and budgeting tools to underpin adaptation actions, involving communities to participate in and take ownership for adaptation, and use of both public and private service providers to expand coverage have all proved to be key elements of an effective approach to climate adaptation and mitigation. They were used in past projects (KAPAP and KAPSLIM). They are currently being replicated under the ongoing KCSAP, RPLRP, and NARIG, and are also expected to be incorporated into the proposed ECAAT to address similar issues. Furthermore, KACCAL results are expected to be consolidated and scaled up and out to benefit the rest of the ASALs under the ongoing and possible future projects.

3.2 Achievement of Project Development Objectives Overall Rating: Substantial 53. The project substantially achieved its objective as measured by the expected outcomes and results indicators. Combined, the project directly benefitted a total of 37,977 people, which is three times above the planned target of 10,000. It also indirectly benefitted over 32,000 people and thousands of livestock from non-project counties.

Table 4: Summary of Achievement of Outcomes

Expected Outcomes Rating 1. Improved climate change integration into national development planning. Substantial 2. Improved capacity at national and county level to plan adaptation measures Substantial 3. Improved capacity to implement climate adaptation measures. Substantial 4. Livelihoods adapted to climate and non-climate factors Modest Overall Achievement of Project Development Objectives Substantial

Outcome 1: Improved climate change integration into national development planning. Attainment of this outcome is rated “substantial” because of the strengthened institutional, technical and policy framework for mainstreaming climate into the development perspective as elaborated below. 54. The national institutional framework for mainstreaming climate knowledge and interventions into national development policies and programs, weak at project start, was strengthened through capacity building, reinforced institutional structures and an enhanced policy agenda. Targeted capacity building and TA equipped national and county government officials with the know-how to access and use the project generated climate knowledge products. This has helped enhance the dialogue on the inter-related climate change and land agendas for the ASALs. Tools developed for climate planning and budgeting, and the established institutional structures to coordinate climate change interventions, are serving as mechanisms to influence the mainstreaming of climate risk management into county planning and budgeting processes. The Climate Smart Agriculture Strategy22 (CSAS) was developed to guide investments and implementation of climate-smart agriculture (CSA) programs in order to improve productivity and food security, while addressing climate change adaptation and

22 Supported by the World Bank, FAO and UNDP

20

mitigation. It can now serve as a practical and actionable policy foundation to advance climate mainstreaming. Outcome 2: Improved ability to plan climate adaptation measures. Achievement of this outcome is rated “substantial” given the successful development and use of multiple climate planning and budgeting tools for adaptation measures as summarized below. 55. County level capacity to plan climate adaptation measures was significantly improved through preparation and use of a series of multi-stakeholder planning tools. Project financed TA helped develop 4 critical tools, which assisted county officials to concretely plan and budget for adaptation measures: (i) from a baseline of zero, 4 county management plans were developed which provided budgeted climate risk management programs. These were subsequently incorporated into County Integrated Development Plans; (ii) 15 CRPs were developed to guide the development of county climate management investments, of which 11 were specifically prepared for the KCSAP; (iii) 82 Community Action Plans (CAPs) helped identify the climate risk management activities that were implemented under the project; and (iv) 1 Methodology and tool kit for screening agricultural investment programs was developed to strengthen the ability of agriculture sector ministries to detect, advice, and mitigate negative impacts from planned or implemented programs. While the kit was not used under KACCAL because it was developed right before its closure, it is expected to be used under the ongoing IDA supported KCSAP, NARIG, and RPLRP, the proposed ECAAT, and other future programs. Outcome 3: Improved ability to implement climate adaptation measures. Realization of this outcome is rated “substantial” based on three factors, strengthened adaptation capacity of public and private service providers, and farmers; successful implementation of a significant number of trainings and demonstrations on adaptation measures; and farmer uptake of adaptation technologies and practices during a very short period. 56. From a baseline of zero, the project trained close to 500 public and private advisory agents23 in community climate risk assessment and management to support project beneficiaries. Climate risk awareness and the capacity of public and private advisory agents to assess climate risk was significantly enhanced through training and workshops. Through CRAs, the trained county officials and service providers, in collaboration with farmer CIGs, could identify what they believed to be the most critical climate related problems, and prioritize key areas for intervention. Together, they produced 8224 budgeted CAPs, which guided the selection of the county interventions implemented under the project. In addition, targeted training on safeguards and M&E improved the ability of county officials to integrate environmental and social safeguards into community planning, and community ability to monitor and evaluate community adaptation activities. The impact of the training is reflected in the quality of the demonstration activities, which received a satisfaction rate of over 96% from the participating farmers. 57. Improved capacity to implement climate adaptation measures is reflected in the successful planning, construction and operationalization of the 14 mega ICIs, the participation of counties and project beneficiaries in the development and implementation of 156 adaptation

23 County Agriculture Sector Programs Steering Committee (CASPSC), County Technical Teams (CTTs), Service Providers (SPs), Sub-County Technical Teams (SCTTs), value chain leaders and communities. 24 of which 2 were IP focused

21

demonstration micro-projects25, and farmer uptake of adaptation technologies and practices. At project end, the milk processing plant was operational, all 13 water pans were completed, fully charged and servicing both project and non-project community members and livestock. Out of 8,159 participating farmers who attended the demonstrations, 5,389 or 66% adopted a range of the technologies and practices because of the desire for both climate resilience and improved socio-economic status. Based on interviews conducted during the end of project impact evaluation, there is strong indication the technology uptake would have been higher had the implementation timeframe been longer. Project beneficiaries also indicated they would continue technology uptake even after project closure, and especially more so after the drought. This is also due in part to the multiplier effect of neighbors seeing the benefits of technologies and or practices they may not have previously taken up, but were prepared to, given the associated visible benefits and impacts. Outcome 4: Livelihoods adapted to climate factors was the outcome expected to result from the full range of project investments and combined benefits of outcomes 1, 2, and 3. The project generated important benefits and its overall preliminary impact is evident as summarized below. Although climate resilience takes time to accrue, it is clear that those participating in the project demonstrated much improved adaptation to climate variability impacts compared to non-project counties. 58. A detailed description of the benefits resulting from the project investments is presented below. (i) Increased skills, climate knowledge at national, county, community and individual farmer adaptive capacity. The project demonstrations and training which also had spill-over effects into non-participating community members, TA and capacity building, helped develop a range of adaptive and management skills for the different stakeholders. Capacity building also resulted in improved interactions between national and county government, and between county officials, private service providers and community members. (ii) Increased production. Beneficiary adaptation of technologies and practices (see annex 2, table 12) helped increase production in some of the target value chains. While climate resilience was a desirable factor in determining the uptake of adaptation technologies, socio-economic elements also played a significant and perhaps more prominent role. In general, because of anticipated greater production and yield potential, there was a higher uptake level of technologies for value chains26 with higher commercial value and an existing market for the related produce. For instance, in Tana River county, mango production increased three-fold (minimum 100 fruits per tree), following the uptake of improved varieties and proper mango husbandry techniques. In West Pokot, daily milk yields went up to 15-20 liters per day, per cow, despite drought, due to training and use of silage, a non-weather dependent feed storage method, and modern livestock feeding methods (enclosed pasture feeding, pasture and fodder planting of drought resistant crops27). The uptake of improved feeding methods and improved farm structures (livestock housing, local poultry housing, light diffuser store) allowed some farmers to expand the rearing of new animal breeds (cattle, chicken). In Tana River county, honey quality and production increased from 5-8 to 12-15 kg per hive, per harvest, from the uptake of the

25 Project funding covered the cost of extension service provision by service providers to farmers, the purchase of demonstration materials. 26 Such as livestock, commercially oriented crops and bee keeping. 27 Sweet potato, cassava, sorghum

22

Langstroth technology, bee housing and providing bees with planted forage. The result was an increase in revenue from 100 – 150 KES per kg to 300-500 KES per kg. (iii) Increased access to better quality water for people and livestock. The uptake of water harvesting techniques for crop enterprises (e.g. bore holes), construction of water pans for water storage for household and livestock use, and improved water management practices, allowed the harvesting of water which increased access to improved quality water for thousands of people and livestock. This was a critical achievement and very beneficial especially during the 2016\2017 drought. Over 29,000 direct beneficiaries, more than 24,000 indirect beneficiaries and thousands of livestock in both project and non-project counties benefitted from these investments. The number of animals with access to the Ruqa water pan alone for example increased from the estimated 12,000 to 28,000 because of animals migrating from non-project areas during the drought. In terms of the impact of improved water quality on beneficiaries’ health, while data was not collected, in all probability, it also contributed to improved hygiene and general health. (iv) Reduced distance and time to water points. Pre-water pan construction, the average distance - one way on foot, to Tana River for example (Garissa and Tana River Counties) was 35km and took on average 7-8 hours one way. Post construction, the distance and time for the residents of Ruqa and Koriyow Arbadabolo in Garissa county for instance, was reduced to 5kms one way and a maximum of 1 hour. This significant reduction was not only drastic in terms of time allocation, it also substantially reduced the overall stress and physical strain on those responsible for collecting water, in many cases women. Furthermore, the freed-up time allowed community members to tend to other family activities. (v) Reduced animal loss. While there was no official collection of the actual figures, the beneficiaries are convinced the improved water access for livestock in the target communities helped reduce some of the massive and devastating animal losses incurred during the drought. For pastoral communities whose only source of livelihood and wealth is livestock, this was a significant feat. This is in stark contrast to the livestock owners outside the project areas who faced the daily agony of either selling, killing or watching their animals die due to the drought. (vi) Reduced produce loss. The camel milk value addition facility, directly reached over 83 milk producers, mostly women. By project closure, it had reduced spoilage and wastage of camel milk by 20-30%. Reduced milk spoilage combined with improved collective marketing of the camel milk, is reported to have increased revenues from KES 50/lt to KES 80/lt. In West Pokot, the use of pheromone traps to reduce fruit flies28 and mango processing techniques significantly reduced mango losses. This allowed more mangos to reach the markets, thereby increasing the income earning potential for mango farmers. In some parts of Tana River county, mango farmers earned more from selling processed rather than fresh mangos. For instance, two fresh mangoes weighing roughly 1 kilo could be sold for 4 KES. The same two mangoes if processed would yield 3 packages of 150 grams each, which combined would bring in 75 KES. Better animal disease management and use of the innovative chepkube breeding technique significantly increased the survival rate of local indigenous chickens, which are both a food source and an income generator, in particular for women. (vii) Reduced incidences of water related conflicts. Training in leadership capacity, group dynamics and conflict resolution, coupled with the establishment of water pan management

28 On average, Kenya looses up to KES 5.5 billion annually due to the invasive species of the fruit fly. The fruit fly reportedly accounts for 70% destruction of mango and citric fruit production.

23

committees, user entrance/membership fees, and security measures for the beneficiary communities, are reported to have helped reduce incidences of water related conflicts. 59. KACCAL’s overall impact on the performance of targeted value chains in the project counties is visible. A comparison of overall performance of some of the targeted value chains in the project versus non-project counties with comparable agro-climatic and socio-economic status for the year 201729, indicates the project counties considerably out-performed the non-project counties (see table 5 below).

Table 5: Summary of Value Chain Performance

Value Chain Project County Non-Project Counties

Dairy (cattle)

Kilifi Kwale Annual Milk Production (MT) Annual Milk Production (MT)

25,961 11,700 West Pokot Elgeyo-Marakwet

85,588 49,148

Honey Production30 (Kg)

Kilifi Lamu

19,704 1,035

West Pokot Baringo

1,040,000 650,994

Mango (tons)

Tana River Monetary Value (KES)

Elgeyo Marakwet Monetary Value (KES)

32,066 211,268,000 14,343 132,782,530

Kilifi Monetary Value (KES)

Kwale Monetary Value (KES)

108,139 1,844,181,000 53,339 934,555,000 Source: Kenya Economic Review of Agriculture, 2017 60. Despite the notable benefits and promising impacts from the project investments, it is difficult to assess the extent of livelihoods adapted to climate factors as a result of the interventions, given the reduced project lifespan. Resilience takes time to accrue. A longer timeframe would have allowed for higher uptake rates. Presumably a longer duration would have also exposed the technologies and practices to potentially more variations in climatic and economic conditions. This would have allowed for a more robust assessment of the impact of the technology uptake on adapted livelihoods and overall contribution to enhancing resilience. 3.3 Efficiency Overall Rating: Modest

61. Efficiency (spending well for the expected results) from a cost-benefit and economic perspective is positive and is rated substantial. Project funds were fully disbursed, all activities were implemented and the project largely delivered its outputs in line with the original project costing. Overall the project has a positive economic perspective. However, the administrative

29 Source: State Department of Livestock Statistics, MoALF, 2018. 30 Aggregated production from traditional log hives, the Kenya Tob-bar Hive (KTBH) and the modern Langstrothe hives.

24

efficiency is modest, because of the project closing extensions, procurement and fund flow issues, and inefficiencies resulting from frequent institutional changes. This therefore brings the overall efficiency rating down to modest. 62. The ex-ante review assessed the financial viability31 (internal rate of return [IRR] and net present value [NPV] of incremental net benefits) of four selected CDD micro-projects32. The IRRs were estimated between 13% and 30% assuming a discount rate of 10%, and were considered favorable. 63. At project completion, the ex post financial analysis, using the methodology employed at appraisal, assessed the financial viability of the demand-driven CDD activities under component 333, also at a discount rate of 10%. It compared traditional cultivation practices (“without project”), with best practices presented on demonstration plots (“with project”), over a period of 20 years34. The results are largely positive and confirm the profitability of nearly all the demonstrations, except for dairy in Kilifi and apiculture in West Pokot, which have a negative net present value (NPV) and/or benefit-cost ratio below 1. Demonstration plots for (Tomatoes in Garissa, Local poultry in Kilifi and West Pokot, Mango in Tana River, Apiculture in Tana River and Kilifi) have cost-benefit ratios between 1.07 and 2.86 and IRRs between 18% and 262% (see table below).

Table 6: Results of financial analysis (NPV, IRR, BC-ration) for selected demonstration plots per county

Enterprises NPV IRR Benefit-cost ratio

Project areas

Tomatoes US$1,971 55% 2.86 Garissa Local poultry US$667 18% 1.12 Kilifi Local poultry US$3,472 - 1.33 West Pokot Mango US$5,006 52% 1.43 Tana River Apiculture US$6,041 262% 1.76 Kilifi Apiculture US$10,871 90% 2.15 Tana River Apiculture - US$5,254 - 0.36 West Pokot Dairy - US$1,309 - 0.53 West Pokot Dairy US$3,191 45% 0.5 Kilifi

64. The analysis shows that interventions for component 3 and component 2 were cost-effective in reaching many beneficiaries. Component 3 reached 8,159 beneficiaries at a total cost of US$275 per beneficiary over the project duration of 32 months (or US$8.5 for each month of project implementation and beneficiary). Component 2 reached 29,165 direct project beneficiaries which are expected to benefit from 14 intercommunity investments at a total cost of US$68 per beneficiary for the entire project duration of 32 months (US$2.2 per beneficiary for each month of project implementation). Under Component 1 the project developed 15 county risk profiles at a total cost of US$600,000. Eleven of the CRPs were a condition for the

31 The analysis did not include a cost-benefits analysis for the entire project. 32 small-scale irrigation, woodlots, beekeeping and sustainable land management practices 33 (most popular enterprises for which data was available - apiculture, dairy, local poultry, mango, tomatoes and for which demonstration plots showed the best practice examples), 34 The ex-ante economic analysis does not specify a time frame.

25

implementation of the KCSAP. KACCAL therefore enabled 15 counties a head-start into the KCSAP implementation and allowed savings of US$440,000 for the KCSAP.

3.4 Justification of Overall Outcome Rating Rating: Moderately Satisfactory (MS).

Table 7: Justification of Overall Outcome Rating

Rating Class Rating Relevance High Achievement of Project Outcomes Substantial Efficiency Modest Overall Rating Moderately Satisfactory 65. The overall project outcome is rated as “Moderately Satisfactory”. The project relevance remains high and is expected to generate net positive returns against cost. All project activities were implemented and the grant funds were fully used, which resulted in substantial achievement of the expected outcomes. All results targets were met or exceeded despite a compressed timeline. KACCAL has provided options for adaptive technologies and practices unavailable or limited at project start, climate planning and budgeting tools, and critical climate information and lessons useful for the development of future agriculture smart projects. Above all, KACCA has generated valuable results that are currently being harnessed under other ongoing climate smart agriculture operations, and which will likely continue to be made available to other parts of the country through future operations. However, overall efficiency is rated modest, given the implementation delays resulting from frequent institutional changes, lengthy procurement times, and slow release of funds for project implementation.

3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 66. Poverty Impacts. The project by design was pro-rural poor and pro the vulnerable and marginalized, given the selection of the ASALs as the geographical intervention area. Poverty and unemployment rates in the ASALs are among the highest in Kenya, where the weather variability has exacerbated the poverty levels, especially among women. Project investments along the various value chains were selected by the beneficiaries for their potential for high economic returns activities, potential to reduce the risk of losses to livelihoods, and improve income security. This logically was expected to bring improvements in their lives and reduce poverty levels in the long run. However, there was no systematic data collection on the overall impact of the project on poverty reduction. This makes it difficult to present a consolidated figure on poverty reduction attributable to the project. 67. Gender and youth. The Project made deliberate and conscious efforts to target women and youth as direct project beneficiaries by implementing discreet measures to increase their participation. This included: (i) gender awareness training for all beneficiaries to promote and facilitate incorporation of gender and vulnerability concerns into all the project activities; and (ii) identification and inclusion of enterprises appropriate for women and youth, such as local poultry breeding, cassava and sweet potato production. In terms of achieving gender parity, women made up 69% of the project beneficiaries at project end. This exceeded both the 50% project target goal and the 35% gender inclusion target stipulated in the Kenyan Constitution, indicating the gender sensitization efforts payed off. The project therefore contributed to the increased participation of women in the project’s various economic activities. On the other hand, while also specifically

26

targeted, findings from the project impact evaluation report indicate the participation of unemployed youth in many of the interventions was lower than expected, reportedly due to lack of land ownership. 68. Social Development. Social inclusion and community based social protection were major aspects of the project. VMGs\IPs35 were key project beneficiary targets. Four groups of IPs (the Warta in Kilifi, the Wailuana in Tana River and the Sangwer in West Pokot) were identified along with targeted interventions. Social protection and conflict management mechanisms were introduced and beneficiaries trained on how to use them. Lessons learned from implementation of these mechanisms stand to benefit other government development interventions. (b) Institutional Change/Strengthening 69. KACCAL strengthened institutional structures in the agriculture sector and established mechanisms to support planning and implementation of adaptive measures. The capacities of government officials (CASPSC, CTTs, SCTTs, MoALF’s CCU) created by predecessor programs\projects – KAPP, KAPAP and the KAPSLM and target communities considered limited and\or non-existent at appraisal were strengthened. At the county and community levels, the project instituted and strengthened additional structures such as the various grant management committees that facilitated project implementation. It likewise supported the strengthening of private and public SPs which promoted and further embedded beneficiary empowerment and ownership of the project among the beneficiaries. The capacity building of almost 500 public and private advisory agents in community climate risk management, while critical for facilitating the KACCAL implementation, has strengthened the human resource base for climate adaptation. Finally, the project supported the development of the Strategy for Climate Smart Agriculture which is expected to further strengthen and support the adaptive capacity and resilience of communities; establish mechanisms to minimize greenhouse gas emissions from agricultural production systems; and create an enabling regulatory and institutional framework. (c) Other Unintended Outcomes and Impacts (positive or negative) 70. There were no reported unintended negative impacts from the project. 71. There are several unintended positive spill-over effects of the project. The considerable reduction in distance and time to water because of the water pans has helped reduce strain on family relationships and interactions, as women no longer must be away from the home for extended periods. Over 24,000 community members and thousands of livestock from outside the project target areas equally benefitted from the water pans, especially during the severe 2016/2017 drought. Community members from outside the target areas attended and benefitted from the KACCAL trainings.

3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 72. The beneficiary satisfaction rate with the project was very high. The GoK commissioned a beneficiary assessment in 2016 in anticipation of the original closing date of

35 Included unemployed youth, Indigenous Peoples (IP), elderly women and men, widows/orphans, the differently-abled, recovering substance abusers, and people living with HIV/AIDS

27

October 2016 and an independent end of project impact evaluation in June 2017. Because the ICIs were still under construction, the former focused exclusively on assessing beneficiary satisfaction of the project investments on the community micro-projects. The assessment is based on interviews with project implementers at national and county levels, field surveys of project beneficiaries, MIS data and beneficiary and stakeholder perceptions. 900 households with both male and female headed households, representative of the project’s agro-ecological zones, and the project supported value chains and specific enterprises were randomly selected to participate in the field surveys. The end of project impact evaluation was conducted when the ICIs had been completed and therefore takes a comprehensive look at all the project interventions 73. Both assessments yielded high positive satisfaction rates above and beyond the project target. The main areas assessed were the overall project’s CDD approach to service delivery; the proposed menu of options for climate smart technologies and practices; the implemented ICIs; the capacity building initiatives; the deliberate measures to reach the VMGs\IPs; training on environmental and social safeguards, and the gender sensitization efforts and incorporation of gender perspectives into the project activities. Beneficiaries however also expressed disappointment with the overall reduction in scope and fund allocations to micro-projects and the short project implementation timeframe. They believe they would have benefitted more from the micro-projects, had they been implemented over a longer time. 74. Tables 6 and 7summarize the percentage of beneficiary satisfaction rates for different project aspects as reported in both the 2016 and 2017 assessments.

Table 8: Beneficiaries’ Ratings of Satisfaction or Higher for the Various Project Aspects

(Beneficiary Assessment, 2016

Project Aspects % Beneficiaries Rating Satisfaction or Higher

Garissa West Pokot Kilifi Tana River

Trainings 76.4 89.4 77.4 98.2

Demonstrations 70.9 82.4 77.5 87.0

Service provider (GRP) 59.3 95.7 75.3 96.3

CWG Leadership 83.7 93.1 86.7 87.1

Service fee committee 69.3 81.4 86.6 36.9

Procurement committee 63.0 83.1 62.8 57.5

Audit committee 42.5 68.6 42.9 29.0

Investment committee 66.7 66.1 66.7 62.5

CSU 66.0 86.7 84.0 91.3

CTT 37.2 70.6 73.4 48.9

MEAN 63.3 81.8 71.8 69.5

Source: Beneficiary Assessment, 2016

28

Table 9: Summary of % of beneficiaries’ satisfaction ratings (end of project evaluation)

Satisfaction Mean Beneficiary Satisfaction Rating Trainers 93.8 Adaptation activities 96.7 Usefulness of activities 97.3 Mean 95.9 Source: Gok end of project Impact Evaluation Report, June 2017

4. Assessment of Risk to Development Outcome Rating: Moderate 75. Key risks to the effectiveness and sustainability of the project interventions in the project areas are mainly weather and environmental, continued insecurity and conflict, and resources. 76. Weather and environmental risks. The ASALs are expected to remain vulnerable to severe and unpredictable weather conditions. The main risks to KACCAL’s development outcomes in the ASALs are therefore linked to the long-term viability of the project investments and climate resilient economic activities. A continued and strong climate smart agriculture portfolio will be crucial to scale up support to investments in climate change adaptation and risk management, to continue to reduce vulnerabilities to climate variability in the ASAL communities. The ongoing NARIG, KCSAP and RPLRP and future operations can play a significant role in further contributing to these efforts. 77. Policy and institutional risks. Kenya’s climate change policy environment is solid as demonstrated through a series of climate related policies and strategies. KACCAL further contributed to the policy foundation through development of the Climate Smart Agriculture Strategy. KACCAL also contributed to moving the climate mainstreaming agenda forward by strengthening some of the county institutional mechanisms. In conjunction with these efforts, resurrection of MoALF’s Climate Change Unit would further strengthen and likely re-energize the national climate mainstreaming agenda in the agriculture sector. 78. Continued insecurity and conflict risks. Conflict due to land and water use disputes and incidences of cattle rustling and theft of produce and small stock are often exacerbated by weather events, as was demonstrated during project implementation. Construction of the community water pans has helped reduce water scarcity and access for the direct beneficiaries in the respective project areas. Nevertheless, the water pans are also seen as a potential source of conflict in the event other communities are forced to migrate to find water sources. KACCAL instituted and implemented a variety of conflict mitigation measures which are reported to have successfully resolved small scale conflicts among the direct beneficiaries. Given the short project duration, it is difficult to reasonably assess the long-term effectiveness of the conflict management mechanisms, and their success in larger scale conflicts beyond the targeted counties. Institutionalization of conflict management at county level throughout the ASALs will need to be rigorously supported and scaled up to pre-emptively prepare for potential larger scale conflicts. 79. Lack of actual budget allocations for climate adaptation. It is expected that the annual CIDPs and CAPs will be used as intended, in the planning and budgeting for climate adaptation measures. However, counties must go beyond just budgeting for them and ensure that funds are made available for the selected adaptation activities on a regular and long-term basis.

29

5. Assessment of Bank and Borrower Performance

5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Unsatisfactory 80. The project’s design and rationale were sound. The project had a strategically and highly relevant development objective, logical components and activities that could reasonably contribute achievement of the PDO and realistic results targets. The project design built on the existing successful institutional architecture and approaches, first of the ARLMP II, and then KAPAP. It incorporated lessons learned, experiences and best practices from operations in Africa and other regions. However, the critical drawback was the extraordinarily long inception\preparation phase with considerable delays between the design stage and project approval. Furthermore, the lengthy INT forensic audit of the ALRMP II left the project in limbo for too long a period which launched the cascade of delayed effectiveness and project launch. (b) Quality of Supervision (including of fiduciary and safeguards policies) Rating: Moderately Satisfactory 81. Once the initial difficult phase passed, implementation support missions in collaboration with the GoK counterparts, were conducted on a regular basis. They included a variety of technical and operational expertise from the Bank’s headquarters and the country office. Missions provided concrete and constructive recommendations for improving project performance, as seen through the series of restructurings which helped improve performance and accelerate project progress. The MTR was instrumental in correcting the course of implementation and project performance. FM supervision identified weaknesses and recommended corrective measures which strengthened KACCAL’s internal controls and helped the project avoid some of the fiduciary issues that had surfaced under the ALRMP II. FM reporting indicates there were no identified incidences of fraud. M&E challenges such as data collection delays were followed up on closely. The environmental and social safeguards staff provided timely advice and guidance when needed and monitored the Grievance Redress Mechanism on a regular basis, which accommodated complaints at both the community and county levels. All complaints are recorded to have been satisfactorily addressed. 82. Overall supervision ratings were candid. While the project was rated “Unsatisfactory” at the start of the project, improved implementation allowed progressive upgrading of the ratings. It closed with an overall rating of “moderately satisfactory”, fully completed activities and achievement of all the expected results targets.

83. There were two drawbacks during supervision. The inability to get resolution of the GoK’s slow flow of funds issue, an ongoing problem reportedly common throughout the entire Kenya portfolio continued to plague the project up to its closure, despite consistently being raised as a critical issue in aide-memoires and management letters to the GoK. Bank management has yet to get satisfactory resolution of the problem which is cause for strong concern for future operations if not resolved. There was an unusually high TTL turnover (5 over the course of the project duration). However, the presence of 3 of them in the country office, together with the fiduciary and safeguards staff, offset some of the negative impacts as it allowed for closer interaction with the GoK counterparts and quick, real-time support to resolve issues.

30

(c) Justification of Rating for Overall Bank Performance Rating: Moderately Unsatisfactory 84. Overall Bank performance is rated “moderately unsatisfactory” based on a combination of the “moderately unsatisfactory” rating for the Bank’s performance for quality at entry and “moderately satisfactory” for the quality of supervision. 5.2 Borrower Performance (a) Government Performance Rating: Moderately Unsatisfactory 85. From a policy perspective, the GoK’s commitment to its climate change agenda was exceptionally strong as demonstrated by the variety of policy framework documents available at project appraisal. Per the GEF grant agreement (GA) between the World Bank and GoK, the government honored its commitment to provide counterpart funding36 and the necessary institutional, administration, implementation and monitoring infrastructure. However, its overall performance was significantly marred by the extensive delays in facilitating project effectiveness, the delays in recruiting critical climate change expertise at the national and county level, the baffling delayed opening of a designated project account, persistent delays in the flow of funds from Treasury to the project account throughout implementation, and long procurement delays. On this basis, the GoK’s performance is rated “moderately unsatisfactory”. (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 86. The KAPAP Secretariat had a slow start in assembling the climate expertise needed to support the KACCAL implementation. Nevertheless, despite staffing challenges at different intervals of implementation, the KS adequately played its part in accelerating project implementation once the enabling conditions were in place. Support mission aide-memoires indicate that most mission recommendations were carried out. The KS provided the necessary coordination for mission field visits, coordinated regular internal national & county level meetings, monitored safeguards compliance and evaluated the performances of the SPs. Overall, responses both at the field level and to the Bank were timely. While it is likely that implementation may have been rushed in some areas (e.g. beneficiary consultations) to align with the reduced project timeframe, the KS, CSUs and general implementation apparatus managed to navigate the numerous challenges to deliver on the expected results. On this basis, the implementing agency’s performance is rated “moderately satisfactory”.

36 which covered staff salaries, internal and external auditing, project monitoring services, and implementation review mission costs.

31

(c) Justification of Rating for Overall Borrower Performance Rating: Moderately Unsatisfactory 87. The “moderately satisfactory” rating for the implementing agency performance is offset by the “moderately unsatisfactory” rating for borrower performance. The overall borrower performance is therefore calibrated to a rating of “moderately unsatisfactory”.

6. Lessons Learned 88. Several key lessons learned have been fleshed out. They include:

(i) Frequent institutional changes are disruptive to project implementation and need to be minimized as much as possible or avoided. The elimination of the original ministry responsible for KACCAL, i.e. Ministry of State for the Development of Northern Kenya and Other Arid Lands; the transfer of KACCAL to the MoAFL; reassignment of project implementation to a different implementing agency; and Kenya’s devolution process were the cause of major delays and implementation challenges at different stages of the project.

(ii) Enhancing resilience is a long-term process and requires more than one program of

intervention. As a pilot project, KACCAL was an important launch pad for introducing adaptive technologies to a population critically in need for them. It is expected that ASAL community members who adapted some of the KACCAL technologies will continue to use them, especially if the economic returns are also significant. The KCSAP and the Resilience Project are well placed and equipped to consolidate and expand the groundwork done under KACCAL on a large scale to benefit the wider population in the ASALs and other areas of the country.

(iii)CDD approaches require time for adequate consultations, community mobilization and

sensitization before the roll out of project activities to ensure beneficiary understanding and internalization of the proposed activities. Sufficient time should be an integral element of project design. The long delays in getting KACCAL launched are noted to have significantly reduced critical time for consultations and sensitization initiatives.

(iv) Budgetary planning and the flow of funds for agricultural projects need to be aligned with

climatic patterns/seasonality and not the budgetary cycles to ensure implementation of interventions at the appropriate time. The delays in the flow of funds from the GoK’s Treasury to the MoALF project account had drastic impacts on implementation including for example the disruption of the implementation sequencing of activities such as production of the CRPs and delivery of inputs such as seeds at inappropriate time (dry\drought season).

(v) Fully adapting livelihoods to climatic factors from the uptake of climate adaptive

technologies takes time and is determined by more than just the desire for resilience. Other factors such as type of technology and ease of adoption, physical and economical accessibility, expected economic impacts, socio-economic status of the adopters and environmental factors economic viability of the value chain, markets, profit levels, individual interests play an important part in the decision-making for uptake. Project interventions therefore also need to keep in mind the local socio-economic conditions in the choice of adaptation technologies made available to project beneficiaries.

32

7. Comments on Issues Raised by Grantee/Implementing Agencies/Donors (a) Grantee/Implementing agencies None (b) Cofinanciers/Donors None (c) Other partners and stakeholders None

33

Annex 1. Project Costs and Financing

(a) Project Cost by Component (in USD Million equivalent)

Components Appraisal Estimate (USD millions)

Actual/Latest Estimate (USD

millions)

Percentage of Appraisal

Component 1: Climate information products, policy, and advocacy

1.46 2.25 65

Component 2: Climate risk management at district and local levels

1.37 2.00 68.5

Component 3: Community-driven initiatives for climate resilience 2.67 1.25 213

Total Baseline Cost 5.50 5.50 100

Physical Contingencies 0.00

0.00

0.00

Price Contingencies 0.00

0.00

0.00

Total Project Costs 5.50 5.50

(b) Financing

Source of Funds Type of financing

Appraisal Estimate

(USD millions)

Actual/Latest Estimate

(USD millions)

Percentage of Appraisal

GEF Grant 5.50 5.50 100

Government of Kenya Budget allocation 0.69 0.13 19

34

Annex 2. Outputs by Component

1. Below is a summary table of the component outputs as linked to the results framework.

Table 10: Summary of Component Outputs as Linked to the Results Framework

Intermediate indicators

Component outputs and linkages

Component 1. Climate information products, policy, and advocacy Sub-component 1.1: Development of climate-related knowledge products to inform climate risk management (CRM) strategies in ASALs

Result indicator One: Climate risk profiles developed and used for district management plans (number)

15 CRPS were developed, the original 4 for the targeted project areas and an additional 11 for counties outside the project areas. However, due to procurement delays, development of the CRPs was not completed early enough during implementation for them to inform preparation of the county climate risk planning and management processes for KACCAL as originally intended. However, the CRPs are being used under the KCSAP to guide climate smart agriculture county investments and priorities and the RPLR Project. They are also expected to be used in the proposed ECAAT project currently under preparation.

Subcomponent 1.2.: Integration of climate action into national ASAL development plans and programs

Result indicator: Methodology and tool for screening agricultural investment programs for climate risk developed

A tool kit to enhance the ability of the agricultural sector ministries to detect, advice and mitigate any adverse impacts from programs implemented or planned at the sectoral level was developed. However, it was developed some months before project closure due to procurement delays from unavailability of the required technical expertise among firms which competed for the contract. The tool will be used under the ongoing KCSAP and RPLRP, and in future operations such as the proposed ECAAT.

Component 2: Climate risk management at the district level Subcomponent 2.1: Capacity building to integrate climate risk management into district planning processes.

Result indicator: Public and private advisory agents37 trained in community climate risk management. Result indicator: Percentage of Kenya Agriculture and Agribusiness Project, sub-projects in the Participating Districts screened for improving response to climate risk.

A total of 49738 agents were trained in various aspects of climate change interventions39. The original target was 80. 100%. 156 KAPAP sub-projects.

37 County officials KS/CSU, CTTs CASSC and public and private advisory/extension agents (i.e. service providers at county level). 38 Data from June 2017 GoK ICR. 39 Climate change risks & management modules (profiling, assessment tools & methodologies); Project sensitization at county and sub-county levels; Project implementation modules (planning, budgeting, & grant management)

35

Subcomponent 2.2: Support for “climate-smart” public and private investments.

Result indicator: Public and private sector investments rated satisfactory or better by beneficiaries (%) (beneficiaries assess whether outcomes have been achieved).

96%. Beneficiary satisfaction with all aspects of the project investments was very high. The expected project outcomes were achieved and all results targets were met or exceeded. There is indication of some level of improved resilience from the uptake of various adaptation technologies and practices.

Component 3: Community-driven initiatives for climate resilience Subcomponent 3.1: Support for community capacity building.

Result indicator: Community Action Plans (CAPs) with concrete climate risk management activities reflected in the budget.

82 CAPs out of a target of 32 were prepared.

Subcomponent 3.2: Support for community-based micro-projects.

Result indicator: Community adaptation projects developed and implemented.

Number of direct beneficiaries (of which % are female).

KACCAL funded 156 micro-projects out of the original target of 80 of the potential interventions identified in the CAPs. A total of 37,977 out of the target of 10,000 beneficiaries were reached by project interventions, of which 69%40 were women.

Table 11: KACCAL supported value chains and enterprises

Value Chain Enterprises County Total Garissa Kilifi Tana

River West Pokot

Dairy Dairy-Camel Dairy-Cow Dairy-Goat

1 2 1

Fruit/Nuts/Vegetables Mango Potato Tomato African Bird’s Eye Chili Bulb onion

2 1 1 1

Meat Local poultry Sheep & goat Beef

2 2 1

NRM Apiculture 3

Total 3 4 4 7 18 Source: KACCAL Final Impact Evaluation Report, June 2017

40data from last ISR as updated by the GoK in June 2017.

36

Table 12: Summary of Climate Change Technologies and Practices Demonstrated to Project Beneficiaries

Technology Justification

Pasture and fodder planting, & conservation To improve availability of feed during dry periods

Water harvesting technologies for crop enterprises, & water pans for water storage for livestock, apiculture & domestic use

To harvest rain water for use during drought periods for human consumption & to sustain crops, livestock and bees

Superior crop varieties and management practices for tomatoes, mangos, potatoes and bulb onions

To improve yields

Improved livestock breeds To increase livestock, milk & meat productivity

Integrated pest and disease management & control (mango & ABEC), field hygiene (mango); crop rotation (Tomato), push & pull, Indian crow traps (ABEC)

Decrease incidences of pests and diseases

Drought resistant crops (sweet potato, cassava, maize and sorghums)

To enhance availability of livestock feeds

Value addition for crop & livestock produce (with focus on mango & milk value chains)

to reduce spoilage resulting from high temperatures as well as increasing shelf-life

Sustainable land management technologies (farm yard manure, composting, foliar feeds, sunken beds, river bank protection, tree nursery establishment & agro-forestry)

To strengthen integrated soil fertility management

Improved farm structures (improved apiary houses and hives, livestock housing, Light diffuse store, Chepkumbe Jiko brooding technology, local poultry housing, Biogas digesters)

To minimize absconding of bees from hives and increase bee productivity; reduce movement & strengthen management of fragile ecosystems, & pest & disease management/control; prolong shelf life potatoes seed; increase survival rate of chicks; deter pests & predators; & reduce deforestation & maximize benefits from dairy farming.

Source: KACCAL Final Impact Evaluation Report, June 2017 Table 13: Service Providers Trained in Climate Change Risks per County

Type of training West Pokot

Garissa Tana River Kilifi* Totals

Climate change risks & management modules (profiling, assessment tools & methodologies)

27 33 66 Data not available

126

Project sensitizations at county and sub-county levels


256

Project implementation modules (planning, budgeting, & grant management)


115

Total 167 180 150 497 Source: GoK Impact Evaluation Study 2017

37

Table 14: Water Pan Distribution per County

County Water Pans Size Direct Beneficiaries

Status

Garissa Hara Khundi Arabadabalo

Ruqua

20,000 m 20,000 m 20,000 m

3,000 600 2,245

Completed Completed Completed

Kilifi Kipawa Makutano Bibithole

23,000 m 20,000 m 20,000 m

3,200 2,000 3,000

Completed Completed Completed

Tana River Shirikisho Kitole

20,000 m 20,000 m

1,500 1,400

Completed Completed

West Pokot Yemtin Lapaten

Cheptapa Kaptisio

Chelarachpogh

24,000 m 24,000 m 22,000 m 23,000 m 20,000 m

2,000 2,500 3,750 1,500 2,500

Completed Completed Completed Completed Completed

Total 13 29,195 Source: GoK Impact Evaluation Study 2017

38

SAMPLE BEFORE AND AFTER PICTURES OF SOME OF THE MEGA COUNTY WATER PANS

Kitole Water Pan – Tana River County Prior to construction of the water pan, Kitole residents traveled extremely long distances to access limited quantities of water daily.

39

Kitole Water Pan – Tana River

The completed and fully charged Kitole water pan directly serviced about 1,400 residents and hundreds of livestock. It significantly reduced time and distance to water.

40

Roqa Water Pan – Garissa County Prior to construction of the Roqa water pan, Roqa residents, mostly women, traveled up to 35 kms a day and could only carry 1 small container of water because of the distance.

41

Roqa Water Pan – Garissa

The completed and fully charged Roqa water pan, directly serviced about 2,245 residents and hundreds of livestock. It reduced the distance and time for the Roqa residents to an estimated 5kms one way for a maximum of 1 hour.

42

Annex 3. Economic and Financial Analysis Appraisal assumptions 1. The Economic and Financial Analysis (EFA) at project appraisal included the following: (i) an overview of the socioeconomic importance of the ASALs; (ii) a summary of general issues in economic analyses of climate change adaptation projects; (iii) a summary of the literature on the economic impacts of climate change; (iv) calculation of the internal rate of return (IRR) and net present value (NPV) for potential community-driven micro-projects; and (v) conclusions and recommendations. Elements (i)-(iii) are based on literature reviews. 2. The ex-ante analysis did not include a cost-benefits analysis for the entire project. The analysis stated that while a range of approaches and methodologies exist to quantify economic costs and benefits of climate change adaptation in the agricultural sector, they have a range of constraints, require intensive data collection, and lead to uncertain results. In addition, project investments which directly target agricultural land management and could be used to calculate economic benefits are relatively few. Instead a large proportion of resources is allocated to capacity building and institutional strengthening, for which cost-benefit analyses are difficult to conduct due to the long-term nature of activities and difficulties in linking causes and effects. The analysis elaborates that the assessment of benefits of climate change adaptation is difficult. To understand whether adaptation to climate change has been successful, a control group is needed which faced the same climate change events but did not choose the same adaptation activities. This is not the case in this analysis. Irrespective of this constraint, the ex-ante EFA assessed the overall financial profitability of certain CDD investments. 3. The ex-ante EFA assesses the financial viability (internal rate of return and net present value of incremental net benefits) of four selected KACCAL CDD micro-projects: small-scale irrigation, woodlots, beekeeping and sustainable land management practices. The activities were selected based on discussions with communities in the project area (particularly Marsabit41 and Garissa) and on evaluations of ALRMP I and II. Data was collected during field visits and included evaluations of the baseline project. The analysis assumes a discount rate of 10%. 4. The ex-ante results show that the CDD enterprises have a potential to provide sizable benefits. For irrigated tomatoes and kale cultivation a farmer’s financial IRR and NPV was 13% and KES14,150 per hectare (US$136, in 2018 US$)42. For woodlots, which include benefits from fuelwood collection, the IRR was estimated to be 19% (NPV of US$818 per hectare); without fuelwood benefits the IRR was estimated at 10% and US$13 per hectare). For communities engaged in beekeeping, the IRR could reach up to 14% and US$476 per hectare for an enterprise with 10 beehives. And for selected sustainable land management interventions and a period of 50 years an IRR of 30% could be achieved. ICR analysis: profitability and IRR for selected demonstration plots 5. KACCAL took a demand-driven CDD approach under Component 3: Community-driven initiatives for climate resilience and supported the establishment of demonstration plots and capacity strengthening through targeted advisory services delivered by service providers to communities, to adopt the practices on their farms. KACCAL financed 156 demonstration plots

41 Which was later dropped after the transfer of KACCAL to KAPAP. 42 All US$ values in this paragraph are presented in 2017 terms.

43

(Table below presents the type of enterprises and number of beneficiaries in each target area). For this analysis, a similar approach was chosen for the ex-ante analysis and financial analyses for popular enterprises such as - apiculture, dairy, local poultry, mango, tomatoes – to assess financial viability of the respective demonstration plots.

Table 15: Types of Beneficiaries Selected Enterprises per County

Value chains

Number of beneficiaries in community

No. of Demonstration Plots

Counties Reached Adopted (a)

Mango 751 643 15 Tana River, West Pokot

Milk 1,261 742 25 Garissa, Tana River, Kilifi, West Pokot

Meats 1,918 1,211 36 Garissa, West Pokot Tomatoes 862 689 11 Garissa

Apiculture 1,682 1,078 36 Tana River, Kilifi, West Pokot

ABEC 360 312 10 Kilifi Local poultry 832 498 16 Kilifi, West Pokot Irish potatoes 373 216 6 West Pokot Onions 120 0 1 West Pokot

8,159 5,389 156

Notes: (a) adopted refers to number of beneficiaries who have adopted some of the presented CSA practices at household level. Figures were provided by the PIU.

6. For each enterprise, traditional cultivation practices (“without project”) were compared with the best practices demonstrated on the demonstration plots (” with project”). A discount rate of 10% as in the ex-ante analysis is used and a timeframe of 20 years is assumed43. An exchange rate of 101KES to US$1 is used. The financial analyses present the net present value (NPV) of incremental net benefits, the financial internal rate of return (IRR) and the benefits-cost ratio (ratio between discounted benefits and cost streams). Due to the project delays and adversities like drought events, demonstration plots were carried out in the last two and a half years (October 2014-June 2017) of the project and thus may not have realized the maximum potential benefits at the time of closing. For this financial analysis, benefits are projected towards year 20. All data and information was received from the Project Implementing Unit and Service Providers. The model descriptions are provided in the following paragraphs. 7. Tomatoes: On the demonstration plot for tomatoes in Garissa, the project supported investments in inputs, such as improved seeds, fertilizer, fungicides, manure, mulching, staking and pruning, which affected labor cost; supplementary irrigation from a nearby river, the establishment of a nursery, application of mechanization and machinery (tractor) during land preparation, and grading and sorting after harvesting. These changes in management practices led to a yield increase from 2.8 tons/acre to 5 tons/acre. The net present value of the incremental net benefit was US$2,010 (US$4,964 per hectare), with a cost-benefit ratio of 2.86.

43 The ex-ante analysis does not specify a time frame except for investment in sustainable land management practices for which a time frame of 50 years is assumed. 20 years is an assumption frequently used in agricultural projects and therefore adopted here.

44

Table 16: Results of Financial Analysis (NPV, IRR, BC-ration) for Selected Demonstration Plots per County

Enterprises NPV IRR Benefit-cost ratio

Project areas

Tomatoes US$2,010 55% 2.86 Garissa Local poultry US$304 14% 1.07 Kilifi Local poultry US$3,514 - 1.33 West Pokot Mango US$5,065 52% 1.43 Tana River Apiculture US$6,112 262% 1.76 Kilifi Apiculture US$11,000 90% 2.15 Tana River Apiculture - US$5,358 - 0.36 West Pokot Dairy US$3,263 - 2.03 West Pokot Dairy US$1,565 30% 0.45 Kilifi

8. Local poultry: The project promoted local poultry in Kilifi and West Pokot, where appropriate hen housing units were constructed and improved local poultry breeds were introduced. Additional interventions included vaccinations which reduced mortality rates, improved brooding technology and production of improved feed formulation. The benefits stem from increased egg production, reduced mortality and improved quality and therefore increased sale prices of the laying hens. In Kilifi, the demonstration plot with 20 improved hens (compared to 3 hens “without project” scenario) provides a return of US$304 and an IRR of 14%. In West Pokot, a demonstration plot with 50 hens (compared to 25 hens in “without” project scenario) results in an NPV of incremental net benefits of US$3,514. 9. Mango: The demonstration plot in Tana River promoted climate-smart mango production practices such as grafting and pruning mangoes, top-working of unproductive mangos, water harvesting and integrated pest management (pheromone traps to trap fruit flies) which improved fruit quality and increased production. In addition, value addition activities were introduced and a solar dryer, sealer and packaging material, storage facility and related electricity costs were covered by the project. Benefits are derived from an increased yield by 60%, increased commodity price per piece of fruit from 2 to 3 KES which reflects the improved quality of the fruit, and selling 1,500 packages of dried mango. The operation reaches benefits in year 2 and achieves an NPV of US$5,065 and an IRR of 52%. 10. Apiculture: The project facilitated the construction of improved bee keeping houses (houses constructed to provide cool space for bees), use of appropriate harvesting kits, honey extractors, honey processor, and, in some sites, planting of sunflowers for use by bees and establishment of ponds. The benefits of the demonstration site in Tana River, which established 15 improved Langstroth hives, were compared with the “without” project scenario of sorghum production and resulted in an NPV of net benefits of US$11,000 and an IRR of 90%. Similarly, in Kilifi, 10 modern beehives were estimated to achieve an NPV of US$6,112 and an IRR of 262%. Both sites reported average yields of 15 and 12 kg per hive and harvest twice a year. In contrast, West Pokot reported an average harvest of 14 kg/hive but only one harvest per year, which would result in a net present value loss of -US$5,358. This was a consequence of the drought during project implementation, empty water ponds, failure to cultivate sunflowers to attract bees and subsequently low or, in some sites, no yield. 11. Dairy: In ASALs dairy cows are often held outside and farmers walk long distances to find adequate pastures for feeding. On dairy demonstration farms the project introduced the production of nappier grass with the “tumbukiza” technology, the establishment of a small stable

45

and promotion of fodder conservation and storage, the establishment of watering holes, and on selected demonstration plots (e.g. in Kilifi in three out of five demonstration farms) the establishment of biogas units to promote the use of slurry44. For a demonstration plot with 1 cow in Kilifi, the financial analysis showed that annual average income in “without” and “with” project scenario is negative45 and the expected increase in productivity did not outweigh the increase in cost. The NPV over 20 years results in US$1,565 and 30% but a benefit-cost ratio of 0.45. Similarly, in West Pokot, the average annual income in “without” and “with” project scenario results in a NPV of US$3,263 and a benefit-cost ratio of 2.03. For Kilifi a lower average dairy yield (6 liters/cow) was reported compared to West Pokot (7.5 liters/cow), which may explain the negative benefit-cost ratio. Both demonstration sites presented a set of potential but costly practices, which explains why the net present value of cost did not outweigh the net present value of benefits in Kilifi. 12. The results of the demonstration plots are mixed, but largely show positive results except for apiculture in West Pokot and dairy in Kilifi. When interpreting the results, the following should be taken into consideration: (i) The productivity gains reported for some of the demonstration plots, were observed at a time of drought and therefore far below the potential, or they have been projected to times with normal weather conditions. (ii) The objective of the demonstration plot may not be maximizing the demonstration plot’s net benefits. Instead the objective is to present a variety of potential climate-smart agriculture practices of which farmers can pick the most suitable and affordable for their private farms. Thus, the interventions and cost items which were considered in the financial analysis of demonstration plots may lead to higher cost than an average rural household would incur46. This may explain why net present value of costs outweighed net present value of benefits on several demonstration sites. 13. Technology uptake is reported to have been quite high. 8,159 beneficiaries were reached by the demonstration plots and learned from these interventions. Even though the project did not provide funding to individual households for the adoption of climate-smart agriculture practices, the end of project impact evaluation study reported that 66% (5,389 beneficiaries) adopted some of the demonstrated practices/technologies at household level. The component cost of US$2.25 million (41% of total cost), results in US$275 spent per beneficiary for training and capacity over the project duration of 32 months, or US$8.5 for each month of project implementation and beneficiary.47 14. However, as in the ex-ante analysis, an aggregate cost-benefit analysis of project investment, which considers beneficiaries’ project benefits, is not conducted due to following reasons: (i) The extent of adoption is insufficiently documented - beneficiaries have selectively adopted demonstrated technologies. If and how each technology translates into tangible benefits (e.g. increase in productivity, increase in production, increase in stability of income) and thus increases in income was not assessed. (ii) The project timeframe was deemed too short for beneficiaries to fully internalize the demonstrated content and achieve tangible income increases. (iii) As stated in the ex-ante analysis, KACCAL introduced climate change adaptation measures for existing commodities which were previously promoted by KAPAP. The incremental value of KACCAL

44 The biogas unit is not considered in the analysis. 45 In Kilifi the annual average net income without project was -US$1,632 and with project -US$734 and in West Pokot without project -US$21 and with project -US$205. 46 At the same time, the effect on productivity of a single practice versus a combination of practices, or an understanding what the most effective practice is to enhance productivity, was not established on the demonstration sites, nor in this analysis. 47 Even if the cost of US$8.5 per month would accrue to beneficiaries for a period of 20 years – to approximate the value of learning and capacity building - the NPV of incremental net benefits would decline by between 8% (Apiculture in Tana River) to 57% (Dairy, Kilifi) but remain positive for all of enterprises, expect for local poultry in Kilifi (NPV of -585) and apiculture in West Pokot.

46

would be the additional benefit of climate change resilience, which could not be documented due to the lack of a control group.48

15. While no cost-benefits analysis can be conducted, the cost-effectiveness of Component 2 Climate risk management at the county level is considered. Component 2 reaches 29,248 direct project beneficiaries of which 29,165 are benefitting from the 13 water pans and 83 beneficiaries are benefitting from the camel milk value addition plant in Bura- Tana River. Considering the investment of over US$2 million (36% of total investment) this implies that US$68 were spent per beneficiary for the entire project duration of 32 months or US$2.2 for each month of project implementation, which is equivalent to daily rural wage rate.

16. Component 1 Climate information products, policy, and advocacy produced, amongst others, 15 county climate risk profiles at a total cost of US$2.25 million (41% of total project investment). Eleven of the profiles were instrumental for the project design of the KCSAP, which targets 24 counties, and are conditional for counties to plan their project activities and commence implementation. The CRPs were prepared by the International Center for Tropical Agricultural (CIAT) at US$40,000 per profile, resulting in a total of US$440,000. While the 11 CPRs don’t directly support KACCAL’s project beneficiaries, the project provided budgetary support to 15 in the KCSAP to provide it a “head-start” into project implementation, thus saving them not only money but also time. Conclusion

17. Climate change adaptation practices as promoted in KACCAL, have the potential of achieving sizable incremental net benefits compared to traditional practices. This was demonstrated by the analysis of a sample of demonstration plots in the project areas which achieved benefits-cost ratios between 1.07 and 2.86 and IRRs between 14% and 262%. Two demonstration plots (apiculture in West Pokot and dairy in Kilifi) led to a negative NPV of incremental net benefits and/or benefit-cost ratio below 1. The findings of the appraisal stage analysis, that selected CDD enterprises are financially viable for households, is confirmed. 18. The project investments did not provide funding to adopt the climate-smart agriculture practices to individual households in the targeted communities. However, 66% of the beneficiaries are reported to have adopted the demonstrated technologies and or practices on their private farms. However, the extent of adoption (number and type of technology, timing) is insufficiently documented, and attributable and quantifiable benefits (e.g. increased yield or income) at household level, could not be credibly estimated in this analysis. As in the appraisal stage analysis, no cost-benefits analysis was conducted. 19. The analysis shows that interventions for component 3 and component 2 were cost-effective in reaching many beneficiaries. Component 3 reached 8,159 beneficiaries at a total cost of US$275 per beneficiary for training and capacity over the project duration of 32 months (or US$8.5 for each month of project implementation and beneficiary). Component 2 reached 29,248 direct project beneficiaries which are benefitting from the 14 intercommunity investments at a total cost of US$68 per beneficiary for the entire project duration of 32 months (US$2.2 per beneficiary for each month of project implementation). Under Component 1 the project developed 15 county risk profiles at a total cost of US$600,000. Eleven of these CPRs were for

48 In addition, beneficiaries referred to a range of driving factors, e.g. personal interest, neighbors, other projects such as KAPAP, as reasons for having taken up the technology or commodity.

47

implementation of the Kenya Climate-Smart Agriculture project. KACCAL therefore gave 11 counties a head-start into implementation and allowed it savings of US$440,000.

48

Annex 4. Grant Preparation and Implementation Support/Supervision Processes

(a) Task Team members

Names Title Unit Responsibility/ Specialty

Lending/Grant Preparation Christine E. Cornelius Lead Operations Officer AFTAR TTL Johannes Woelcke Senior Economist AFTAR Economics Arati Belle Natural Resource Economist AFTEN Natural Resources

Management Ingrid Mollard Consultant AFTAR Banu Setlur Operations Analyst MNSSD Environmental

Safeguards Nyambura Githagui Sr. Social Development Specialist AFTCS Social Safeguards Jorge Uquillas-Rodas Consultant AFTEN Henry Amuguni Financial Management Specialist AFTFM Financial Management Efrem Fitwi Procurement Specialist AFTPC Procurement Robert Njagi Frank Sperling Ian Burton Benjamin Guillon Christophe Crepin GEN2B Team member Jorge E Uquillas Rodas Consultant GSUGL Masud Mozammel Sr. Communication Officer ECRGP Team member Fredrick Kamande Karanja Ingrid Mollard M&E Consultant GFA13 Team member Fredrick Semazzi Tom Mboya Owiyo

Supervision/ICR Christine E. Cornelius Lead Operations Officer AFTAR TTL Johannes Woelcke AFTAR TTL Banu Setlur Sr. Environmental Specialist GFAGE Team member Arati Belle Disaster Risk Management Spec. GSU18 Team member Daniel O. Otunge Andrew Mwihia Karanja Sr. Agriculture Specialist AFTAR TTL Gibwa A. Kajubi Sr. Social Development Specialist GSU04 Team member Wilson Kinyua Julius Githinji Muchemi Ladisy Komba Chengula Lead Agriculture Economist GFA07 TTL Joseph Oryokot Sr. Agriculture Specialist GFA07 TTL Nora Kaoues Sr. Agriculture Economist GFA07 Team member Valens Mwumvaneza Sr. Agriculture Specialist GFA13 Team member Sophie Nelly Rabuku Program Assistant AFRVP Administration Support Abel Lufafa Sr. Argriculture Specialist GFA12 Team member

49

Marketa Jonasova Operations Officer GFAGE Team member Harideep Singh Lead Operations Officer GFAGE Team member Mark Owuondo Odiaga Consultant GEN01 Team member Irene Nambuye Musebe Agriculture Specialist GFA07 Team member Irene Bomani Operations Analyst GFA07 ICR Author Christine Heumesser Economist GFAGE Team member *Blank spaces= information no longer available as individuals have left the Bank

(b) Staff Time and Cost

Stage of Project Cycle Staff Time and Cost (Bank Budget Only)

No. of staff weeks USD Thousands (including travel and consultant costs)

Lending FY07 3.78 70,186.00 FY08 9.67 81,357.00 FY09 19,343.00 FY10 9.20 38,136.00

Total: 22.65 208,945.00 Supervision/ICR

FY09 3.96 29,073.00 FY10 0.00 0.00 FY11 0.45 1,733.00 FY12 8.03 50,294.00 FY13 11.41 45,630.00 FY14 8.73 24,866.00 FY15 0.00 0.00

Total: 65.58 151,596.00

50

Annex 5. Beneficiary Survey Results 1. The objectives of the beneficiary survey report commissioned by the GoK in July 2016 in anticipation of the original closing date of October 2016, were to assess and document levels of beneficiaries’ satisfaction on: (i) project investments; (ii) climate change adaptation capacities gained; and (iii) preliminary uptake by beneficiaries of climate change information and knowledge. Its findings are based on: (a) quantitative and qualitative data collected from interviews with farmer leaders, service providers, CTTs, CSUs, local county administrators and a sample 50 CWGs, including IPs (361 respondents), randomly selected from the 156 supported CWGs; and (b) a comparison of the interventions per county, target value chains and socio-economic status. 56% of the respondents were female and 44% were male. The survey claims a 100% response rate.

Key Findings:

2. Overall satisfaction with project interventions. 80% of the respondents are reported to have participated in both the establishment of demonstration sites and trainings. Overall, the project was rated satisfactory or higher by most the beneficiary respondents (West pokot-81.8 %; Kilifi 71.8%, Tana river-69.5%; Garissa-63.3%). Among the key shortcomings were the overall insufficient project duration and the short training and demonstration periods.

Figure 2: Beneficiaries’ Satisfaction with Project Aspects Figure 1: Beneficiaries' Satisfaction rates on Various Project Aspects

Source: GoK Beneficiary Assessment, July 2016

3. Climate change adaptation measures, prioritized by the respondents in order of highest expected impact on enhanced resilience to climate change challenges across the value chains, were water harvesting technologies and planting of forage trees, fodder and drought tolerant crops. However, respondents believed the small-scale nature of the water conservation structures would not sufficiently shield farmers from prolonged droughts such as the one experienced during the project period. The water pans and bore holes would have significantly more impact.

0

10

20

30

40

50

60

70

80

90

100

Train

ings

Dem

onst

ratio

ns

Serv

icepr

ovide

r (G

RP)

CWG

Lead

ersh

ipSe

rvice

fee

com

mitte

ePr

ocur

emen

tco

mm

ittee

Audit

com

mitte

eIn

vest

men

tco

mm

ittee

CSU

CTT

Train

ings

Dem

onst

ratio

ns

Serv

icepr

ovide

r (G

RP)

CWG

Lead

ersh

ipSe

rvice

fee

com

mitte

ePr

ocur

emen

tco

mm

ittee

Audit

com

mitte

eIn

vest

men

tco

mm

ittee

CSU

CTT

Male Female

Unsatisfactory Marginally satisfactory Satisfactory Highly satisfactory

51

4. Uptake of Climate Change Information and Knowledge Products (IKPs). There was an overall uptake level of the IKPs of over 55% for all the value chains except apiculture and ABEC in Kilifi county which had 49% and 51.3% respectively. The report attributes the lower rates to the persistent drought during the project period which reportedly discouraged beneficiaries from investing in the IKPs. Per the report, the value chains with the highest uptake levels were; Tomatoes (100%); Mango in Tana River (94%); Bulb onion (93%) and dairy cow in West Pokot (81.4%). This is attributed to the high commercial value of these value chains and good market access. It also captures a higher level of willingness towards higher adoption of technologies which bring in higher incomes. Value chains with the lowest uptake levels included modern/improved bee hives in West Pokot and Kilifi; bee forage and drought tolerant maize in Kilifi; and tree nurseries in Tana River. For Kilifi and Tana River, drought is considered the contributing factor. For the apiculture farmers, the cost of modern/improved bee hives proved to be prohibitive.

5. Factors determining the uptake of technologies included the technology type, physical and economic accessibility, commercial value and anticipated economic impacts, socio-economic status of the adopters and environmental aspects. Commercial value ranked high as a factor as seen in the uptake for goat meat in Tana River (95.7%), mango in Tana River (94.1%) and Bulb onion in West Pokot (93%). On the other hand, the 48% and 29% uptake levels for drip irrigation kits among ABEC and Irish potato beneficiaries demonstrate the impact of availability of technology and costs respectively.

6. Information Delivery Channels. The project’s use of the public private partnership service delivery system was considered more effective than traditional public extension services in providing information to beneficiaries. The respondents’ preferred information channels were the project trainings and demonstrations, government extension services, radio programs and neighbors; and the least preferred were television programs, extension leaflets and Non-Governmental Organization. Participation in trainings was rated (very good-18.9%, good-61.3%).

7. Challenges. The respondents highlighted a number of difficulties: (a) group dynamics; (b) gender inequalities including women’s limited access to and control over resources especially land and livestock; (c) unmet and unclear beneficiaries’ expectations from the project; (d) language barriers between beneficiaries and service providers and other implementing agents due to shortage of native technical personnel in the counties; (e) a short implementation period that led to rushed trainings and demonstration sites; (f) tedious and lengthy procurement procedures; (g) market dynamics; (h) poor road infrastructure paired with wide geographical coverage where some areas lacked internet and mobile connectivity; (i) human-wildlife conflicts and human-human conflicts; and (j) prolonged drought which predominantly negatively affected the water and soil moisture interventions and demonstrations.

52

Annex 6. Stakeholder Workshop Report and Results N/A

53

Annex 7. Summary of Borrower's ICR 1. Overall Project Goal and Relevance to National and Global Development Objectives The overall goal of Kenya Adaptation to Climate Change in Arid and Semiarid Lands (KACCAL) Project was to enhance the resilience of communities and the sustainability of rural livelihoods threatened by climate change in the arid and semi-arid lands of Kenya. The Project Development Objective was to improve the ability of participating districts and communities to plan and implement climate change adaptation measures. To achieve that objective, KACCAL was expected to support and enhance the climate information base, strengthen the adaptive capacity of stakeholders and mainstream climate risk management into development plans and investment programs. 2. The mainstreaming approach was expected to fit into Government of Kenya (GOK) and Bank’s strategy and their global commitment to scale up efforts to address the risks posed by climate change. In this regard, the Project was expected to contribute towards several policy objectives of the government and its international commitments. The project was to contribute towards the achievements of Vision 2030 where the government acknowledges that ASALs require special attention to achieve sustainable poverty reduction and economic growth. KACCAL was also linked to the objectives of the Kenya Strategy for Revitalizing Agriculture (2005) and its successor the Agricultural Sector Development strategy (ASDS) both of which emphasized the importance of reducing risk and vulnerability for groups whose livelihoods rely on the use of natural resources. Other key policies and strategies that the Project was expected to contribute included the Draft Land Policy, the Draft ASAL Policy, and the Draft Disaster Management Policy. Apart from the national policies, KACCAL was to contribute towards the country’s global environmental commitments especially the United Nations Framework Convention on Climate Change (UNFCCC) and the United Nations Convention to Combat Desertification (UNCCD). 3. Project Design and Quality at Entry - At operational level, KACCAL was restructured in June 2012 to be closely linked to Kenya Agricultural Productivity & Agribusiness Project (KAPAP). The project was to add value to KAPAP’s farmer and community level interventions by supporting capacity building and investments towards climate change adaptation and resilience strategies. The Project was also designed with shared implementation and institutional arrangements with KAPAP. This was to reduce the operation costs of the Project and ensure the climate change agenda is taken on board in a more sustainable manner. The linking of the two projects was expected, among other benefits, to shorten the learning period and offer valuable lessons especially on implementation of community micro-projects. The linkage was also to offer the Project the institutional capacity that was already in existence under KAPAP. The restructured KACCAL project was set to be implemented over the period 2012-2016 but the actual implementation did not commence until 2014 due to several factors including delays in staff recruitment and opening of project account. 4. Achievement of PDO and Overall Assessment - The interventions proposed and implemented under KACCAL were carefully identified to deal with challenges of mainstreaming climate change policies and strategies at national and county levels. The achievement of the project development objectives was being tracked through two main indicators: PDO 1 – Number of District management plans with concrete climate risk management activities reflected in the budget and PDO II – Percentage of Community adaptation projects rated satisfactory or better by participating communities. The two PDO indicators should have been changed during the restructuring of project in June 2012 to reflect the institutional changes that had taken place and impacting on county planning and budgeting. Furthermore, due to the long delay in the

54

commencement of the project, the assessment is based on project implementation in the last 18 months. It is also noted that the project has initiated processes for establishment of structures that will be responsible for mainstreaming climate change adaptation into national policies, and planning and budgeting at county level. Considering these developments, both PDO indicators are rated as ACHIEVED. The project intermediate outcomes were being tracked through 8 intermediate indicators of which 6 were achieved or exceeded during the same period. Only two intermediate indicators were rated PARTIALLY ACHIEVED and the rating for one is likely to improve when the requisite information is collected before the closure of the project to confirm initial impressions. Overall performance of the project is rated as ACHIEVED. 5. Key Achievements - The project team implemented 156 community driven initiatives for enhancing climate resilience and 14 intercommunity investment micro projects. The water conservation structures will improve the resilience of the communities against scarcity of water caused by frequent droughts in the operational areas. The project also completed the development of Climate Risk Profiles for 15 counties of which 4 have been disseminated and shared with stakeholders in the project operational areas. Other key achievements include development of Methodology and Tool for screening climate risk for agricultural projects and formulation of the Kenya Climate Smart Agricultural Strategy. In addition, the project was during the short period able to reach 13,424 direct beneficiaries. It is estimated that by the end of the extended implementation period the project may benefit more 20,000 beneficiaries and several livestock. 6. Factors that Affected Performance - Several factors undermined the performance of KACCAL. These factors included: i. The project was designed to be implemented for about 4 years, but it was effectively

implemented over the last 18 months. The delay in project start-up was caused by two factors: Change in domicile of the project from Ministry of State for the Development of Northern Kenya and Other Arid Lands to Ministry of Agriculture Livestock and Fisheries; and delay in recruitment of climate change experts to build capacity for KAPP and CSUs levels. The staff were expected to be on board by Dec 31st, 2012, but were recruited in January and September 2014, for the KS and CSUs staff respectively.

ii. The climate change interventions were new and complex and hence the delay in the start-up denied the implementers adequate time to put in place structures and build capacity for implementation of the project.

iii. Convoluted financial flows and bureaucracy delayed release of funds to the project adversely affecting implementation of planned activities. The opening of the designated account at the Central bank presented logistical challenges resulting in long delays between time of raising withdrawal applications and the time the resources were received in the project account and disbursements made to project beneficiaries.

iv. Low levels of literacy among the beneficiaries handicapped capacity building efforts and methodologies by service providers.

v. The prolonged droughts experienced in the project areas hindered the success of climate adaptation interventions. Farmers were expected to implement what they learnt during group demonstrations and upscale on their own farms to tackle climate smart related challenges at household level. However, despite SPs mounting good demonstrations, not all farmers were able to implement what they had learnt due to the prolonged drought.

55

vi. The lengthy procurement processes delayed the delivery of key project outputs which

depended on material inputs and consultancy services.

vii. Poor communications infrastructure in the counties and vast distances between implementation sites.

viii. Insecurity in some project areas affected not only project interventions but also M&E. 7. Project Management and M&E – Following the collapse of the ASDS coordination mechanism, the Agricultural Sector Programmes Steering Committee (ASPSC) continued to provide guidance and oversight functions. It was observed that the assumption in the design of the project that KACCAL would ride on the institutional infrastructure of KAPAP was not fully met due to staff changes and work overload for KAPAP staff. Thus, collection of data for monitoring and reporting suffered. Despite these weaknesses, the project staff, service providers and communities made commendable efforts in achieving current project results in the last few months of the project implementation. 8. Sustainability and Risks to development outcomes - The key risks to the effectiveness and sustainability of KACCAL outcomes fall into three main areas namely: i. Policy and institutional risks: At the national level the mainstreaming of climate change

issues was to be spearheaded by the climate change unit at the MOALF. The unit has however been moribund and almost rendered ineffective through several institutional changes and staffing challenges. Thus, although the NCCRS and climate change action plan are in place, there is lack of an effective institutional mechanism for climate change mainstreaming at sector level. This poses a critical risk to the initiatives and expected outcomes from KACCAL. However, since resilience of rural ASALs communities is a key objective of the project, KACCAL has initiated at county level the establishment of institutional structures that will ensure the integration of climate change issues in the future county planning and budgeting processes. These institutional structures are underpinned in the Climate Change Act No 11 of 13 May 2016.

ii. Weather and environmental risks: ASALs are prone to severe weather conditions mainly floods and droughts. The investments made especially the water pans are therefore exposed to risks of siltation, contamination and spread of diseases due to increased congregation of human and livestock. To mitigate against these risks KACCAL has trained the beneficiary communities to conserve the catchments, stabilize the embankments, and regulate access to the facilities. Regarding the incidences of crop and livestock diseases and post-harvest losses that are associated with severe weather conditions, KACCAL has trained the beneficiary communities in integrated pest management and other technologies for ameliorating weather conditions or increasing shelf life for commodities.

iii. Insecurity: In the ASALs, insecurity manifests in two main ways: Firstly, there is the common banditry that entail cattle rustling and theft of produce. Secondly, there water and pasture related conflicts. In times of drought, the search for water and pasture leads to migration of pastoral communities in ASALs. The water pans constructed under KACCAL could therefore attract pastoral communities beyond the immediate beneficiaries leading to conflict. While KACCAL has trained the beneficiary communities in conflict resolution this may not be adequate in dealing with conflicts emanating from communities outside the project areas. Therefore, there is a need for national and county governments to invest in

56

wider inter-community conflict resolution mechanisms and more water pans to reduce such conflicts.

9. Lessons Learned – The challenges faced in the implementation of KACCAL provide useful lessons for the design and implementation of future related projects. These lessons include: i. The CRA process is a powerful tool for mobilizing community support and ownership of

identified interventions. This was instrumental in implementation of micro projects despite delays in release of funds.

ii. The use of Force Accounts in PPP arrangements can considerably reduce cost of infrastructure projects for communities. With adequate facilitation and GOK technical support, communities can implement and supervise projects at a cheaper cost than private contractors.

iii. The disbursement of funds for agricultural projects should be synchronized with climatic patterns/seasonality rather than fitting to the budgetary cycles. For instance, the planting of grasses was affected by drought because planting materials were distributed during drought.

iv. Project interventions should contextualize local socio-economic conditions in the choice of technologies. For instance, drip irrigation technology for onion production in West Pokot and ABEC in Kilifi were unaffordable for most farmers while quality polythene liners for water harvesting were easily accessible.

v. There is a need for county and national governments to coordinate agricultural programmes. This will support joint planning, implementation and monitoring of various projects and initiatives. It will further help to harmonize approaches and avoid duplication of efforts. Currently, different programmes are perceived to be in competition instead of complementing each other.

vi. The consortia approach has worked relatively well in delivery of services to stakeholders. However, to attracting competent service providers and ensure equitable delivery of services to all stakeholders there is need to improve management and financial package.

vii. It is observed that agricultural interventions in many projects are relatively small in scale and not adequate to help farmers’ transition subsistence to commercial farming.

viii. The project MIS function needs to be designed, tested and capacity built early in the

project implementation for it to play the critical role of providing information for effective monitoring and evaluation of project performance. Once a project commences without an effective MIS it becomes thereafter difficult to design one because priority is given to implementing interventions.

57

Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders N/A

58

Annex 9. List of Supporting Documents 1. Project Concept Note

2. Project Appraisal Document

3. Grant Agreement between GoK and IDA

4. Country Partnership Strategy 2014-2018

5. Preparation and implementation mission reports

6. Gok’s Impact Evaluation Report, June 2017

7. GoK’s Implementation Completion Report, June 2017

8. Restructuring Paper June 2012

9. Restructuring Paper March 2016

10. Restructuring Paper October 2016

11. Restructuring Paper April 2017

12. Project ISRs

13. Project IFRs

14. Audit Reports

15. Kenya Climate Smart Agriculture Strategy

16. National Policy for the Development of the Arid and Semi-Arid Lands of Kenya

17. Kenya National Climate Change Response Strategy

18. Economic Review of the Agriculture Sector, 2017, Ministry of Agriculture, Livestock and Fisheries

19. State Department of Livestock Statistics, 2018; Ministry of Agriculture and Irrigation

20. https://youtu.be/XSicsxmUeXI

https://youtu.be/XSicsxmUeXI

59

MAP 1: KACCAL PROJECT COUNTIES

60

MAP 2: MAP OF KENYA