Document of The World Bank Report No: ICR2265 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-H0810 IDA-H5270) ON A GRANT IN THE AMOUNT OF SDR 18.1 MILLION (US$27 MILLION EQUIVALENT) AND AN ADDITIONAL GRANT IN THE AMOUNT OF SDR 6.4 MILLION (US$10 MILLION EQUIVALENT) TO THE REPUBLIC OF MALAWI FOR A COMMUNITY-BASED RURAL LAND DEVELOPMENT PROJECT March 30, 2012 Sustainable Development Department Agriculture and Rural Development Unit Country Department AFCS3 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective January 31, 2012) Currency Unit = Kwacha (MKw) MKw 163 = US$1 US$ 1.5500 = SDR 1 FISCAL YEAR ABBREVIATIONS AND ACRONYMS ADMARC Agricultural Development and Marketing Corporation ASWAp Agricultural Sector Wide Approach ASWAp-SP Agricultural Sector Wide Approach Support Project BG Beneficiary Group CAS Country Assistance Strategy CBRLDP Community Based Rural Land Development Project CEM Country Economic Memorandum DGA Development Grant Agreement EIA Environmental Impact Assessment EIRR Economic Internal Rate of Return EMP Environmental Management Plan EPA Extension Planning Area ESA Environment and Social Analysis ESMF Environment and Social Management Framework ESSRP Environmental and Social Screening and Review Process EU European Union FAO Food and Agriculture Organization of the United Nations FISP Farm Input Subsidy Programme GIS Geographic Information System GoM Government of Malawi ha Hectare GPS Global Positioning System HH Household IDA International Development Association IEC Information – Education – Communication km Kilometer LA Local Assembly LDF Local Development Fund MAIWD Ministry of Agriculture, Irrigation and Water Development MASAF Malawi Social Action Fund M&E Monitoring and Evaluation MDGS Malawi Growth and Development Strategy MFDP Ministry of Finance and Development Planning MKw Malawi Kwacha MLHUD Ministry of Lands, Housing and Urban Development MLGRD Ministry of Local Government and Rural Development MT Metric Ton (1,000 kg) MVAC Malawi Vulnerability Assessment Committee NAO National Audit Office NPV Net Present Value NSO National Statistics Office OP/BP Operational Procedures/Bank Policies PAD Project Appraisal Document PDO Project Development Objective PIM Project Implementation Manual PMP Pest Management Plan PMU Project Management Unit PVA Poverty and Vulnerability Assessment SIL Specific Investment Lending SLWM Sustainable Land and Water Management Vice President: Obiageli Katryn Ezekwesili Country Director: Kundhavi Kadiresan Sector Manager: Karen McConnell Brooks Project Team Leader: Hardwick Tchale ICR Team Leader: Olivier Durand REPUBLIC OF MALAWI Community-Based Rural Land Development Project CONTENTS Data Sheet A. Basic Information........................................................................................................ i B. Key Dates .................................................................................................................... i C. Ratings Summary ........................................................................................................ i D. Sector and Theme Codes ........................................................................................... ii E. Bank Staff ................................................................................................................... ii F. Results Framework Analysis ..................................................................................... iii G. Ratings of Project Performance in ISRs ................................................................... iv H. Restructuring (if any) ................................................................................................. v I. Disbursement Profile .................................................................................................. v Main Document 1. Project Context, Development Objectives and Design ............................................... 1 2. Key Factors Affecting Implementation and Outcomes .............................................. 6 3. Assessment of Outcomes .......................................................................................... 13 4. Assessment of Risk to Development Outcome......................................................... 19 5. Assessment of Bank and Borrower Performance ..................................................... 20 6. Lessons Learned ....................................................................................................... 22 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 23 Annexes Annex 1. Project Costs and Financing .......................................................................... 25 Annex 2. Outputs by Component ................................................................................. 26 Annex 3. Economic and Financial Analysis ................................................................. 29 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............ 38 Annex 5. Beneficiary Survey Results ........................................................................... 41 Annex 6. Stakeholder Workshop Report and Results................................................... 44 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 45 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ....................... 55 Annex 9. List of Supporting Documents ...................................................................... 56 A. Basic Information Community-Based Country: Malawi Project Name: Rural Land Development Project P075247 IDA-H0810,IDA- Project ID: L/C/TF Number(s): P115226 H5270 ICR Date: 03/30/2012 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: MALAWI Original Total XDR 18.10M Disbursed Amount: XDR 22.79M Commitment: Revised Amount: XDR 24.50M Environmental Category: B Implementing Agencies: Ministry of Lands, Housing and Urban Development (MoLHUD), Private Bag 111, Lilongwe. Cofinanciers and Other External Partners: None B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 07/20/2001 Effectiveness: 07/12/2004 07/12/2004 Appraisal: 12/02/2003 Restructuring(s): Approval: 04/13/2004 Mid-term Review: 11/30/2007 03/30/2007 Closing: 06/30/2009 09/30/2011 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: Moderate Bank Performance: Satisfactory Borrower Performance: Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Moderately satisfactory Implementing Quality of Supervision: Satisfactory Satisfactory Agency/Agencies: Overall Bank Overall Borrower Satisfactory Satisfactory Performance: Performance: -i- C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry Yes None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes None time (Yes/No): Supervision (QSA): DO rating before Satisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 5 4 General agriculture, fishing and forestry sector 89 89 Sub-national government administration 6 4 Public administration – Agriculture, fishing and forestry 3 Theme Code (as % of total Bank financing) Decentralization 17 12 Land administration and management 33 38 Participation and civic engagement 17 13 Rural markets 33 26 Social analysis and monitoring 5 Tax policy and administration 3 Personal and property rights 3 E. Bank Staff Positions At ICR At Approval Vice President: Obiageli Katryn Ezekwesili Callisto E. Madavo Country Director: Kundhavi Kadiresan Hartwig Schafer Sector Manager: Karen Mcconnell Brooks Richard G. Scobey Project Team Leader: Hardwick Tchale Rogier van den Brink ICR Team Leader: Olivier Durand ICR Primary Author: Olivier Durand - ii - F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) To increase the incomes of about 15,000 poor rural families by implementing a decentralized community-based approach to land acquisition and farm development in four districts. Revised Project Development Objectives (as approved by original approving authority) To increase the agricultural productivity and incomes of about 15,000 poor rural families through the implementation of a decentralized, voluntary community-based land reform program on eligible land in the Project districts. (a) PDO Indicator(s) Original Target Actual Value Baseline Formally Values (from Achieved at Indicator Value Revised approval Completion or (20081) Target Values documents) Target Years Increase in real incomes of participating families as compared to those of control groups and of 4,530 n/a 11,330 30,500 such families prior to the project (Malawi Kwacha /month). Increase in agricultural Maize: productivity of participating 450 1,500 1,800 families as compared to that of n/a such families prior to the Project Tobacco: (Kilograms/hectare). 300 1,000 800 (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Baseline Achieved at Indicator Revised approval Value Completion or Target Values documents) Target Years Component 1 – Land Acquisition and Farm Development Number of farm families established on land acquired 0 n/a 15,000 15,142 through the project. Component 2 – Land Administration Number of beneficiary groups which have received appropriate 0 n/a 452 666 documentation of land ownership. Number of Land Registries 0 n/a 32 32 computerized and functional. 1 Baseline value at time of AF appraisal - iii - Original Target Actual Value Formally Baseline Values (from Achieved at Indicator Revised Value approval Completion or Target Values documents) Target Years Reduced number of days taken to 118 n/a 110 90 register a land transfer. (2008) Component 3 – Capacity Building Number and quality of farm development proposals by eligible beneficiary communities and/or 0 n/a 33,000 33,428 hectares of land received, reviewed, approved and fully disbursed. Increased number of staff trained in surveying and registration of 0 n/a 30 702 land parcels including monitoring and collection of land tax/rent. Component 4 – Project Management, Monitoring and Evaluation 4 quarterly Improved timeliness and quality of Quarterly reports reports generated by the M&E n/a n/a reports per year submitted on system. on time time Complete and acceptable Final impact Timely and acceptable project n/a n/a impact assessment impact evaluation report. assessment delivered report G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 06/29/2004 Satisfactory Satisfactory 0.00 2 11/30/2004 Satisfactory Satisfactory 1.29 3 12/16/2004 Satisfactory Satisfactory 1.29 4 06/28/2005 Satisfactory Satisfactory 1.42 5 11/07/2005 Satisfactory Moderately Satisfactory 2.93 6 12/30/2005 Satisfactory Moderately Satisfactory 3.28 Moderately 7 06/29/2006 Moderately Unsatisfactory 7.18 Unsatisfactory Moderately 8 12/11/2006 Moderately Satisfactory 9.78 Unsatisfactory 9 04/27/2007 Satisfactory Satisfactory 11.48 10 11/26/2007 Satisfactory Satisfactory 17.60 2 Includes staff trained at MSc, Diploma and certificate levels (see Table 4 in Government ICR). - iv - 11 02/13/2008 Satisfactory Satisfactory 18.20 12 06/15/2008 Moderately Satisfactory Moderately Satisfactory 22.89 13 11/30/2008 Moderately Satisfactory Moderately Satisfactory 24.83 14 06/09/2009 Satisfactory Moderately Satisfactory 26.76 15 12/10/2009 Satisfactory Satisfactory 27.00 16 06/15/2010 Satisfactory Satisfactory 28.44 17 03/16/2011 Satisfactory Moderately Satisfactory 33.22 18 11/03/2011 Satisfactory Moderately Satisfactory 34.74 H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions Additional financing, reformulation of PDO and 10/27/2009 Y S MS 27.00 change to performance indicators. If PDO and/or Key Outcome Targets were formally revised (approved by the original approving body) enter ratings below: Outcome Ratings Against Original PDO/Targets Satisfactory Against Formally Revised PDO/Targets Satisfactory Overall (weighted) rating Satisfactory I. Disbursement Profile -v- 1. Project Context, Development Objectives and Design 1. Malawi's agriculture remains dualistic. At independence in 1964, it inherited an agriculture sector comprising a few, large foreign-owned, export-oriented estates and a dominant smallholder sector of mostly poor, subsistence farmers. Colonial laws allowed estates to obtain the full value of export sales, but prevented smallholders from producing burley tobacco which is the most lucrative export crop. These laws also forced smallholders to sell through the parastatal, the Agricultural Development and Marketing Corporation (ADMARC). The situation has improved, but much still needs to be done with respect to improved distribution and more efficient utilization of land, and more effective support to smallholders. About 84 percent of agricultural value-added comes from about 1.8 million smallholders who on average own 1 hectare (ha) of land. 2. Land pressure is particularly severe in the south of Malawi (where average holdings can be as low as 0.1 ha). About 1.1 million hectares of land are held in some 30,000 estates, with an average size of landholding ranging between 10 and 500 ha. Land distribution is sharply unequal, and overcrowded arable land exists next to under-used land parcels. It is estimated that one third of arable land is under-utilized, while household surveys demonstrate a direct link between poverty and access to land. In addition, Malawi's history of the carving out of large-scale farms (estates) from customary land, combined with increasing land pressure, has led to the emergence of localized tensions around land, sometimes expressing themselves through encroachments and invasions. 1.1 Context at Appraisal 3. The Community-Based Rural Land Development Project (CBRLDP) was designed to make a start in addressing emerging social conflicts related to unequal access to land by piloting a government assisted, community-driven land transfer program to land-deprived small scale farmers. Malawi is a landlocked country with an average population growth rate of 3.3 percent per year. It remains significantly dependent on agriculture which contributes about 36 percent to Gross Domestic Product (GDP), employs 85 percent of the work force, and contributes 90 percent of foreign exchange earnings. 4. Therefore, one of the key constraints to improved smallholder productivity is the small and declining land holding sizes. In 2002, a new National Land Policy (NLP) was adopted by the Government of Malawi to correct some of the historical land issues and inequality. The favourable policy environment, together with the availability of land for sale by willing estate owners, provided an opportunity for the introduction of a land redistribution project based on voluntary land transfers between landowners (willing sellers) and the land-poor (willing buyers). The CBRLDP aimed at addressing this challenge with a particular focus on reducing unequal land distribution, land market failure and land administration weaknesses, while supporting decentralization, community participation and programmatic approaches. The project was aimed at piloting a government assisted, voluntary and community-based land reform approach. -1- 5. In the Bank 2003 Country Assistance Strategy (CAS), the CBRLDP was identified as an operation with high strategic importance towards contributing to the second CAS pillar to “establish a platform for sustainable poverty-reducing growth�. Linkage was also established with the 2004 Structural Adjustment Credit that included land elements, namely the dissemination of the Government’s Land Policy adopted in 2002, the drafting a new Land Bill and the raising and improvement of collection of leasehold rents, and the future implementation of a land tax on freehold land as incentives to the release of unused land to the market (rental or sale). Bank involvement in the proposed operation relied on extensive experience and lessons learned from community- driven development land reform in Brazil. 1.2 Original Project Development Objectives and Key Indicators (as approved) 6. The original Project Development Objective (PDO) as stated in the Development Grant Agreement (DGA) was “to increase the incomes of about 15,000 poor rural families through the implementation of a decentralized, voluntary community-based land reform pilot program on eligible land in the Project districts�3. 7. The Project outcome indicators were defined as follows: (a) Incomes of participating families compared to control groups (“with- without�) and pre-project income levels (“before-after�); (b) Increased and sustainable agricultural production on participating farms (“with-without� and “before-after�); and (c) The effective evaluation of the piloted approach. 8. The original key performance indicators are presented in the following table: Table 1 - Original key performance indicators Original intermediate results Original intermediate result indicators Component 1 – Land Acquisition and Farm Development Land acquired, number of  Number of farm families established on land acquired beneficiary families achieved and through the project. agricultural production achieved.  Amount, speed and cost of land acquisition (per beneficiary family and per hectare).  Speed and cost of establishment of agricultural production (per beneficiary family and per hectare). 3 It should be noted that the PDO in the original PAD was slightly different from the DGA in terms of formulation, but not in substance: “to increase the incomes of about 15,000 poor rural families by implementing a decentralized, community-based and voluntary approach to land reform in four districts in southern Malawi�. The PDO in the PAD annex 1 is also different: “To increase the incomes of about 15,000 poor rural families by implementing a decentralized, community-based approach to land acquisition and farm development in four Districts�. -2- Original intermediate results Original intermediate result indicators Component 2 – Land Administration Decentralized, transparent land  Number of beneficiary groups or beneficiaries which administration system in place. have received appropriate documentation of land ownership.  Number, speed and cost of titles, cautions and hectares registered and sketch maps completed. Component 3 – Capacity Building Capacity to implement community  Number and quality of land acquisition and farm driven land acquisition and farm development proposals by eligible beneficiary development exists or is created. communities and/or hectares of land (i) received; (ii) reviewed; (iii) approved; and (iv) fully disbursed. Component 4 – Project Management, Monitoring and Evaluation Monitoring and evaluation system  Timeliness and content of reports generated by M&E provides appropriate information system. for effective project management and scaling-up decision after pilot project completion. 9. Project outputs were defined as: (a) Providing access to resources for land acquisition and farm development to about 15,000 poor beneficiary families; (b) Providing beneficiaries with secure title to the land acquired through the piloting of decentralized land administration; (c) Building capacity at community, district and national levels to implement a community-driven approach to land reform; and (d) Implementing a monitoring and evaluation system to assess project inputs, outputs, and outcomes, and provide a sound basis for the decision to scale-up the project. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 10. The PDO was slightly revised in October 2009 when processing an additional financing to include the “agricultural productivity� dimension. The revised PDO is “To increase the agricultural productivity and incomes of about 15,000 poor rural families through the implementation of a decentralized, voluntary community-based land reform program on eligible land in the Project districts.� 11. The additional financing gave the opportunity to refine the PDO level indicators as follows: (i) the income indicator (a) to be reported as real and not nominal income in order to control for price increases or inflation in the income changes; (ii) the outcome indicator on increased production (b) was changed to “increased agricultural productivity� to align it with the revised the PDO; and (iii) the third outcome indicator (c) was moved to the intermediate result indicator level as it was more aligned to component 4 outputs. -3- 12. Four new intermediate results indicators were added to better track progress achieved in strengthening land administration capacity. Two original intermediate results indicators were dropped as they were deemed difficult to monitor and measure, namely “Amount, speed and cost of land acquisition per beneficiary family and per hectare� and “Speed and cost of establishment of agricultural production per beneficiary family and per hectare�. 13. The revised outcome and intermediate results indicators are summarized in the following table: Table 2 – Revised outcome and intermediate results indicators Refined outcome indicators  Increase in real incomes of participating families as compared to those of control groups and of such families prior to the project (Malawi Kwacha/month).  Increase in agricultural productivity (maize and tobacco yield) of participating families as compared to that of such families prior to the Project (Kilograms/hectare). Revised intermediate results Revised intermediate result indicators Component 1 – Land Acquisition and Farm Development Land acquired, number of  Number of farm families established on land acquired beneficiary families achieved. through the project. Component 2 – Land Administration Decentralized land,  Number of beneficiary groups which have received transparent administration appropriate documentation of land ownership. system in place.  Number of Land Registries computerized and functional (28 Districts, 3 Regions and 1 national).  Reduced number of days taken to register a land transfer. Component 3 – Capacity Building Capacity to implement  Number and quality of farm development proposals by community driven land eligible beneficiary communities and/or hectares of land acquisition and farm received, reviewed, approved and fully disbursed. development exists or is created.  Increased number of staff trained in surveying and registration of land parcels including monitoring and collection of land tax/rent. Component 4 – Project Management, Monitoring and Evaluation Monitoring and evaluation  Improved timeliness and quality of reports generated by the system provides appropriate M&E system. information for effective project management and  Timely and acceptable project impact evaluation report. scaling-up decision after pilot project completion. -4- 1.4 Main Beneficiaries 14. The primary beneficiaries of the Project were defined as poor, land-poor and food insecure families from the pilot districts. The original project was initially implemented in four pilot districts (Mulanje, Thyolo, Machinga, and Mangochi), then extended in October 2008 to two other districts (Balaka and Ntcheu) as requested by the Government of Malawi (GoM). Beneficiaries were to be self-selected according to democratic and local customary principles, and with the capability to farm. The objective was to support 15,000 land poor households among the 1.8 million smallholders who on average own less than 1 hectare of land. Land administration services and staff represent a second important category of Project beneficiaries whose capacities were to be strengthened through training and equipment. This second category received a stronger attention under the additional financing. 1.5 Original Components (as approved) 15. The Project had the following four components: (a) Land Acquisition and Farm Development (US$16.78 million4) implemented with direct community involvement through identification, negotiation and acquisition of land and the preparation of farm development proposals, approved by a district-level multi-stakeholder entity, taking into account legal, technical, environmental, and poverty reduction criteria. These sub-projects are carried out through the award of land acquisition and farm development grants to project beneficiaries who are the poor, landless and food insecure households from the pilot districts. (b) Land Administration (US$8.00 million) to facilitate the transfer of secure title to the beneficiaries by supporting the strengthening of land administration institutions in the project area. (c) Capacity Building (US$6.19 million) to support public information campaigns, community mobilization, participatory rural appraisal, training and technical support to communities, district and national participating institutions and stakeholders, as well as environmental and social impact assessments and policy analysis. (d) Project Management, Monitoring and Evaluation (US$8.81 million) for overall project administration, coordination, supervision, monitoring and evaluation. 1.6 Revised Components 16. There was no revision to the Project components. 1.7 Other significant changes 17. The original Project was initially implemented in four pilot districts (i.e., Mulanje, Thyolo, Machinga, and Mangochi). In October 2008, Government requested for an amendment to the DGA to include two other districts (i.e., Balaka and Ntcheu) into the pilot project. In May 2009, a six-month extension of the Project closing date was granted from June 30, 2009 to December 31, 2009 to give time to prepare an additional financing. 4 Original Grant and additional financing. -5- 18. In October 2009, the Project benefitted from an additional grant of US$10 million to cover an over-run related to the costs of relocating 1,000 households to reach the initial target of 15,000 households, as well as to provide additional funding to the Capacity Building component to strengthen the capacity of the land administration institutions, and further support required reform of the legal framework for land administration. The additional grant came with the reformulation of the PDO and the revision of the results framework presented above. It included a new extension of the closing date to September 30, 2011. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 19. Project preparation was sound and relied on solid background analyses developed with other donors for the Presidential Land Commission of Inquiry for Land Reform in 1999. These helped prepare the Land Policy adopted by the GoM in 2002 and translated into the Malawi Land Reform Programme Implementation Strategy (2003-2007). The CBRLDP constituted one of the key components of this strategy and helped anchor the Project in the right institutional set-up and get support from the Government. 20. The approach proposed for the CBRLDP came from the experience of the “Land reform and poverty alleviation pilot project� developed in Brazil. As part of the preparation process, a Malawi Delegation, led by the Minister of Lands, visited Brazil in 2002. The Brazilian experience had achieved convincing results with the establishment of family-scale farms but on a much larger scale than proposed in the case of Malawi. It provided the core “community-driven land transfer� principle of the CBRLDP and the following implementation guidance to target land-deprived small-scale farmers 5: (i) A community-based approach to land distribution is considerably more expedient and less costly than the traditional administratively-led process; (ii) Self-selection of project participants has proven effective in pin-pointing the landless rural poor; (iii) Complementary investment sub-projects should be provided concurrently with land allocation; and (iv) Concurrent and ex-post monitoring and evaluation are critical to identify bottlenecks and make adjustments in the course of project implementation. 21. Within the Bank, the design of this critical operation was closely scrutinized, especially because the Bank was considering increased engagement in the region and was carefully monitoring what was going on in Zimbabwe at the time. It was the Bank's first redistributive land reform project in the Africa region 6 and it was the first in the world to use Bank funds for land acquisition and this triggered a change of a decades-old policy. To this effect, the project required an exception to the Bank’s prohibition on disbursement against Land (under Para 2(b) of OP 12.00 on Disbursement). This exception was granted by the Managing Director in March 2002 and a Land Committee 5 However, beneficiaries in Brazil acquire land through loans and receive grants for farm development; beneficiaries in Malawi receive grants for buying land and for farm development. 6 The first project effectively implemented, since a pilot in Zimbabwe although approved, was never implemented. -6- was established to supervise application. This Land Committee reviewed the project on July 29, 2002 and highlighted the key issues the project needed to pay careful attention to. 22. Project design also drew lessons from past Government experience with the redistribution of land bought from private estates. The most recent one was the Makande Estate which was bought by GoM from Lohnro Agribusiness Limited in 2001 for redistribution to the landless in Thyolo District. This Makande experience was jeopardized by political interference and corruption and resulted in social tensions and violence. It convinced the GoM of the need to develop a more peaceful and sustainable land redistribution process that is transparent, voluntary, legal, and resource-supported. 23. The key issues to be addressed were then clearly identified, namely the highly unequal distribution of land, land market failure, community-driven development and land administration weaknesses. Based on lessons learned from Brazil and past Malawi experiences, a set of core concepts and principles was agreed with the GoM as part of Project appraisal insisting on the decentralization process and community empowerment, the selection of land to be redistributed, the assistance package provided to resettled farmers and the legal modalities. The proposed principles aimed at mitigating some environmental risks, such as using land in protected or fragile areas, unsuitable land for agriculture, as well as social risks, such as elite capture, beneficiary self-selection and voluntary process, and integration of new settlers in communities. The critical design and implementation principles are shown in Box 1. Box 1: Core project design and implementation principles  Land distribution will take place only on farm land acquired from willing sellers, on land transferred from government administration, or on land acquired through private donations;  The project will explicitly exclude protected or fragile areas, or areas with restricted/limited agricultural potential;  Beneficiaries will be self-selected, formed in groups on a voluntary basis, and subject to pre- defined eligibility criteria;  Implementation will be decentralized through the existing District Assembly institutions, consistent with the Decentralization Policy;  Project resources for land acquisition and farm development will be transferred directly to beneficiaries and will be managed by them;  Land given to a beneficiary household should be sufficient to meet subsistence and economic viability;  Beneficiaries will decide the property regime under which they will hold the land (i.e., leasehold, freehold or customary estate);  Enhanced capacity at all levels is a pre-requisite for successful implementation of the project;  Lessons leant from the pilot districts will determine the scope of future interventions. Source: Project Appraisal Document, World Bank, 2004 24. Critical risks to Project implementation were clearly identified and relevant. The risk of failure by the GoM to timely adopt some key land policy measures (ground rents and land tax to suppress speculation in land prices) was well assessed and proved to be one of the critical elements that hampered Project implementation. The same occurred with the risk of insufficient land available for redistribution. Policy measures proposed under parallel adjustment operations were relevant but not enough to mitigate those risks. -7- 25. The capacity of the land administration to properly and timely carry out the land titling process for resettled farmers was however a bit overestimated and remains one the critical limiting factors to scaling-up. The original project design had also included the provision of social amenities to resettled households but it was decided that the Malawi Social Action Fund (MASAF) 7 should, in collaboration with the District Assemblies facilitate the provision of these amenities, through the District Development Planning Framework. Relevant linkages were established to effectively deliver through MASAF the land acquisition and farm development (LAFD) grant to relocated beneficiaries but modalities were not sufficiently clarified for the provision of social amenities. MASAF in particular had its own implementation modalities whereby local governments were responsible for prioritizing social investments. These local government priorities did not always match the project resettlement schedules and it was therefore not possible for resettled communities to quickly access additional funds to build social infrastructure. Integration of resettled communities into the local development planning structures was also slow and limited. This left relocated households facing hardships in terms of access to water, health services and education. Water supply was thus included in the LAFD fund granted to relocated households but this led in some cases to a reduction of the share available for real farm development and productive activities. Solutions were eventually found under MASAF and with other Projects and NGOs on a case by case basis so that in the end few BGs were left with no social services. In summary, the project design should have considered the decentralized approach to MASAF priority setting and agreed with local authorities on giving priority to infrastructure and social services for resettled communities, or come up with an alternative way to finance these investments through the project. 2.2 Implementation 26. The Project experienced a slow start and low disbursement for over a year due mainly to the late recruitment of Project staff, late acquisition of logistical means and poor communication. Estate owners, for instance, were initially skeptical with the Project approach and most of them doubted Government’s credibility to honour payments for land acquisition, and perceived the Project as mere Government ploy to repossess land. Misconceptions on Project objectives were amplified by the high turnover of information, education and communication (IEC) specialists. Obviously, the low rate of literacy among targeted beneficiaries did not facilitate the IEC process. To speed up implementation and improve Project perception, the decision was quickly taken to decentralize Project staff to get information dissemination and operational activities closer to beneficiaries and estate owners. The communication strategy was also revised from outsourcing to internally producing IEC tools. This quickly resulted in speeding up implementation with a first batch of land acquisition and farm development subprojects ready for execution. After the initial four beneficiary groups (BGs) had paid for their farms and relocated in November 2005, the negative perception changed and more estate owners came forward and offered their land for sale. 7 MASAF is a Social Action Fund financed by the Bank through an Adaptable Program Loan (APL) whose first phase became effective in the mid-1990s. -8- 27. Thereafter, Project implementation performance had been satisfactory but the following lingering issues remained that hampered implementation and affected Project achievements: (i) Inconsistent land policy decisions: Although the Government adopted a new land policy in 2002, the new land law which was intended to create a favorable environment for the development of land markets was never enacted. This has resulted in inconsistencies in Government’s land policy decisions, such as the moratorium on payments of the land rents, the non-application of regulations that empower the Government to repossess leased land that is underutilized or an inappropriate taxation policy of freehold land that could encourage the release of such land. To help solve this issue, the Project commissioned a land tax study to review and discuss with various stakeholders the inclusion of land tax provisions in the draft Land Bill. A new graduated land rent structure was prepared and is now implemented on all leased land, including tobacco farms, thus superseding the moratorium. The Project supported the drafting of the land bill, including extensive consultations, but its enactment is still pending. (ii) Availability of land: Poor land policy decisions led to reduced land availability for the Project. The availability of suitable land in Machinga and Mangochi Districts started declining and prices started rising, threatening Project targets achievement. In the Thyolo and Mulanje Districts, no land was proposed for transaction by estate owners8. In addition, the GoM failed to provide 4,000 ha of land as initially planned. Based on a study commissioned to explore land availability, the Project geographic coverage was expanded in October 2008 to cover two additional Districts (Ntcheu and Balaka). (iii) Weak capacities of the Land administration services: From the beginning of implementation, the Project was unable to conduct timely search in the land and deeds registries since land records at district, regional and national levels are manually-based and most of the records are in disarray. Then, the registration and titling of land transactions was extremely slow due to the lack of a computerized system. The lack of modern surveying equipment such as GPS affected the delivery of proper surveying services to the Project in particular and the public in general. Capacity constraints for surveying held up for months the issuance of group titles to Project beneficiaries (65 percent of the relocated BGs had received titles in June 2009). The Project provided training and equipment but capacity in surveying and registration are still low and remain a critical constraint to land reform and potential scaling-up. The 2009 additional financing helped fix some of the capacity gaps. (iv) Monitoring and Evaluation (M&E) deficiencies: Inadequate technical capacity at the PIU and District levels significantly delayed the operationalization of the Project M&E and the establishment of a sound Management Information System (MIS). The delayed operationalization of the MIS affected the initial data capture, 8 In these Districts, most private land is held by multi-national companies growing teas and other high- value tree crops which proved to be unwilling to release land. -9- storage and analysis and the timeliness and precision of monitoring implementation progress by project staff. It also affected their ability to provide feedback data for decision making and to propose correction to implementation, measure impacts and publish implementation reports. In addition, the first independent impact evaluator failed to produce an acceptable set of baseline data, methodology for monitoring, and evaluate Project impact. Another independent evaluator finally managed to collect data around the fourth year of Project implementation, to provide a good set of economic and financial analyses and to finalize the overall project impact evaluation. (v) Project management: Towards the mid-term of Project implementation, some issues of salaries and conditions of service emerged in addition to delayed renewal of contracts. While, the GoM took time to clarify and solve these contractual arrangements, staff morale was affected and some activities were slowed down. 2.3 M&E Design, Implementation and Utilization 28. At closure, the Project M&E system was rated satisfactory. It started from a low base and was still a bottleneck at mid-term review mainly due to inadequate technical capacity at the Project Management Unit (PMU) and District level. The initial design laid out clearly the types of data and analyses that would be required throughout Project implementation but was vague on the exact institutional set-up. More specifically, the linkage with the MIS to provide reliable data on the land acquisition and farm development was not formally designed and it took time to become fully operational. The mid-term review (MTR) mission provided key recommendations as part of an action plan to bring the MIS and the M&E systems up to full operation. 29. The project eventually managed to undertake a first beneficiary baseline survey for the first 50 BGs which had just relocated or were about to relocate in 2006. Later on, with the operationalization of the MIS, the project managed to provide regular data and reports. The project also contracted out key independent studies on topical issues including independent project impact evaluations that were useful in monitoring project implementation and results on the ground. 30. Although implementation started in 2004, the first substantial number of beneficiaries was relocated in 2005. In 2006, a first impact evaluation was conducted for the first relocated beneficiary groups. In 2007, a second impact evaluation created a panel data set of approximately 1,500 households. Follow-up surveys were further conducted in mid-2007 and in November 2008 which led to the creation of a panel data set. However, only 47 percent of the households were interviewed in the baseline survey, thus failing to create a panel data set suitable for the analysis. In 2008, the Independent Evaluation Group (IEG) of the Bank noted some data gaps and the small sample size of the households. IEG eventually contracted out an independent evaluator which helped improve the data and sample size. Then a full annual household survey was conducted for the third year, building on the baseline sample to construct a panel data set. Finally, based on the established data panel, a final independent impact evaluation was conducted in 2010 and 2011 which provides a sound economic analysis of the Project impact on relocated households. - 10 - 31. The M&E system was originally conceived as a classic Project M&E system and its integration to the Ministry of Lands, Housing and Urban Development (MoLHUD) was not clearly envisaged and anticipated. At the end of the Project, the project database and MIS have been left under the responsibility of the Department of Policy and Planning in MoLHUD for purposes of future use in the design of any scale-up land reform program as well as further medium and long term impact evaluation of CBRLDP. 2.4 Safeguard and Fiduciary Compliance 32. The CBRLDP was classified as an environmental category B operation which required partial assessment. It triggered two safeguard policies, i.e., Environmental Assessment and Pest Management, given the focus on agriculture and the expectation that farmers would use inputs, including fertilizer and pesticides. The initial Environment and Social Analysis (ESA) concluded that the Project was not expected to have any significant environmental or social impacts and would in fact contribute to the sustainable development of the rural sector in Malawi through piloting an approach to land reform which would: (i) increase the income of poor households; (ii) improve agricultural productivity and sustainable land management practices; (iii) strengthen community empowerment, inclusion, and social cohesion; and (iv) mitigate the localized tensions around the land issue. 33. Environment impact: At appraisal, a number of risks linked to the acquisition of farms that could encroach or be too close to protected areas, national parks, wetlands and other sensitive areas were identified. The Project implemented the ESSRP to ensure that farms that would be a potential environmental risk were excluded. Implementation adhered to strict environmental and social safeguards. To comply with the policies, the project implemented an environmental management plan (EMP) and identification of project sites (farms) closely followed the environmental and social screening process. As part of the EMP, an integrated pest management plan (PMP) was developed and use of these tools was facilitated through a capacity strengthening plan. 34. The project also developed a number of manuals focused on specific issues such as prevention and control of human/animal conflicts. The 2009 Bank supervision mission concluded on safeguard compliance that the screening procedure was sufficiently applied as it ensured that land acquired was free from encroachment. The ESSRP adopted the definition of safe distance from the relevant legislative provisions (Forestry Act and Wildlife Policy) for protected areas and this led to using 5-10 km, as a buffer zone between beneficiary settlements and protected areas. However, due to lapses in enforcement, there are leasehold titles, not under the project, that are issued on land that is too close to protected areas. Capacity gaps in the Department of Forestry and National Parks and Wildlife contributed to limited protection of reserves. In a few areas where this was identified under the project, corrective measures were taken during implementation. 35. Social issues: There were no major negative social impacts and conflicts as a result of relocations to newly acquired farms. The settlement approach involved purchase of farms from willing sellers and voluntary settlement of beneficiaries. All the land acquired by the beneficiaries was bought from voluntary sellers. Former estate workers had not been forced to leave the acquired farms and, in cases where they had expressed a desire to join the Project, they were subjected to eligibility criteria and allowed to join the - 11 - Project. In some cases, estate owners gave away part of the estate to surrounding communities before selling. 36. A few disputes between beneficiary groups and surrounding communities erupted over boundaries. Some surrounding communities did not initially recognize estate boundaries. The absence of beacons to mark estate boundaries exacerbated the problem as local communities took advantage of the situation. There were also disputes between the beneficiary groups and estate owners whereby some estate owners came back to claim part of the land or some trees on the already paid for land. 37. Activities were undertaken to mitigate emerging social issues such as disputes over farm boundaries and inequitable sharing of farm land, including demarcation of estates into 2 ha land parcels; reaffirmation of estate boundaries; replacement of beacons; training in conflict management; and development of guidelines for sharing vacated land. 38. Fiduciary: Overall the project adhered to terms of fiduciary compliance, and performed satisfactorily with respect to compliance to financial covenants through regular and timely submission of financial reports and annual audits. Audit Reports were submitted slightly late but were always unqualified. Despite some delays, procurement was rated moderately satisfactory to satisfactory throughout the implementation of the project. Some issues emerged with community procurement and were fixed through training, closer supervision and technical audits. 2.5 Post-completion Operation/Next Phase 39. Through its IEC program and its social approach, the Project ensured that BGs will be fully integrated in their relocation area and able to develop productive activities and interact with other communities in the neighborhood. The Agricultural Development and Marketing Corporation (ADMARC), Auction Holdings Limited (tobacco) and other produce vendors or service providers will remain present and available to the relocated households. It is expected that district councils will incorporate the needs of beneficiary groups and surrounding communities as regards to social amenities in their development plans. Furthermore, the Project supported the participating district councils to develop project proposals detailing the issues affecting Project beneficiaries and surrounding communities which could be submitted to non-governmental organizations, the Local Development Fund and other possible financiers for financial assistance. 40. Based on the promising results of this pilot experience of land acquisition and redistribution for smallholders, the GoM is willing to scale up the approach to the entire country with an objective of resettling at least 100,000 households. A concept note has already been communicated to the Bank and a feasibility study was about to start but was not initiated as it would not have been completed before project closure. 41. The Bank is willing to support this GoM’s commitment to address one of the most critical constraints to productivity increase and to agricultural growth. To keep the momentum and the capacity established under the CBRLDP while addressing some of the critical prior actions, the scaling-up strategy should be gradual and phased looking first at a quick replication of the approach to the benefit of around 20,000-25,000 households using all the CBRLDP features that have worked effectively. Looking at the unused and already encroached estates could be an option to start a follow-up operation. - 12 - 42. While contemplating the preparation of a new land redistribution project, the Bank is already committed to help the Government address some of the critical elements required for a more ambitious scaled-up investment. The recently approved additional financing to the Agricultural Sector Wide Approach – Support Project (ASWAp-SP) will for instance provide support to further strengthen the Land Administration services and to thoroughly analyze the issue of land availability, as well as other issues such as fiscal sustainability. Detailed activities will include: (i) Strengthen existing land administration structures at district, regional and national level through training, provision of equipment, cleaning of land records, modernization of land and deeds registries, and reorganizing lands registries. Training in land parcel surveying, physical planning, valuations and registration. Procurement of the required equipment such as GPS, mapping and GIS software, for field staff, as well as satellite imagery to undertake a comprehensive study of land availability in the country. (ii) Studies and analyses with a particular focus on: land availability to update the land records obtained in the mid-1990s; provision of credit for farm development as the basis for a scaled-up land reform program; and environmental and social impact assessment. 43. The Bank will also continue monitoring the adoption of the new Land Bill. Although CBRLDP proved that the voluntary and community-based approach to land reform can be implemented without the ideal policy framework, having the right land taxation incentives in place would however facilitate land release and thus increase land availability for a scaled-up operation. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation 44. The PDO was and remains highly relevant to Malawi’s overall development objectives. It has two constitutive elements of high relevance to the country: (i) increase the incomes of poor rural families; and (ii) promote a participatory approach to land redistribution as a means for the poor rural households to increase their incomes. Reducing poverty by increasing incomes remains obviously of high importance. Poverty remains widespread and concentrated in rural areas despite the good economic growth of the recent years. According to the 2007 Malawi Poverty and Vulnerability Assessment (PVA), 52.4 percent of the population lives below the poverty line and 22 percent is classified as ultra poor or unable to meet recommended food needs. 45. High population density and poverty have led to significant human pressure on the environment and degradation of Malawi’s natural resource base, notably land and forests. According to the National Statistics Office census data, population density has grown from 43 inhabitants per square kilometer in 1966 to 139 in 2008. With this growing population, customary land has become more fragmented and the land holding sizes have declined. Average land size holding per household in Malawi is 1.2 ha while the average land per capita is 0.33 ha. More than half of the agricultural households cultivate less than one hectare. The growing population increases the land area under cultivation, reduces the average land size holding and exploits forests and woodlands for - 13 - firewood and charcoal production. The resulting deforestation, run-off, flash floods, soil erosion and sedimentation, are serious threats to the environment and natural resource base for future production. Over the last 20 years, 80 percent of the production growth has been achieved through expansion of crop areas. Intensification through productivity increase as stated in the PDO is thus of high urgency as well. 46. The need to address land issues is one of the top priorities of the GoM. The Malawi Growth and Development Strategy (MGDS) recognized land as a critical factor of production and an important source of livelihood, while highlighting the numerous land challenges, including increased demand for land emanating from rapid population growth, high rate of urbanization and insufficient public awareness on land laws. The GoM’s objective is to ensure equitable access to land and tenure security; efficient management and administration system; and an ecologically balanced use of land and land-based resources. The promotion of sustainable agricultural land management practices is also one of the three strategic pillars of the GoM’s Agricultural Sector Wide Approach (ASWAp) investment framework. 47. The Project design and implementation approach were also original and relevant. The participatory and voluntary approach has proved to be appropriate to the Malawi context and highly relevant to avoid social conflicts on a highly sensitive issue. The government assisted, community based land distribution model (willing seller-willing buyer principle) is perceived as essential to a successful and sustainable land reform program as it lowers transaction cost and land prices while efficiently distributing land. The land acquisition and farm development methodology has also proved to be well elaborated and efficient to help agricultural households relocate and restart a new farm. The only major design flaw was the lack of attention to the provision of social amenities. The project was too much relying on other projects or other funding mechanisms. 48. The land reform agenda and land administration capacity strengthening remain fully relevant as well, despite the relative modest achievements on those two fronts. As generally observed in any sector and country, reforms and capacity building take time to shape and yield result on the ground. These efforts should be sustained as currently proposed under the ASWAp-SP additional financing. 49. If the Project succeeded to deliver the planned outputs and to achieve its expected outcomes (see next section 3.2), it failed to establish a fully operational and sustainable monitoring information system. Project data and analyses have not really been integrated to or incorporated by the MLHUD. 3.2 Achievement of Project Development Objectives 50. The Project has fully achieved its PDO by succeeding in increasing both the incomes and the agricultural productivity of 15,000 rural families through the proposed decentralized, voluntary community-based land reform approach. As shown in the table below, the achievement of the key performance indicators has surpassed the target values: (i) incomes of relocated families have been multiplied by 6 as compared to control groups; and (ii) yields for maize and tobacco have reached an average level 50 to 60% higher as compared to control groups in the surrounding areas; average maize and - 14 - tobacco yields have been multiplied by 4 and 2.6 respectively as compared to the previous situation of the relocated households. Table 3 – Achievement of PDO indicators Baseline Value Value Achieved at Indicator Target Values (2008) Completion Increase in real incomes of participating families as compared 30,500 Malawi to those of control groups and of 4,530 11,330 Kwacha /month such families prior to the Project. Increase in agricultural Maize: productivity of participating 450 kg/ha 1,500 kg/ha 1,800 kg/ha families as compared to that of Tobacco: such families prior to the Project. 300 kg/ha 1,000 kg/ha 800 kg/ha Source: PAD, GoM’s ICR and Final Independent Evaluation Report by Italtrend. 51. The successful achievement of the PDO stemmed from the following results and outputs attributable to the Project: (i) the successful relocation of 15,142 households (against a target of 15,000) and the effective implementation of some 666 land acquisition and farm development plans (LAFD) on 33,428 ha against a target of 33,000 ha under component 1; (ii) each household has received approximately 2.2 ha on which to farm and reside, expanding from an average land holding size of less than 0.5 hectare before participating in the Project; (iii) security of land tenure has been provided to 663 beneficiary groups against a target of 452 through the issuance of land title transfer registration; and (iv) a slight reduction in the number of days for land registration transfer thanks to training and equipment provided by the Project to the land administration services. 52. Beneficiary households reported that due to improved production of maize, their food self sufficiency improved significantly. Maize stocks lasted 3.6 months in average before relocation, but now last 10.7 months after relocation. This production increase is a direct result of the land holding size increase, but the concomitant productivity increase observed on relocated households is probably influenced by Project interventions (provision of fertilizers and improved seeds for instance) combined with other factors, such as favorable weather conditions, post-fallow cultivation and technical advice provided by extension services9. 3.3 Efficiency 53. Detailed economic and financial analyses were prepared at appraisal and revised at mid-term review to quantify the economic benefits of the project. These analyses were based on the assumption that the primary economic benefits as a result of the Project accrue from increased agricultural productivity due to the redistributed land and the use of improved crop varieties and agricultural inputs, as well as the distributional benefits 9 The Project hired six extension workers who complemented the public officers in the provision of extension services to beneficiaries as well as surrounding communities. - 15 - gained from increasing the incomes of about 15,000 poor and land poor rural families. They were based on two representative farm models related to the agro-climatic zones within the four pilot districts and anticipated to be established by smallholders under the project: subsistence farmers (80 percent) and semi-commercial farmers (20 percent). 54. Even under conservative assumptions, ignoring the indirect, non-measurable benefits of the Project (reduction of social tensions for instance), the end-of-project internal rate of return (IRR) and net present value (NPV) are higher than those estimated at appraisal (including costs for capacity building and land administration). The results indicate that the Project was financially and economically viable under all scenarios tested given the positive net present values. The economic rate of return (ERR) of 20 percent was based on the actual build-up of the 15,142 households under the Project. Table 4 – Summary of the economic and financial analysis Baseline (PAD) 2011 (ICR) Financial Rate of Return 13% 17% Economic Rate of Return 15% 20% Financial NPV US$0.7 million US$7.6 million Economic NPV US$4.1 million US$14.2 million Source: PAD and Final Independent Evaluation Report by Italtrend. 55. As the Project was piloting new approaches to land redistribution, in particular, land transactions (willing buyer and willing seller) and community-driven approaches, a high proportion of the total Project costs (47 percent - see section 57) were allocated to “overhead costs� i.e., costs not in the land acquisition and farm development component. The analysis considers the Project net benefit for the land acquisition and farm development (which was allocated 53 percent of the grant), and the results indicate that the economic benefits are much higher and significantly higher than the appraisal estimates. Table 5 – Economic and financial analysis for the land acquisition and farm development component Baseline (PAD) 2011 (ICR) Economic Rate of Return 27% 32% Financial Rate of Return 30% 38% Financial NPV US$11.9 million US$18.9 million Economic NPV US$15.1 million US$25.2 million Source: PAD and Final Independent Evaluation Report by Italtrend. 56. The Project spent 53 percent of the financial resources for land acquisition and farm development. The Project management used 32 percent of the Project resources, while 8 percent and 7 percent were spent on capacity building and land administration, respectively. The share of actual expenditures diverged substantially from the planned expenditures. Specifically, the planned share of 21 percent for project management - 16 - increased to 32 percent in the actual expenditure, with the cost for vehicle operating costs, per-diems and other travel expenses effectively doubling, and this was exacerbated by the project area being located over 200km from Lilongwe (or over 100km from Blantyre) where the project implementation team was located. As a consequence, the share of expenditure for the other three components declined. The reasons indicated for the increase in the budget allocation for Project management were mainly the rise in fuel price, in costs for hotel accommodation and for vehicle maintenance. However, it should be noted that no funds were re-allocated from the LAFD grant (because they were “ring- fenced� at design), preserving field investments and support to relocated farmer households. 3.4 Justification of Overall Outcome Rating 57. Rating: Overall project outcome is rated as satisfactory. This rating is based on the high relevance of the Project objectives and the achievement of the Key Performance Indicators, as well as a good overall economic rate of return. Two minor shortcomings have been highlighted above and can be summarized as follows: (i) The improvement of the land administration efficiency has been tangible but remains limited and far from optimal (capacity building is still required and will be provided by the ASWAp-SP additional financing); and (ii) the provision of social amenities to relocated households and the supporting policy measures have been inadequate. However, the provision of social amenities was not part of the project as per the design, and out of control of the implementation team. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 58. Targeting and beneficiary selection: Average land holding size of Project beneficiaries before relocation was 0.4 hectare, much lower than the control groups (between 1.1 and 1.5 ha) as measured by the final independent evaluation. Average maize yields of households to be relocated were also much lower than those of the surrounding families. This suggests that the Project approach to beneficiary selection has been efficient in targeting the poorest households that should be candidate for relocation. Most of those poor families had to sell their work force, after the relocation some of them became employers of occasional waged workers. 59. Effect on diversification: As a result of the increased landholding, resettled households started growing a wide range of crops, including maize, tobacco, groundnuts, sweet potatoes, cassava, and pigeonpea. Impact studies indicate that after relocation, the ownership of large livestock remained low, but that there was a rise in the proportion of beneficiary households owning chickens, with a much higher increase over time in the number of livestock units owned by households as compared to the control groups. 60. Gender: Out of the 15,142 beneficiary households, 3,687 (24 percent) were female-headed. It should be noted that participation of women in the Project was higher than this rate, as women took also part to IEC and training activities. Even though crop production and incomes were not different from those of male-headed households, female-headed households generally had problems in opening up new gardens, constructing shelters/houses, among others and they had to hire men to assist them. The - 17 - GoM’s own ICR recommends that future projects might consider providing additional resources to disadvantaged beneficiaries such as female-headed, orphan headed, elderly headed households. 61. Land disputes and tensions: It is a positive achievement that the Project did not encounter any political interference, land access disputes or other social tensions confirming the relevance of the Project approach to land acquisition based on the willing seller - willing buyer principle. Land being a sensitive matter, it was well known at Project design that movement of people from one area to another could lead to some social tensions. Extensive public awareness campaigns and training sessions contributed to avoid misconception and prevent social conflicts. All relevant institutions or structures at district level (District Councils, magistrates, the police, traditional leaders and Community Oversight Committees in both the sending and receiving areas) had benefitted from training on conflicts management and resolution. In general, the Project had been implemented with very few instances of land disputes between beneficiaries and surrounding communities and within the beneficiary groups. Outstanding boundary demarcations within and between beneficiary groups were attended to by the Project Land Surveyor, assisted by Government Land Surveyors and Physical Planners. 62. Spillover effects: According to different studies, the Project has not had significant spillover impact on the vacated areas or on the surrounding areas where beneficiaries have been relocated. It should be noted however that there was a slight increase in the average holding size from 0.6 ha to 0.9 ha as neighbors were able to take over pieces of land left behind by relocated Project beneficiaries. 63. Access to social services: As per its initial design, CBRLDP did not provide for social amenities to resettled households. After relocation, beneficiary communities were to apply to MASAF through the District Assemblies (DA) for the provision of community assets such as boreholes, access roads, schools and clinics in line with MASAF principles and criteria for demand driven development and community participation. In very few cases though, MASAF provided such community infrastructure to project beneficiaries and surrounding communities on time (see section 2.1 – paragraph 25). Solutions were found for the most pressing need for access to potable water through utilization of part of the Land Acquisition and Farm Development (LAFD) grant. The disbursement proportions for the farm development grant were changed in order to provide more funds for the first payment part of which could be used by communities to cater for the provision of potable water. As a result of this change, about 102 boreholes, 189 protected shallow wells and 28 piped water installations were provided for domestic use by beneficiaries and the surrounding communities. However, access to health facilities remains a challenge for communities which have been relocated in very remote areas. (b) Institutional Change/Strengthening 64. Land administration capacity: Throughout implementation, the surveying procedures and processing of survey data to produce deed plans was slow and seriously affected the issuance of group land title transfers. Training and survey and office equipment have been provided by the Project and capacities have improved and are expected to remain beyond Project closure. However, there is need to pursue training - 18 - activities and equipment of District services, including the full computerization of the land registries, to enhance the efficiency of land titling transactions. The Project provided financial support to MLHUD for consultative meetings and IEC campaigns related to reforms in land law. The University of Malawi received some modern survey equipment and desktop computers for use in training undergraduates in surveying, physical planning and land economy. (c) Other Unintended Outcomes and Impacts (positive or negative) 65. Recovery of bad debts: As the Project was not supposed to buy estates with encumbrances, the owners agreed that outstanding debts should be recovered at source before the transactions were completed with Project beneficiaries. As such, the Project contributed to recover more than US$1 million of arrears in land rent due to the Government and in outstanding loans due to commercial banks. 66. Land price increase: As per recommendation of the Bank’s Land Committee, project’s impact on land price changes was closely monitored and a land market study was commissioned. It confirmed that the demand created by the Project had a significant impact of the price of land over the three first years of Project implementation: while the project was developing its approach and sensitizing stakeholders, land prices first doubled from an initial level of US$50-55 per hectare then stabilized around US$110 per hectare and remained at this level until Project closure. The study recommended the inclusion of 2 additional districts (Ntcheu and Balaka) to the original four pilot districts in order to relieve the pressure on land prices. In the Mulanje and Thyolo Districts, the increase in land prices was however not sufficient or still below the reservation price to facilitate to the release of freehold land by private tea and coffee estates. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 67. There were no beneficiary surveys or stakeholder workshops but the Project prepared a compilation of beneficiary success stories that are presented in the GoM’s ICR and included in annex 4. The Malawi Government’s ICR was discussed at a broader stakeholder conference held as part of the final implementation support mission. 4. Assessment of Risk to Development Outcome 68. Rating: The risk to development outcome is rated moderate. Resettled households have received land titles and are fully secured on their new land holding. Both the 2009 IEG impact evaluation and the final independent impact evaluation provides econometric analysis that shows a significant improvement of well-being of beneficiary households, especially in land size, agricultural output, crop productivity, food security, assets holdings and agricultural income. In general, these impacts are significant but they slightly decrease over time, with the exception of assets value. 69. Resettled households’ productivity level and food security, as for any other smallholder households in Malawi, will rely on rainfall and, especially in the Southern Region on the occurrence of dry spells. A risk of bad harvest will thus persist as long as improved technologies (such as drought resistant varieties) and resilient farming techniques are not adopted by Project beneficiaries. In that sense, the on-going Bank- funded ASWAp-SP contributes to technology dissemination and a recently approved - 19 - additional financing will expand the coverage of research and extension services provided to farmers to access new technologies and sustainable land and water management techniques. 70. The econometric analyses conducted does not provide reassuring evidence with respect to the Project long term impact on durable productivity increases and total income improvement, as well as on sustainable access to social services such as schools, health and water facilities. Project beneficiaries have been resettled in remote areas on former private estates that were located far away from neighboring secondary towns or even villages. As a consequence, households are far from basic social services but also far from reliable markets and formal or informal social support networks. This may explain the limited prospect on the long run. The ASWAp-SP AF will ensure that resettled households receive adequate extension services, including market oriented advice, and will help them better connect to market opportunities through contract farming arrangements. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry 71. Rating: Satisfactory. The Bank team brought a wealth of experience and knowledge to help the GoM design this pilot experience in Malawi. Despite the complexity and sensitivity of land issues in Africa in general and in Malawi in particular, the design remained simple, well conceived and straightforward, focusing primarily on relocating poor agricultural households. The proposed government assisted, voluntary and community-based approach was innovative and quite risky for the country but it proved to be fully relevant and in the end fully efficient. The preparation phase took time (almost two years) but the Bank took the time required to agree with the GoM on the approach and to lay out some key principles for implementation. 72. The only design shortcoming relates to the inadequate attention to the provision of social services and the delayed implementation of land policy reforms. The provision of social amenities was logically left to other social projects within the framework of the overall decentralized development planning framework. However, it was clear during implementation that the District Assemblies, due to their limited resources, and existing already agreed investment priorities could not adequately cater for the development of all the project areas, and particularly for the rapid increase in investments required for the resettled areas. (b) Quality of Supervision 73. Rating: Satisfactory. The Project benefitted from a close monitoring and implementation support by the country office. Supervision continuity was provided with supervision missions twice a year, low TTL turnover and presence of a co-TTL then TTL in the country office. The Bank pro-activity and flexibility to quickly adjust Project design and implementation arrangements, and provide additional funding should be highlighted. From the early beginning, ISRs flagged key issues related to M&E, policy decision or access to water for relocated communities. - 20 - (c) Justification of Rating for Overall Bank Performance 74. Rating: Satisfactory. Based on the performance ratings at entry and during supervision, the overall Bank performance is rated satisfactory. It reflects a well designed and relevant operation with respect to country development priorities, as well as a close and professional implementation support provided to the Project. 5.2 Borrower Performance (a) Government Performance 75. Rating: Moderately satisfactory. The rationale for this rating is that although the GoM's was not entirely successful in passing specific legislation, it managed to successfully implement the project, as designed, under existing legislation. The enactment of the new land bill was dependent on political will in Parliament and therefore not entirely under the control of the Government. At the time difficult land policy issues, such as the moratorium on land tax collection or the non-application of regulations that empower the Government to repossess leased land that is underutilized, were being discussed which made the passage of any land law difficult. In fact many of the policy proposals, that could have ended up in the law were indeed not consistent with the Project orientation and requirements. Passage of a law with provisions not consistent with the requirements of the project could have been more detrimental to land reform than the status quo. 76. Within GoM, MLHUD was the official head of the Project and provided Project leadership and policy guidance on land administration and management services. It housed the PMU, assisted the Project with vehicles during the first two years and when need arose provided funds in lieu of 4,000 ha of land pledged to the Project; provided funds for a six month no cost extension; and actively participated in the joint implementation support missions. MLHUD was not always proactive in solving some emerging Project management or implementation issues (staff recruitment and conditions of services, provision of social amenities, land administration delays, etc.). The Ministry was however fully successful in implementing this complex as well as socially and politically sensitive project in a weak institutional context with limited competences on the ground at the beginning of the project. 77. The other key Malawi Government stakeholders in the Project included the Ministry of Finance, Ministry of Agriculture, Irrigation and Water Development and the Ministry of Local Government and Rural Development (MLGRD). All ministries, both at the centre and local levels, were very supportive of the Project and provided guidance that enabled the Project team to achieve commendable progress. 78. Although the GoM was in the beginning sceptical on the applicability of the proposed principles, it did manage to apply them over the course of the Project. In the end, the GoM has fully adopted the voluntary and community-based land reform approach developed under CBRLDP and fully adheres to its key principles while preparing for a scale-up operation as described above (see section 2.5). - 21 - (b) Implementing Agency or Agencies Performance 79. Rating: Satisfactory. The PMU had highly dedicated staff that was proactive and responsive to Project implementation issues. As evidenced in the Project implementation performance, it fully contributed to the PDO achievement. The PMU adhered to implementation procedures and guidelines, maintained good working relationship with both the donor and the government, successfully supervised independent Project impact evaluation, filed audited financial statements, maintained financial integrity, filed financial monitoring reports with the Bank, timely production and submission of annual progress reports. The PMU failed however to deliver an operational and efficient MIS. (c) Justification of Rating for Overall Borrower Performance 80. Rating: Satisfactory. Based on the combination of the two ratings above, overall Borrower performance is rated as satisfactory. It reflects a real Government’s willingness to facilitate land redistribution from private estates to smallholder producers based on a sound adherence to Project objectives and approaches. 6. Lessons Learned 81. Overall, the CBRLDP experience confirmed the relevance and replicability of the core project design and implementation principles (presented in box 1 page 7). The following lessons should be highlighted as key for any scaling-up operation in Malawi or similar investments in other countries in Africa: 82. The voluntary community-based approach to land reform prevents social tensions and political interferences but requires time and sound capacity building efforts: The pilot Malawi experience developed under CBRLDP confirms the relevance of the overall voluntary community-based land approach. Land reform aiming at redistributing freehold or underutilized private land to smallholder producers is feasible through this methodology that prevents political interferences and preserves social peace. It requires however, sound and large scale efforts in terms of information, education and communication to avoid misconception on land reform objectives and to prevent misinformation among stakeholders. The process was successful because it was initiated on the ground at the grassroots level not from the central level. By involving from the very early stages local authorities, traditional leaders and District-level public services, the process was fully pragmatic and efficient. Rural communities have strong social networks which make them hide some crucial information about prospective beneficiaries applying to participate. This calls for adequate lead-time to be factored in to assess the eligibility of the beneficiaries. 83. Beneficiaries prefer to relocate within or close to their original homes: It was learnt that, while beneficiaries need land, the majority preferred to relocate within their original homes or close to their original homes. One of the key factors that act as a disincentive to relocate is the need to preserve cultural and social ties that get to be weakened when the relocation happens further and across cultural lines. The importance of socio-cultural factors should not be undermined in the design of subsequent projects in order to reduce disputes that might arise from divergent cultural practices. Any replicate operation on a larger scale should avoid transgressing social boundaries between regions. - 22 - 84. The willing seller – willing buyer principle is replicable as it minimizes political interference, corruption and disputes. It facilitates the integration of the resettled households within receiving communities. Under CBRLDP experience, even though there was capacity limitation at the beginning with Land Administration services, land transactions from surveying to titling proved to be much faster than other land redistribution methods such as expropriation. 85. There is a need for a supportive policy environment and efficient Land administration services: The CBRLDP experience confirms the need for a sound conducive policy and regulatory environment. Policy measures must be consistent with the objective of facilitating or stimulating land release on the market for acquisition by smallholders. These include consistent rent collection of leased land by the Government and consistent taxation of freehold/unused land, enforcement of lease covenants to recover abandoned or under-utilized estates. A sound capacity building program for land services (training and equipment) should be prepared and implemented upfront as a pre- requisite to avoid delays with land surveying and registration that could discourage beneficiaries and reduce the credibility of the land reform. This includes a modernization of the existing land registries and clear assessment of land availability by District. 86. Land reform programs should be embedded within broader rural development support (social and technical): As described above, resettlement of smallholders occur in most cases on remote estates located far from basic social services. Coordination with projects or decentralized services that can provide social amenities is not always effective and easy to implement, nor is the integration of resettled communities is local development planning processes. Provision of social amenities should therefore be part of the resettlement package offered to relocated communities. This package should also include provision of market-oriented agricultural advice as existing public agricultural extension agents are not always available for newcomers. Experience shows also that resettled people are generally more receptive to changes and innovations and could more quickly adopt new technologies and cropping practices. Fully dedicated private extension agents as part of the resettlement package could yield quick result on productivity improvement and diversification of farming systems. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies 87. The implementation completion results report prepared by the PMU and the MLHUD on behalf of the Borrower is summarized in annex 7. 88. By letter dated March 16, 2012, the MLHUD Principal Secretary commented on the Bank draft ICR recognizing that it “has very well captured key issues pertaining to the project. Among them is the issue of how the project started, on a weak note, but ended with remarkable achievements and also that the Bank will support this Ministry to up- scale the project by funding preparatory activities like studies, training and equipment procurement through the ASWAp-SP - Additional Financing.� 89. In this letter, MLHUD however asked the Bank to consider reviewing the Government’s performance rating. It was felt that despite the weaknesses, rightly pointed - 23 - out in the draft ICR in justification of the moderately unsatisfactory rating, the GoM put efforts and provided intensive support to Project implementation. Therefore, rating government performance as satisfactory would better reflect these efforts and achievements. 90. The ICR review meeting discussed this issue and agreed that the Government’s performance rating should be upgraded to moderately satisfactory and the overall Borrower performance rating to satisfactory. As explained in section 5.2, this indeed better reflects GoM’s success in implementing a complex and sensitive Project in a weak institutional context with limited competences, despite not being fully successful in improving the policy framework. (b) Cofinanciers Not applicable. (c) Other partners and stakeholders Not applicable. - 24 - Annex 1. Project Costs and Financing (a) Project Cost by Component (in US$ Million equivalent) Additional Total Project Initial Actual/Latest Appraisal Estimate Percentage of Components Financing Estimate Costs Estimate (US$ millions) Appraisal (US$ millions) (US$ millions) (US$ millions) 1 – Land Acquisition and Farm 15.55 1.23 16.78 16.39 98% Development 2 – Land Administration 4.00 4.00 8.00 6.50 81% 3 – Capacity Building 4.26 1.93 6.19 5.22 84% 4 – Project Management, 5.97 2.84 8.81 9.84 111% Monitoring and Evaluation Total Project Costs 29.78 10.00 39.78 37.95 95% Project Preparation Facility 0.57 0.00 0.57 0.52 91% Total Financing Required 30.35 10.00 40.35 38.47 95% Source: Estimates from GoM’s ICR, Project M&E and financial management data (b) Financing Additional Financing Total Project Actual/Latest Type of Appraisal Estimate Percentage of Source of Funds Estimate Initial Costs Estimate Cofinancing (US$ millions) Appraisal (US$ millions) (US$ millions) (US$ millions) Borrower Parallel 1.50 0.0 1.50 1.09 73% Local Communities Parallel 1.28 0.0 1.28 1.28 100% IDA Grant Parallel 27.57 10.00 37.57 36.01 96% Total 30.35 10.00 40.35 38.47 95% Source: Estimates from GoM’s ICR, Project M&E and financial management data - 25 - Annex 2. Outputs by Component Component 1 – Land Acquisition and Farm Development Baseline Activities and Output Indicator Target Results Achievements Value Beneficiary farms/households 0 15,000 15,142 101% Beneficiary groups which have received appropriate 0 452 641 147% documentation of land ownership. Number of households in BGs - 15,000- 14,725 - with land titles Acquired land 0 33,000 33,428 101% Farm development proposals 0 452 666 147% (subprojects) Source: GoM’s ICR and Project M&E data Number of water point installations financed by BGs and other organizations Boreholes Shallow wells Piped water installations District BG Other BG Other BG Other sources financed sources financed sources financed Mangochi 14 33 65 - - - Machinga 46 8 117 - 17 - Balaka 0 1 2 - 9 - Ntcheu 0 - 5 - 5 - Totals 60 42 189 - 31 - Source: GoM’s ICR and Project M&E data Component 2 – Land Administration Trained staff: Masters in Land Surveying 1 Masters in Urban Planning 1 Masters in Geo-Informatics 3 Masters in Land Administration/Real Estates 1 Masters in Land Economy 1 Diploma in Land Administration 6 Certificate course in GPS/GIS Training of Trainers 30 Certificate course in database design, management and networking 27 Training in land registries management 31 Training in modern computerized valuation techniques 19 Source: GoM’s ICR and Project M&E data - 26 - List of major equipment procured and issued out Equipment Organization The Physical Polytechnic Lands Surveys Planning of the PMU Districts Total Department Department Department University of Malawi 4x4 4 0 0 0 11 6 21 vehicles Server 2 0 0 0 1 9 12 Desktops 25 11 6 10 36 12 100 UPS 25 11 6 10 36 12 100 Laptops 3 2 1 0 20 5 31 Software 0 5 0 0 2 0 5 Desk Jet 11 3 0 1 1 0 16 Printer Office Jet 1 1 2 4 Printer PABX 1 0 0 0 1 Photocopiers 7 2 1 0 10 Fax 2 0 0 0 2 machines GPS sets 0 4 1 2 1 8 Hand held 1 4 5 GPS sets Total 0 6 0 1 7 stations Theodolite 0 2 0 1 1 4 AO plotter 0 2 0 1 3 AO scanner 0 1 0 0 1 Diazo 0 4 0 0 4 printer Source: GoM’s ICR and Project M&E data - 27 - Component 3 – Capacity Building Training provided to various Project stakeholders Level and number of participants Type of training National District Community Total PRA training 5 56 0 61 Conflict resolution 10 20 0 30 Procurement methods 18 8 0 26 Team building 38 20 0 58 Ethics 38 20 0 58 Video production 5 6 0 11 MSc (various fields of the land economy) 7 0 0 7 Diploma in Land Administration 6 0 0 6 Certificate course in computerized valuation 13 6 0 19 techniques Certificate course in use of modern GIS and GPS 27 3 0 30 Training in Sun Accounting System 9 5 0 14 Certificate course in database design, 23 4 0 27 management and archiving Use of hand-held GPS 14 0 0 14 Use of Surpac survey software 10 0 0 10 Sun accounting system 7 2 0 9 Policy makers course 2 0 0 2 Financial management and Procurement 30 16 1,286 1,332 Conflict/dispute management and resolution 2 18 0 20 Conflict/dispute management and resolution 0 40 26 66 Cluster training in finance and procurement 0 0 10,483 10,483 Monitoring & evaluation 2 9 0 11 Human resource management 1 0 0 1 Secretarial management 6 0 0 6 Crop storage 0 145 12,600 12,745 Plant based pesticides 0 19 874 893 Problem animal control 0 15 1,700 1,715 Water management & sanitation 0 0 989 989 HIV/AIDS mitigation 0 0 6,000 6,000 Land redistribution 9 2 0 11 Resource mobilization and fund raising 0 45 0 45 Local Environmental Management Plan 0 40 396 436 Study tours 40 20 0 60 ICT training 11 7 0 18 Legal skills for paralegals 3 0 0 3 South - South visit 1 1 0 2 Property management 6 0 0 6 Source: GoM’s ICR and Project M&E data - 28 - Annex 3. Economic and Financial Analysis Project costs and funding sources 1- The Project internal reports indicate that the total income budget for the Project was originally US$28,958,940 with IDA financing US$27,000,000 and Malawi Government financing US$1,958,940. In December 2005, the financing arrangement for the Project was amended to allow IDA to finance 100 percent of all Project costs. This reduced the Malawi Government contribution to US$304,192. Following the amendment in financing arrangements, there was a reduction of income available to the Project by US$1,654,748 because IDA financing still remained at US$27,000,000. The GoM made the following contributions as evidence of commitment to the Project: US$304,192 before retrofitting, US$571,429 for land acquisition in lieu of the 4000 ha pledged at Project design. During the July-December 2009 period, when Additional Financing to the Project was being processed, GoM contributed US$785,714. The total Government contribution was, therefore, US$1,661,335. Hence the total budget of the original phase of the pilot Project was around US$29.0 million. 2- In 2007 the MTR mission conducted an economic and financial analysis of the Project to quantify the Project benefits based on actual information collected during the first three years of implementation. The analysis was based on the assumption that primary economic benefits of the Project accrue from increased agricultural productivity due to the redistributed land and the use of improved crop varieties and agricultural inputs, as well as the distributional benefits gained from increasing the incomes of about 15,000 poor and land poor rural families. This analysis was only a partial analysis because two important benefits had not been included: (i) the relieving of the negative externalities associated with tensions around the land issue, and (ii) the value in piloting new approaches to land redistribution and community-driven approaches. These were left out because there were minimal tensions around the land issues hence less externalities from potential tensions. 3- The analysis was based on two representative farm models i.e., subsistence farmers (80 percent) and semi-commercial farmers (20 percent), based on agro-climatic zones within the six pilot districts. The internal rate of return for each of these farm models was calculated, taking into account all financial and economic costs and benefits. The change in household income, due to own-consumption and cash sales of crop surplus were examined. The analysis aggregates from the individual farm models to determine Project benefits, based on a Project benefit build-up as beneficiary households enter the Project. The Project costs were based on the cost estimates that result from the detailed Project costing. The sensitivity of results to changes in key assumptions was analyzed to test the robustness of the results. Finally, the fiscal impact of the Project was assessed. Two farm models were considered:  Subsistence model: Eighty percent of targeted beneficiaries were expected to grow primarily food crops for their own consumption and a small proportion of cash crops. The crops and cropping pattern were expected to differ between agro-climatic zones, however the average smallholder, based on aggregate data from the six districts, was - 29 - assumed to include some quantities of local and hybrid maize, groundnuts, sweat potatoes, pigeon peas (in intercrop with maize), cassava and sorghum;  Semi-commercial model: Twenty percent of beneficiaries were expected to grow a higher proportion of cash crops in agro-climatic zones where this was possible, in addition to some food crops. Expected crops were: hybrid maize, tobacco (especially burley), sweet potato, cassava, rice, cotton, chillies and paprika. In the original model, this category comprised 10 percent of the farmers, with another 10 percent allocated to the commercial group which was assumed to grow tea. In this analysis, the two categories were combined because none of the Project beneficiaries had grown tea. Mostly, this was because the resettlement only took place in Mangochi, Machinga, Balaka and Ntcheu where agro-ecological conditions and marketing constraints, did not support the growing of tea. 4- The updated analysis was based on the same original farm models but subjected to sensitivity analysis to assess the robustness of the earlier findings. Furthermore, the analysis used the actual information collected during the five years of the Project implementation and used the actual beneficiary build-up of 15,142 beneficiaries and a total land size 33,428 ha. The other key improvement in the model related to the inclusion of livestock production, in particular, goats and chickens. The earlier analyses excluded livestock valuation because it was assumed that beneficiaries were just in the process of acquiring livestock. The inclusion of livestock valuation in the present analysis, had led to a slight change in household incomes while no changes were observed in the major economic and financial indicators. 5- It was assumed that in the first year all beneficiaries would grow hybrid maize obtained from the Project’s farm development grant. However, the land area allocated to maize was reduced from 0.6 ha to 0.5 ha reflecting the actual area that was allocated to maize by beneficiaries in the first year. The land for maize increased to 0.6 ha in the second year and remained constant over the years. The consultant diverged from the earlier analysis by conducting a sensitivity analysis taking into account the following: (i) Asset accumulation: The major asset included in the present analysis was the accumulation of livestock (poultry and goats) over the years. They focused on livestock assets due to their productive nature and potential contribution to household income. (ii) Actual beneficiary build up: An analysis was conducted based on the actual beneficiary build-up. By December 2010, 15142 households had relocated. About 455 households relocated in year one, 4067 relocated in year two, 3,700 in year three, 4,434 in year four, 1,488 in year five and 998 in year six. (iii) The different proportion of beneficiaries in each farm model: The consultant ran a model in which the proportion of semi-commercial households was adjusted from 20 percent to 30 percent. The change in the proportion of households under the semi commercial model was fairly consistent with the actual data as about 30 percent of the households grew at least one of the cash crops such as tobacco, cotton, cassava. - 30 - 6- Based on the explanations above, three key scenarios were considered in the analysis: - Baseline Scenario I: Maintaining the full target of 15,000 households, but with a 12 month extension with cost without livestock as in the earlier MTR analysis and with the same proportion of households in each of the farm models as in the MTR (20 percent in the semi-commercial model and 80 percent in the subsistence model). - Scenario II: Using the actual build-up of households of up to 15,142 with costs including livestock, with the same proportion of households in each of the farm models as in the MTR (20 percent in the semi-commercial model and 80 percent in the subsistence model). - Scenario III: Using the actual build-up of households of up to 15,142 households, but with a different proportion of households in each model (30 percent in the semi- commercial model and 70 percent in the subsistence model). Summary of key economic and financial indicators 7- The results indicated that the Project was financially and economically viable under all scenarios tested given the positive net present values; the financial and economic rates of returns were above the 12 percent threshold. The economic rate of return (ERR) based on the actual build-up of the 15,142 households under the Project returned an economic rate of return of 20 percent (the same as the 20 percent reported by the MTR) which was still above the 12 percent threshold. 8- Changing the proportion of households under subsistence and semi-commercial models produced significantly better results. When the proportion of semi-commercial households was raised to 30 percent, the economic rate of return was 22 percent which was higher than the 20 percent obtained under Scenario I. 9- Based on these indicators, scenario III seemed to be more favourable because of the larger number of beneficiaries participating in semi-commercial agriculture. Apparently this scenario was consistent with the actual situation in the Project area as 30 percent of the households participated in the growing of at least one of the commercial crops (tobacco, cotton, rice or cassava). The summary economic and financial performance indicators are presented in Table 11. Table 11: Summary results Scenarios NPV (US$ million) Rate of Return (%) Scenario I: Maintaining the full target of 15,000 households, but with a 12 month extension with cost Financial 7.3 Financial 17 without livestock (20% in the semi-commercial model Economic 13.3 Economic 20 and 80% in the subsistence model) Scenario II: Using the actual build-up of 15,142 households with costs including livestock, with 20% Financial 7.6 Financial 17 in the semi-commercial model and 80% in the Economic 14.2 Economic 20 subsistence model Scenario III: Using the actual build-up of 15,142 Financial 9.5 Financial 17 households but with 30% in the semi-commercial Economic 17.6 Economic 22 model and 70% in the subsistence model Source: Independent Impact Evaluation - Italtrend - 31 - Main Assumptions 10- The consultant adopted all the assumptions used in the earlier analysis during the midterm review which included: - The actual yield estimates from the beneficiaries as well as input and output prices had been used. - A 20 year time horizon was considered for the full Project build-up of costs and benefits based on an individual farm-level horizon of 15 years. - Current income levels (and assumed “without Project� incomes) were assumed to be US$100 per family per year, of which own consumption was a component. - Beneficiaries in the subsistence model derived a small proportion of their income from off-farm employment (mainly as ganyu labor, as is the current situation) which was come into effect by year 3, once shelter and basic infrastructure had been constructed. - It was assumed that no credit was available for smallholders. - Smallholder production yields of all crops were expected to increase by 2 percent per year as productivity performance and technology adoption increased. - Input costs include fertilizer, seeds, establishment costs and farm implements. Output prices were assumed low quality, farm gate prices. All yield data was based on the previous impact evaluation study. - Family labor was valued at the informal labor (ganyu) rate of MK10/day, the opportunity cost in the remote rural areas where alternative gainful employment was scarce. - The land price was assumed to be fixed throughout the four districts at US$175/ha. The Project was assumed not to affect the price of land. Adjustments in land tax and ground rents (expected to be part of the new Land Policy when it gets codified into law) were expected to contain any possible price escalation. Land prices had been increasing but on average, they still remained below the price assumed in this analysis. - A discount rate of 12 percent was used, in line with the figure recommended by the Ministry of Economic Planning and Development This discount rate was used to convert the stream of future net benefits to their NPV. - Significant distortions in the economy in input costs (e.g., fertilizer) and output prices were assumed to be minimal. Therefore, financial and economic costs and prices were assumed to be virtually the same (except for a minor taxable proportion of input costs). - The non-farm multiplier from the linkage effect of a change in farm cash on the local economy was assumed to be 1.5. 11- Costs accruing at the family level include the grant for land acquisition (US$315), settlement allowance (US$84), farm development (US$651) i.e. total of US$1,050. Benefits included in the analysis were: the net cash derived from crop sales (i.e., sold output less purchased inputs), family consumption of own production and a nominal valuation of the asset accumulation from the sweat equity in the shelter and basic infrastructure. - 32 - 12- Overall Project costs included the total per family grant (15,000 beneficiaries at US$1050) and “overhead costs� of land administration (US$2.1m), capacity building (US$3.5m), Project management and M&E (US$5.8m). US$100,000 per year was estimated as the ongoing overhead cost after the Project was completed. This attracted an additional cost estimated at US$3.58 million (US$2.00 million and US$1.58 million being proposed additional funds for operational expenses and LAFD component, respectively). This was included in the analysis. Therefore, the total direct Project cost was US$31.2 million. 13- This analysis had included benefits from capital accumulation such as livestock production, in particular, goats and chicken which were the dominant livestock among beneficiaries. The following benefits were not included in the analysis because they were either difficult to value, or due to unavailability of reliable data: (i) value of land improvements expected, for example, some smallholders planted fruit trees which yielded benefits; (ii) value contributed by social infrastructure constructed under MASAF i.e., schools, clinics, roads. It was not clear what number of these activities were completed and it was difficult to include them in the analysis; (iii) expected health improvements (lower morbidity and less health costs) associated with improved nutrition, water supply and sanitation; and (iv) other benefits associated with increased security of tenure, for example, increased investments in home and consequent capital accumulation. 14- The build-up of beneficiaries had been uneven across Project districts. In fact, all the settlement had occurred in Mangochi, Machinga, Balaka, and Ntcheu Districts. While a sensitivity analysis was conducted using the actual aggregate build-up of beneficiaries annually for the Project, this was not disaggregated by district. As such, the analysis had not taken into consideration the actual differential build-up across districts but rather the aggregate build-up of relocated households each year. Financial and economic benefits 15- The net financial and economic benefit from the individual farm models that represented the net benefit stream for the land acquisition and farm development component are shown in Table 12. These are based on Scenario II in which benefits from livestock were included and based on the actual build-up of beneficiaries and under the assumption of 80 percent subsistence and 20 percent semi-commercial households. The breakdown of the overhead costs was as developed in the Project costing. Project net benefits then subtract out the overhead cost from the net benefit stream from the farm models. Table 12: Benefit build-up for Project (US$ million) economic benefit Year Net benefit from financial benefit benefit land and Overhead costs Net economic land and farm development development Project net Project net administrat Proj. Man. Capacity Building Ongoing farm Subtotal & M&E o/heads Land ion 1 (1.0) (0.9) 1.38 0.56 1.87 3.8 (4.8) (4.8) 2 (2.8) (2.7) 0.50 0.76 1 2.3 (5.0) (4.9) 3 (4.1) (3.9) 0.50 0.9 0.87 2.3 (6.4) (6.2) - 33 - 4 (1.9) (1.5) 0.50 0.78 0.99 2.3 (4.2) (3.8) 5 1.6 2.2 0.50 0.7 0.99 2.2 (0.6) 0.0 6 4.8 5.6 0.1 0.1 4.7 5.5 7 5.4 6.5 0.1 0.1 5.3 6.4 8 5.7 6.8 0.1 0.1 5.6 6.7 9 5.8 7.0 0.1 0.1 5.7 6.9 10 6.0 7.2 0.1 0.1 5.9 7.1 11 6.1 7.4 0.1 0.1 6.0 7.3 12 6.2 7.6 0.1 0.1 6.1 7.5 13 6.4 7.8 0.1 0.1 6.3 7.7 14 6.6 8.0 0.1 0.1 6.5 7.9 15 6.7 8.2 0.1 0.1 6.6 8.1 16 6.9 8.4 0.1 0.1 6.8 8.3 17 7.0 8.5 0.1 0.1 6.9 8.4 18 7.1 8.6 0.1 0.1 7.0 8.5 19 7.1 8.7 0.1 0.1 7.0 8.6 20 7.1 8.7 0.1 0.1 7.0 8.6 Total 86.6 108.1 3.4 3.7 5.7 1.5 14.3 72.3 93.8 NPV 18.9 25.2 2.9 3.0 4.8 0.7 11.1 7.8 14.2 Source: Final Independent Evaluation Report by Italtrend. 16- Tables 13 and 14 show the summary results of the cost-benefit analysis. Financial and economic costs were assumed to be the same because financial costs are generally free of taxes and transfer payments. However, only financial benefits include a small component of incremental taxes (derived from tax on agricultural inputs) and the linkage effects of changes in farm income since economic benefits had to be net of taxes/subsidies and transfer payments such as interest. These benefits had little impact on the subsistence model, however were much greater in the semi-commercial. Table 11: Cost-benefit analysis results- based on Scenario II (80% / 20%) Subsistence Semi-commercial Total Project Proportions of households (%) 80% 20% 100% Number of households 12,114 3,028 15,142 Uniform grant ceiling US$1,050 US$1,050 Area per households 2ha 2ha Total area 24,227ha 6,057ha 30,284ha Land price US$175 US$175 Area under cultivation 1.5ha 1.8ha Annual revenue/ ha planted US$165 US$422 Financial analysis NPV of total costs US$1,289 US$1,289 US$13.0 NPV of total benefits US$2,366 US$3,825 US$20.6 NPV of net financial benefits US$1,078 US$2,536 US$7.6 Economic analysis NPV of incremental taxes US$14 US$15 US$0.2 NPV of linkage effects US$232 US$1,405 US$6.3 NPV of net economic benefits US$1,324 US$3,957 US$14.2 FRR 28% 42% 17% ERR 28% 55% 20% Source: Final Independent Evaluation Report by Italtrend. - 34 - Table 12: Cost-benefit analysis results based on Scenario III (70% / 30%) Subsistence Semi-commercial Total Project Proportions of households (%) 70% 30% 100% Number of households 10,599 4,543 15,142 Uniform grant ceiling US$1,050 US$1,050 Area per household 2ha 2ha Total area 21,199ha 9,085ha 30,284ha Land price US$175 US$175 Area under cultivation 1.5ha 1.8ha Annual revenue/ ha planted US$165 US$422 Financial analysis NPV of total costs US$1,289 US$1,289 US$13.0 NPV of total benefits US$2,366 US$3,825 US$22.5 NPV of net financial benefits US$1,078 US$2,536 US$9.5 Economic analysis NPV of incremental taxes US$14 US$15 US$0.2 NPV of linkage effects US$232 US$1,405 US$7.9 NPV of net economic benefits US$1,324 US$3,957 US$17.6 FRR 28% 42% 18% ERR 28% 55% 22% 17- As the Project was piloting new approaches to land redistribution, in particular, land transactions (willing buyer and willing seller) and community-driven approaches, a high proportion of the total Project costs (43 percent) are for “overhead costs� i.e., costs not in the land acquisition and farm development component. The analysis considers the Project net benefit for just this component, and the results indicate that the economic benefits were much higher. Table 13: Project net benefits for the land acquisition component Land acquisition and farm development component- 80% subsistence 20% semi- commercial based on Scenario II Financial NPV US$18.9 Million FRR 32% Economic NPV US$25.2 Million ERR 38% Land acquisition and farm development component- 70% subsistence 30% semi- commercial based on Scenario III Financial NPV US$20.8 Million FRR 33% Economic NPV US$28.7 Million ERR 40% Source: Final Independent Evaluation Report by Italtrend. - 35 - 18- The results presented in Table 15 are for scenario II (Land acquisition and farm development component- 80 percent subsistence 20 percent semi-commercial) and scenario III (Land acquisition and farm development component – 70 percent subsistence 30 percent semi-commercial). The net benefits were much higher for scenario III than for scenario II. 19- This analysis adopted assumptions proposed by the MTR team for comparisons of the present results. The MTR team found that the LAFD component, which was the key component for the Project was economically and financially viable since it generated positive financial and economic net present values and the rates of return were all well above the threshold of 12 percent. These results were consistent with the MTR findings in that the LADF component was found to generate ERR and FRR of 33 percent and 40 percent, respectively, suggesting that there was a much higher return for every dollar spent in LAFD, mainly due to the resulting higher productivity of the land. Farm income analysis 20- Increases in farm incomes in the two models for the two scenarios (80 percent subsistence and 70 percent subsistence) were quite substantial as shown in Table 16. This was mainly attributed to a sharp rise in consumption (at least compared to the initial levels). Moreover, the impact evaluation results indicated that the majority of beneficiaries had been able, even in their first year of production, to produce substantial marketable surplus for most of the crops. As such there was significant increase in cash income accruing primarily from crop sales coupled with a small proportion from off-farm labour. Table 14: Changes in family income Family Income Compared to base year Subsistence Semi- Weighted commercial average 80% subsistence 20% semi-commercial Year 3 (%) 249 261 251 Year 6 (%) 272 512 320 Year 15 (%) 352 650 411 Household consumption of own maize (kg) 669 123 Annual days family labor per farm (year 5) 116 145 70% subsistence 30% semi-commercial Year 3 (%) 249 261 253 Year 6 (%) 272 512 344 Year 15 (%) 352 650 441 Household consumption of own maize (kg) 669 123 Annual days family labor per farm (year 5) 116 145 Source: Final Independent Evaluation Report by Italtrend. - 36 - 21- Graphs in figures 1 and 2 present the rise in family income due to farm cash income, and off-farm employment over the Project life for the two household categories. Farm cash income for subsistence households sharply declined in year two of the resettlement before rising again in the third year. The sharp drop in the cash income in the second year was compensated by a rise in earnings from non-farm income while aggregate consumption continued to rise. Subsistence households also derived substantial incomes from off-farm sources throughout the years. Semi-commercial households had fairly stable incomes from farm cash in the first two years and in the long run, both subsistence and semi-commercial households benefit from the Project as portrayed by increasing farm cash incomes and aggregate consumption. Figure 1: Subsistence household model – Family income over Project life and compared without Project Total family income - subsitence Family income compared to without project $500 $500 Income (including consumption) ($) $450 $450 Family income ($/year) $400 $400 $350 $350 $300 $300 $250 $250 $200 $200 $150 $150 $100 $100 $50 $50 $0 $0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Years Years Off farm employment Farm cash Consumption With Project Without project Figure2:Semi-commercial household model – Family income over Project life and compared without Project Total family income - semi-commercial Family income compared to without project $800 $800 Income (including consumption) ($) Family income ($/year) $700 $700 $600 $600 $500 $500 $400 $400 $300 $300 $200 $200 $100 $100 $0 $0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Years Years Off farm employment Farm cash Consumption With Project Without project - 37 - Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Task Team Leader, Senior Rogier van den Brink AFTS1 TTL(appraisal) Economist First TTL, Sr Land Jorge Munoz AFTS1 TTL Administration Specialist Second TTL, Agriculture and Stanley Hiwa AFTS1 TTL Land Reform Third TTL, Land Policy and Shem Migot-Adholla AFTS1 TTL Administration Land Law and John W. Bruce Senior Counsel LEG Policy Financial Kanagasabai Loganathan Financial Management Specialist AFTFM Management Jayne Kwengwere Team Assistant AFTS1 Processing Melanie Jaya Team Assistant AFTS1 Processing Meseret Kebede Team Assistant AFTS1 Processing Francis M'Buka Social Development Specialist Social conflict Agriculture and Land Reform Caesar Chidawanyika AFTS1 Land reform Specialist Institutional Nginya Mungai Lenneiye Sr. Social Protection Specialist AFTH1 Development Hope Phillips Volket Senior Operations Officer AFTH1 Operations Development P.C. Mohan Consultant AFTS1 Communication Johnstone Nyirenda Procurement Specialist AFTPC Procurement Tesfaalem Gebreiyesus Procurement Specialist AFTPC Procurement Rural Tijan Sallah Lead Operations Officer AFTS1 Development Economic and Michael John Webster Young Professional AFTS1 Financial Analysis Aziz Bouzaher Lead Environment Specialist AFTS1 Safeguards Financial Donald Mphande Financial management specialist AFTFM Management Muthoni Kaniaru Legal LEGAF Legal Rajat Narula Sr. Finance Officer LOAG Disbursement Sylvester Kofi Awanyo Procurement AFTPC - 38 - Responsibility/ Names Title Unit Specialty Supervision/ICR Sylvester Kofi Awanyo Lead Procurement Specialist EAPPR Procurement Hans P. Binswanger Consultant AFTP1 M&E design Mary C.K. Bitekerezo Senior Social Development Spec AFTCS Safeguards Aziz Bouzaher Lead Environmental Specialist ECSS3 Safeguards John W. Bruce Consultant MNSPS Frank Fulgence K. Operational Operations Adviser AFTAR Byamugisha design Senior Rural Development Watson C. Chidawanyika AFTAR TTL Specialist Simon B. Chenjerani Senior Procurement Specialist AFTPC Procurement Chirwa Financial Fenwick M. Chitalu Financial Management Specialist AFTFM management Lungiswa Thandiwe Senior Environmental Specialist AFTEN Safeguards Gxaba AFTS1- Stanley Hiwa Sr. Agric. Economist TTL HIS Wedex Ilunga Senior Procurement Specialist AFTPC Procurement AFTS1- Ebrahim Mohamed Jassat Senior Social Development Spec Safeguards HIS Annie Kaliati Jere Team Assistant AFMMW Team support Grace Ingrid Chilambo Team Assistant AFMMW Team support Chrissie Kamwendo Senior Operations Officer AFMMW Operations Guo Li Senior Agriculture Economist AFTAR Operations Francis Kanyerere Financial Financial Management Specialist AFTFM Mkandawire management Prasad C. Mohan Lead IEC Specialist AFTDE IEC design Sr Financial Management Financial Donald Herrings Mphande AFTFM Specialist management David Rohrbach Senior Agriculture Economist AFTAR Operations Tijan M. Sallah Manager AFRCP Operations TTL(2007 - Hardwick Tchale Senior Agriculture Economist AFTAR completion) Hawanty Page Senior Program Assistant AFTAR Team support Pascal Tegwa Senior Procurement Specialist AFTPC Procurement Gert Johannes Alwyn Van Lead Financial Management Financial AFTFM Der Linde Specialist management Hope C. Phillips Volker Senior Operations Officer EASHH Operations Rogier J. E. van den Brink Lead Economist EASPR TTL(approval) - 39 - (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) USD Thousands Stage of Project Cycle No. of staff weeks (including travel and consultant costs) Lending FY02 13 64.89 FY03 11 52.87 FY04 93 380.08 Total: 117 497.84 Supervision/ICR FY05 39 118.80 FY06 39 122.62 FY07 53 146.79 FY08 51 92.93 FY09 51 0.00 .FY10 FY11 Total: 233 481.14 - 40 - Annex 5. Beneficiary Survey Results Success stories 1. Household of Ms Asiyatu James Ms. Asiyatu James of Ngatala village, Traditional Authority Chimwala in Mangochi district, is now all smiles as she can afford a peaceful sleep in her permanent house which is very clean and tidy. Being a single parent, it was only a dream that one day she would have a permanent house with corrugated iron sheets. That dream has been achieved through Kudzigulira Malo Project. Ms. Asiyatu James has five children but her main worry was how to feed the children and where to accommodate them mainly during the rainy season since her house was very temporary in nature. She was doing farming on her small piece of land but hardly produced enough to last her family a season. This compelled her to join Kudzigulira Malo Project so that she should have adequate land of her own and her children. She therefore joined other members who formed Nywenywe Trust and purchased part of Chigoti estate at Ngatala village. After getting her 2 hectare piece of land, she went straight into tobacco farming as one of the cash crops that could give her more income. Within the first season, she produced 15 bales of tobacco and every season the number of bales has not gone down below 10. This has made her to realize enough income which has helped her construct the permanent house. Annually she is able to not less than MK 150,000.00 as profit. Apart from tobacco, she is also growing groundnuts, maize, cassava and sweet potatoes to supplement her income. Food security has been attained as she has enough maize all year round. More income is also derived from animal farming and she currently has 10 goats and some local chickens. She now leads a different life compared to some surrounding households who depend on ganyu for survival due to small land holding and are not able to produce adequate food crop and income from cash crops. 2. Household of Mr. Gresam Mwechumu of Ngatuwanya Beneficiary Group Mr Gresam Mwechumu, who hailed from Mpita village, Traditional Authority Mlomba is married and blessed with twelve children. Before relocation, the family was cultivating a small piece of clan land as such it did not harvest enough for the family consumption and income. When the family heard of the intention of the estate owner to sell part of his Mpira estate through Community Based Rural Land Development Project, they agreed to participate and successfully qualified as beneficiaries. The family was amongst members of Ngatuwanya trust that relocated on Mpira estate in 2006. Using the proceeds from his two hectare plot, Mwechumu managed to buy a plot at Mlomba trading centre, where he has constructed several structures being used as dwelling house, grocery, barber shops and rented out to other traders. His compound is supplied with electricity which apart from lighting, he also uses to run his fridge. Amongst other notable assets which the family procured include Digital Satellite television set and pushing bikes. - 41 - On education, the family supports 4 children who were in various forms at secondary schools. He also sponsored one son with passport and transport to work in South Africa and currently is in the process to assist the second son with the same. On the sources of income which enabled the family to have this break-through, the family has broken the odds of considering tobacco as the main source of cash in the farming sector. The family has never grown tobacco and does not dream of doing so. The family testifies that cassava was a magic crop which enabled them to realize high income from a small piece of land. He recalls that in 2010 he was able to realize MK85,000.00 from a very small piece of land. He was all praises to the Project for introducing cassava seed multiplication and wished if it could do more to reach many beneficiaries. The family only grows enough maize for food and concentrates on cassava, livestock and selling of groceries as sources of cash. Mlomba Trading Centre has benefited from a number of beneficiary groups that used the proceeds from their 2 hectare plots to establish businesses and facilitate growth of the centre. 3. Household of Daudi Kamphepo of Mkongomwa Beneficiary Group Mr. Daudi Kamphepo came from Lucius village in Machinga District and relocated on Chigumukire estate in 2006 as one of the 17 families belonging to Mkongomwa trust. Kamphepo has one wife, 3 female and 2 male children making a total of 7. He had about 0.5 of an acre before relocation amidst other economic bottlenecks which led to perpetual household food insecurity. He claims the Project transformed his life beyond his imagination and is determined to sustainably utilize the land for greater achievement. Amongst the assets, Mr. Kamphepo has acquired because of the Project include a motor bike, television screen, push bike and a radio. He built modern house, roofed with corrugated iron sheets and is currently molding bricks to build other structures. Mr Kamphepo has livestock i.e. 10 goats, over 40 chickens, 8 guinea fowls, and 60 pigeons, courtesy of the Project On food security, he is an icon both within and outside the trust and never experiences months of food shortages since he joined the Project. For instance, in the years 2008/09 season he produced and purchased 22 and 10 tobacco bales respectively and realized close to MK500,000 from the sales of agricultural proceeds. Since 2008 the family he has been harvesting maize in excess of 100 bags of 50kg each. In 2010/11 the family harvested 60 bags of cassava, around 20bags of pigeon peas and 110 bags of maize. He, however, lamented of the low tobacco prices in the 2010/11 season. Mr. Kamphepo is paying school fees for his son who is in form one at Chikweo Community Day Secondary School. An examination of the son’s school report indicated that he is a gifted child who had been scooping position one in class. The father however said he is currently searching for a better school and claimed that he could afford school fees of up to MK30,000.00 per term. This showed that the family was capable of spending more on education. General assessment of the Kamphepo family indicated that the family had made tremendous improvement in the socio-economic status after benefiting from the land redistribution Project. The appearance of the family members also testified their good health status which was attributed to improved dietary habits. He owes his success to the prudent utilization of the land and grant provided by the Project. - 42 - 4. Access to Potable Water - Collaboration between District councils and beneficiaries. The Project provided funding for farm development to individual families of the Trusts in form of grants. The funds were particularly for purchase of land and farm inputs such as fertilizers, farm implements and hiring labor. However, the major challenge in the relocated areas has been inadequate social amenities such as potable water. The beneficiaries from Hau and Phikani clusters which relocated in 2009 have however a success story to tell about their determination in changing a seemingly gloomy water situation in the area to the present uninterrupted flow of piped water. The beneficiaries from Hau cluster comprise of Mthetsanjala, Hau and Nsipe Trusts and relocated on Riviridzi estate with a total of 61 farming families. While the Phikani cluster comprise of Nangolongonda and Riviridzi Trusts and relocated on Domwe estate with a total of 35 farming families. The unavailability of water on the newly relocated areas piled a lot of problem to the beneficiaries of both Hau and Phikani clusters. The water scarcity forced women to walk over three kilometers in order to draw water from the nearest water points and this negatively affected their planned farming activities. One day relocated beneficiaries came up with a solution to their water problem. The beneficiaries came up with a proposal of tapping water from the existing Mpira water Project which supplies water to the adjacent villages. The idea was unanimously agreed as the only best way of solving the apparent water problem in the area. As a way of implementing the proposed water Project, the relocated beneficiaries agreed to provide labour by digging trenches from the connecting point of the existing Mpira water Project pipe line (about three kilometers) to the designated supply points where stand pipes were to be erected for each Trust. The beneficiaries also agreed to take advantage of their financial grant for farming allocated to them by the Project and use part of it towards procurement of necessary materials for the proposed water Project. In order to support the brilliant idea, the District Council pledged to provide the necessary technical expertise for the success of the Project. Work at both beneficiary clusters was carried out between May and July 2010. By this time, the laying of pipes, pipe connections and erection of stand pipes at all designated supply points were successfully done. The District Council then facilitated the formation and training of water management committees so that the new facilities are sustainably managed. Currently the beneficiaries from both Hau and Phikani clusters are enjoying the usage of potable water for drinking, cooking, washing and molding bricks. Beneficiaries have clearly expressed positive sentiments regarding the successful implementation of piped water Project which has resulted in increased access to safe water and improved sanitation in the area. Beneficiaries can now access potable without hassles and the majority of them have embarked on molding of bricks with the intention of constructing permanent houses. - 43 - Annex 6. Stakeholder Workshop Report and Results Not applicable - 44 - Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR 1- Key factors affecting implementation and outcomes The Project had been successfully implemented and this was attributed to many factors at various levels: preparation, design and implementation. Project Preparation, Design and Quality at Entry The Project preparation, design and quality at entry had their strengths and weaknesses. Strengths Strong background analysis: The general Project preparation was based on well done analysis, lessons and experience from previous land reform programmes implemented in other countries such as Brazil and Malawi Government’s Makande Resettlement Programme in Thyolo district. Project design flexibility: The Project design provided for flexibility to minimize operational risks. The flexibility allowed for adjustments in the course of Project implementation in order to attain the development objectives. MASAF experience: The link to MASAF provided the opportunity to benefit from the experience and successes of MASAF which was familiar with community participation approaches to development funding. Establishment of PMU: The idea of having a PMU as an implementing agency was good. It fostered speed, commitment, quick decision making and output oriented spirit. The establishment of PMU enabled recruitment of staff who were well trained and experienced. Weaknesses Combination of land acquisition and farm development: The combination of land acquisition and farm development processes under one component exerted competing demands as personnel working under the component had to attend to both. The situation worsened when it was established that government extension system was inadequate. The demands were later on balanced by the engagement of agricultural extension and water and sanitation technicians on temporary basis and creation of additional positions of Lands Project Officers through conversion of other posts. Provision of social amenities/communal infrastructure: The Project design assumed that funding for social infrastructural needs such as potable water, primary schools, health facilities and access roads would be provided to Project beneficiaries through institutional arrangements with MASAF. This did not happen as the two programmes did not work out implementation modalities. For estates which could be made available for the Project, assessment of health, education facilities and access roads should be done prior to relocation of beneficiaries and where they are absent, they should be provided prior to relocation or soon after relocation for use by beneficiaries and surrounding communities taking into consideration sector norms. Provision for water should be part of the Project financing while other services could be financed from other sources such as government, civil society organizations, private organizations, etc. - 45 - Provision of extension, marketing and credit services: Inadequate capacity in district councils to provide extension services (agriculture, forestry, environmental management, health, water and sanitation) negatively affected utilization of Project inputs and beneficiary households’ efforts, hence compromised outcomes. The Project should have provision for extension services. Cadastral surveys: The Project implicitly assumed that all estates had clear boundaries and confirmed hectarage. However, this was not the case, therefore during implementation all estates had to be surveyed to confirm boundaries and hectarage. This brought in additional costs as the assignment was not initially budgeted for and also delayed the title transfer registration. It is recommended that budget for cadastral surveys and physical planning for group and individual land parcels should be part of future project budget. Project implementation The following are the factors which contributed to the successful implementation of the Project. Participatory approach: The Project design promoted participatory approach. Various stakeholders, including beneficiaries, traditional leaders, district officials, government ministries and departments, were actively involved in all processes of Project implementation. This promoted transparency, ownership and accountability of the Project. Donor and government commitment: There was commitment from donor and government as evidenced by timely interventions on a number of issues such as financing, Project extension, etc. Dedication of Project staff: Despite a number of challenges in Project implementation, Project implementing personnel were dedicated to see the Project to its completion. Support from various organizations/institutions: Various institutions such as district councils, non-governmental organizations, faith-based organizations, political leaders, community based organizations, and traditional leaders supported the Project implementation morally and materially. For example, some organizations provided extension services, farm inputs and credit to the beneficiaries while others provided potable water facilities like boreholes. Establishment of Project implementation unit: Establishment of the Project management unit made the staff to concentrate on Project activities although recruitment of staff was staggered putting much pressure on available staff. Capacity building initiatives: Capacity building initiatives, which ranged from provision of relevant equipment and training for various institutions including beneficiaries, traditional leaders, government ministries and departments, the Polytechnic of the University of Malawi and district councils financed by the Project contributed a lot to the success of the Project. The following section discusses challenges and how they were addressed: - 46 - Inadequate IEC messages: At the beginning of the Project implementation, there were a lot of misconceptions and skepticism about the Project. This was due to inadequate information, education and communication (IEC) campaigns due to delayed recruitment and high turnover of IEC Specialists. With assistance of the personnel from Malawi Broadcasting Corporation, the Project mounted intensive sensitization campaigns for potential beneficiaries, traditional leaders, landlords/estate owners, etc, which led to a change in attitude towards the Project. This led to high demand from potential beneficiaries and supply of land by estates owners which made the Project to meet its targets. Low literacy rate among beneficiaries:Low literacy rate among potential beneficiaries led to misconceptions about the Project and inappropriate decision making regarding use of farm development funds and technology adoption. Agricultural extension and water and sanitation technicians were engaged to assist the Project beneficiaries and surrounding communities. The Project also financed activities of non-Project extension personnel in the Project areas to ensure they adequately served both Project beneficiaries and surrounding communities. No offers of land for sale in Thyolo and Mulanje districts: While it is reported in the Project Appraisal Document that there were idle or underutilized estates in Thyolo and Mulanje, not many estate owners offered their land for sale to the Project. Three owners offered their estates to the Project but all the estates had serious encroachment issues. In addition, those who were perceived to be underutilizing it was not a guarantee that they could sell the estates to the Project because they were not asked if they could sell to the Project as such, no estate was available for the Project. Potential beneficiaries from these districts were assisted to get land in other pilot districts. This definitely contributed to the limited participation of potential beneficiaries because of long distances to relocate. Declining availability of suitable arable land: Due to declining availability of suitable farm land in Machinga and Mangochi, the Project was extended to Balaka and Ntcheu after a Land Availability, Land Market and Prices Study. This enabled the Project to reach the target of 15,000 farm families with ease. Manual lands registries: Manual operation and paper based filing and archiving of lands registries significantly delayed searches for land information. Initiatives to improve the situation such as computerization of land records of land registries and deeds registry. Performance of independent Project evaluators: The performance of one of the independent Project impact evaluators was unsatisfactory. As a fall-back position, the Project prepared an interim report and two other firms were engaged to complete the assignment. Enforcement of land policy instruments: Enforcement of policy instruments such as intensification of land rent collection and establishment of equitable land rent structure was inadequate as such they did not have positive impact regarding release of underutilized estate land. Nevertheless, the Project operated within the existing land- related laws. During the Project period, all the land-related laws, however, were reviewed by the Malawi Law Commission and a report was published. The processes to table the land bill in Parliament had started at the time of producing this ICR. - 47 - Sun Accounting System: The Project was required to use Sun Accounting System as an accounting database and Vision as a financial reporting package. However, its application was not full till the final years of the Project. The Project, however, managed to produce acceptable accounts using Microsoft Excel during the first years of implementation. 2- Safeguards and Fiduciary Compliance The Project implementation paid attention to environmental and social safeguards and complied with fiduciary requirements as discussed below. Environmental safeguards: Some of the environmental safeguards coordinated during Project implementation were: Tree planting, integrated pest management (IPM), establishment of clean cassava nurseries, promotion of environmental management plans, soil and water conservation, and human-animal conflict management. More than 800,000 trees had been planted in the Project areas. During the 2010-11, the Project distributed 255,000 tree seedlings to about 120 beneficiary groups for planting. An assessment done in April 2011 established a survival rate of 70 percent for trees planted in the 2010-11 rainy season. There were no major negative social impacts and conflicts as a result of relocations to newly acquired farms. The settlement approach involved purchase of farms from willing sellers and voluntary settlement of beneficiaries. All the land acquired by the beneficiaries was bought from willing sellers. Former estate workers had not been forced to leave the acquired farms and, in cases where they had expressed a desire to join the Project, they were allowed so that they could also benefit from the Project after meeting the eligibility selection criteria. Social safeguards: There were no major negative social impacts and conflicts as a result of relocations to newly acquired farms. The settlement approach involved purchase of farms from willing sellers and voluntary settlement of beneficiaries. All the land acquired by the beneficiaries was bought from voluntary sellers. Former estate workers had not been forced to leave the acquired farms and, in cases where they had expressed a desire to join the Project, they were subjected to eligibility criteria and allowed to join the Project. In some cases, estate owners gave away to surrounding communities part of the estate before selling. Some activities were undertaken to mitigate emerging social issues such as disputes over farm boundaries and inequitable sharing of farm land, included demarcation of estates into 2.0ha land parcels; reaffirmation of estate boundaries, replacement of beacons; training in conflict management; and development of guidelines for sharing vacated land. Procurement: Despite the initial delay in preparing and approving the first procurement plan, subsequent procurement plans were revised accordingly and appropriate procedures were followed in procuring goods and services. Financial management: The Project requested for funds from the World Bank to disburse to beneficiaries and settle obligations to suppliers and service providers on time. Finance monitoring reports (FMRs) were prepared every quarter and submitted to the Bank through the Ministry of Finance on time. The Project submitted audited financial statements and management letters to the World Bank which were generally with unqualified audit opinion within six months after the end of a financial year. - 48 - 3- Post-completion Operation/Next Phase At all times, the Project involved stakeholders at ministry, district and beneficiary group/village levels in implementing activities. In this manner, capacity was built to continue with post-completion operations. Receiving district councils incorporated the needs of beneficiary groups and surrounding communities as regards social amenities in their development plans. Furthermore, the Project supported the participating district councils to develop project proposals detailing the issues affecting Project beneficiaries and surrounding communities which could be submitted to non-governmental organizations, the Local Development Fund and other possible financiers for financial assistance. The Agricultural Development and Marketing Corporation (ADMARC), Auction Holdings Limited and other produce vendors were key buyers of farm produce and would continue to provide such marketing services. It is recognized in the Project Appraisal Document (PAD) that land redistribution was to be scaled up across the country. However, in view of the sensitivity of land issues it was felt important to pilot the land acquisition and redistribution approach on a smaller scale to learn some lessons. Now, a scale-up to the Project would be desirable as the pilot phase had generated a lot of demand for land and there were estate owners who had offered their farms for sale. The Government submitted a Concept Note on the same to the Country Office of the World Bank for consideration to fund a scale-up Project. 4- Assessment of outcomes The Malawi Poverty Reduction Strategy Paper (MPRSP), the Country Economic Memorandum (CEM) and Country Assistance Strategy highlighted land reform activities as core elements for promoting pro-poor growth. The Project development objectives were in line with all these strategies and remained very relevant as they were consistent with Malawi’s current development priorities as espoused in the Malawi Growth and Development Strategy (MGDS). The MGDS recognizes that land is the basic factor of production as well as source of livelihood for the majority of Malawians and inadequate access to land had been identified as one of the critical factors contributing to poverty in the country. Therefore, increasing secure and equitable access to land is a key component of the Government’s strategy to reduce poverty. Target beneficiary households: A total of 15,142 households had relocated against a target of 15,000, representing 101 percent; acquired 33,428 ha against 33,000, representing 101 percent; and group land title transfer registration of 641 against 666, representing 96 percent (Table 4). Each household had received approximately 2.2 ha on which to farm and reside. Security of land tenure: Out of the 15,142 households from 666 beneficiary groups, 14,725 households from 641 beneficiary groups had secure group land titles totalling 32,274.8 ha. Increased agricultural productivity: As a result of the land acquisition and farm development grants, the primary Project beneficiaries were able to realize higher crop productivity. Impact evaluation data showed that maize productivity had improved from - 49 - 456kg/ha at baseline to 1,800kg/ha and tobacco yield from 323 kg/ha to 800kg/ha, respectively after participating in the Project. Food self-sufficiency: As a result of the increased agricultural productivity, household food self-sufficiency improved from about 4 to 10 months. Household incomes: Annual household nominal incomes for beneficiary households improved from an average of MK18,700 at baseline to MK30,500 after participating in the Project. Following the change in income levels, significant household asset accumulation such as decent housing, radios and bicycles had been observed. Enhanced capacity and skills: Various survey and office equipment was provided to technical departments in the Ministry of Lands, Housing and Urban Development. This facilitated faster processing of land surveys and registration of titles. Various cadres of staff underwent formal and informal training within the country and abroad, notable ones being 7 officers who acquired master of science degrees in land-related fields universities and another six who obtained diplomas in land administration from the Natural Resources College in Malawi. Staff from participating district councils, beneficiaries and surrounding communities attended various capacity building sessions. The Polytechnic of the University of Malawi also received some modern survey equipment and desktop computers for use in training undergraduates in surveying, physical planning and land economy. Beneficiaries were also trained in various fields for smooth Project implementation. Access to potable water: Through Project grants and resources from non-governmental organizations, about 102 boreholes, 189 protected shallow wells and 28 piped water installations were provided for domestic use to the primary beneficiaries but also surrounding communities. Support to land law reform: The Project sponsored a feedback session, facilitated by the Malawi Law Commission on Report of the Law Commission on the Review of the Land-Related Laws and a consultative meeting on the Ministry’s contribution towards the second Malawi Growth and Development Strategy. Social infrastructure: The Project had no budget for provision of social amenities such as schools, access roads, boreholes. These were expected to be provided under MASAF 3. However, implementation modalities were not discussed and agreed upon as a result, there were no or inadequate social/communal amenities to cater for Project beneficiaries. As solution, part of farm development funds were used to access potable water; MASAF provided resources for primary school blocks, teachers’ house, boreholes, grinding mills, and access roads. Other organizations also came in to provide similar services. Efficiency: Results from independent Project impact evaluation showed that with about 15,142 households benefiting from the Project and 30 percent practising semi- commercial agriculture while the remaining 70 percent in subsistence agriculture, the financial rate of return (FRR) of 17 percent and the economic rate of return (ERR) of 22 percent were above the threshold of 12 percent. This meant the return was higher for - 50 - every unit of the resource (US$) spent. The Project was, thus financially and economically viable. Poverty Impacts, Gender Aspects and Social Development: Impact evaluation data showed that maize productivity had improved from 456kg/ha at baseline to 1,800kg/ha and tobacco yield from 323 kg/ha to 800kg/ha. As a result of the increased agricultural productivity, food insecurity was reduced and annual household incomes improved from an average of MK18,700 to MK30,500. Significant household asset accumulation had also been observed. The Project provided for potable water through grants on farm development and the Additional Financing. Participating district councils developed project proposals for construction of schools, access roads and environment conservation. Some non-governmental organizations provided 42 boreholes and extension services to Project beneficiaries. About 102 boreholes, 189 shallow wells and 31 piped water installations had been provided in the Project areas serving Project beneficiaries and surrounding communities. Out of the 15,142 beneficiary households, 3,687 (24 percent) were female headed. The participation rate was at par with national trends regarding participation of female headed households in voluntary development programmes. It should be noted that participation of women in the Project was higher than this rate. Female-headed households generally had problems in opening up new gardens, constructing shelters/houses, among others and they had to hire men to assist them. Future projects might consider providing additional resources to disadvantaged beneficiaries such as female-headed, orphan headed, elderly headed households. However, through visual observations, crop production and incomes were not different from those of male-headed households. Land disputes/tensions: Land being a sensitive matter, it was well known at Project design that movement of people from one area to another could lead to some social tensions. During Project implementation, land disputes were adequately monitored and attended to. In general, the Project had been implemented with minimal land disputes between beneficiaries and surrounding communities and within the beneficiary groups. Towards the end of the Project, all relevant institutions or structures at district level were taken through land conflicts management and resolution training as one way of building capacity for post Project period. The participants included officials from District Councils, magistrates, the police and traditional leaders. Outstanding boundary demarcations within and between beneficiary groups were attended to by the Project Land Surveyor, assisted by Government Land Surveyors and Physical Planners. Institutional Change/Strengthening: The Project provided financial support to the Ministry of Lands, Housing and Urban Development for consultative meetings and IEC campaigns related to reforms in land law. Furthermore, various survey and office equipment was provided to technical departments: Lands, Surveys and Physical Planning in the Ministry of Lands, Housing and Urban Development. This facilitated faster processing of land surveys and registration of titles. The Polytechnic under the University of Malawi also received some modern survey equipment and desktop computers for use in training undergraduates in surveying, physical planning and land economy. - 51 - Legal framework: The Malawi Government reviewed its land-related laws to ensure the legal framework would effectively support the implementation of the Malawi National Land Policy. The Project had provided financial support to the Ministry of Lands, Housing and Urban Development to carry out some activities related to the review of the land related laws. Other Unintended Outcomes and Impacts Recovery of bad debts: A number of estate owners had debts with various institutions/organizations such as government, financial institutions and private associations. Most of the debts were in the category of bad debts. As the Project was not supposed to buy estates with encumbrances, the owners agreed that outstanding debts should be recovered at source before the transactions were completed with Project beneficiaries. In the process, the Project helped to recover monies for these organizations. Inadequate availability of social amenities: Right from the outset, the Project was not to provide communal infrastructure or amenities. As such, communal infrastructure was to be provided through MASAF. However, during the first years of settlement, this did not work and majority of Project beneficiaries lacked social amenities. During one of the Project implementation support missions, a decision was made that part of the beneficiaries’ farm development funds should be used for provision of potable water. This improved access to potable water to beneficiaries and surrounding communities but it reduced the financial resources which were meant for farm development. In some cases, other institutions/organizations were requested to come in and assist with potable water facilities. Furthermore, the district councils incorporated the needs of the beneficiaries and surrounding communities in their development plans. Surveying of estates: During the Project design it was assumed that farm sizes and boundaries were well known through cadastral surveys. However, during implementation it was found that cadastral surveys had not been done for almost all estates. The Project had to conduct surveys for all the estates which were to be acquired by the beneficiaries. This exercise was not budgeted for. To speed up the confirmation of estate sizes a Project implementation support mission recommended the purchase of hand held Geographical Position System equipment for the districts. Assessment of outcomes: The attainment of key Project outcomes (beneficiaries, hectarage, productivity and incomes) surpassed the targets as such the achievement is rated Satisfactory. Assessment of Project risks at appraisal, mid-term and at evaluation: Mitigation measures for risks identified at the Project design were generally effective; hence, the risks were not a threat to the achievement of the Project development objectives. However, the following had been identified as risks to sustainability of Project development outcomes. 5- Assessment of Bank and Borrower performance Assessment of the Banks performance: The Bank as a development partner provided Project implementation support through financing, supervision of Project progress, - 52 - identification of implementation bottlenecks and coming up with practical solutions to address the identified constraints. The Bank provided sound Project implementation support that facilitated in turning around what would have been a problem Project. More specifically, this was demonstrated through the Bank’s pro-activity in working with Government to amend the DGA to provide practical solutions to address issues. The Bank also agreed with Government to (i) retrofit its contribution when Government was having fiscal problems; (ii) granting approval to extend the Project to two additional districts; (iii) granting of a no cost extension and provision of funds under Additional Financing; (iv) supporting the implementation of the Project through a PMU; (v) approval of converting some positions to match the workload in other areas; (vi) approval of revised procurement plans to accommodate items which were not in the original procurement; (vii) promoting flexibility of the Project design which allowed these changes to be accommodated. Despite these achievements, there were some areas on which decisions were unfavourable to the Project implementation process. These included delayed approval of the first procurement plan, recommendation to stagger employment of key Project staff, delayed effectiveness of Additional Financing, omission to include provision of social amenities and credit in the Project contrary to similar projects funded by the Bank, for example in Brazil. Overall, the performance of the Bank is rated Moderately Satisfactory. Assessment of Government performance: All ministries were very supportive of the Project and provided the leadership that enabled the Project team to achieve commendable progress. Government through MoLHUD housed the PMU for the first two years; assisted the Project with vehicles during the first two years and when need arose; provided funds in lieu of 4,000ha of land pledged to the Project; provided funds for a six month no cost extension; timely clearing of Finance Monitoring Reports; actively participated in the joint Malawi Government/World Bank Project implementation support missions; accepted supervision missions and chaired wrap up meetings; and acted on issues raised in management letters. Government departments at district level actively participated in the implementation of Project activities. The National Audit Office provided cost effective services in auditing the Project accounts. The Local Development Fund (formerly MASAF) ably managed and disbursed the LAFD Grants. There were some weaknesses on the side of Government. For example, Government dragged on the enactment of the land bill; agreed to retrofit counterpart contribution which casted doubts on commitment to the Project and lengthy recruitment procedures for Project staff. Overall, the performance of the Government is rated Moderately Satisfactory. PMU Performance: The PMU had a delegated authority from the MLHUD to manage public resources to achieve specific development objectives as agreed between the GoM and IDA and in accordance with the provisions of the Project Implementation Manual. The PMU had highly dedicated staff who were proactive and responsive to Project implementation issues as demonstrated or evidenced in the Project implementation performance. For example, the targets were met, disbursement was at 90 percent at the end of the Project, it adhered to implementation procedures and guidelines, maintained good working relationship with both the donor and the government, successfully supervised independent Project impact evaluation, filed audited financial statements with the Bank, maintained financial integrity, filed Financial Monitoring Reports with the - 53 - Bank, timely production and submission of annual progress reports. PMU had inadequate interpersonal skills, especially in the early years of the Project, though the situation improved in the later years. Having turned around a problem Project into a success story, the overall performance of the PMU is rated as Satisfactory. 6. Lessons learned The main lessons learnt for implementing the Project were as follows: � It is possible for people to relocate from one area to another area provided social amenities and support services are available or provided. � While land reform programme addresses the issue of equitable land access and subsequently increases agricultural production, incomes, food security; on its own does not improve the overall beneficiary welfare. Access by beneficiaries to social amenities and other services such as potable water, education and health facilities, passable roads, agricultural inputs and produce markets, credit, extension services should always complement land reform programs in an integrated manner. Integrated development planning with land as an entry point is the best approach to uplifting the welfare of rural communities. � Community based and voluntary land redistribution programme is feasible and cost effective compared to other forms of land redistribution as this approach promotes voluntary participation and minimizes political interference and social tensions. � Adequate capacity in land administration and management services is a pre-requisite for a land reform programme. Reliable land-related statistics should be easy and quick to extract by modernizing land registries. Adequate number of staff with relevant skills and knowledge in various fields of land administration and management services is crucial. � Land reforms require conducive policy environment and political will to support the development of land markets. The environment should have policy instruments which are easy and effective to implement to address land related issues like release of underutilized land into the land markets; and the policy instruments should be supported by a sound legal framework. It is only land reform programmes to which government is supportive legally, technically and financially that are successful. - 54 - Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Not applicable - 55 - Annex 9. List of Supporting Documents Italtrend, 2011 – Independent Project Impact Evaluation of Community-Based Rural Land Development Project – Lilongwe, Malawi Republic of Malawi, 2005 – Project Implementation Manual for the Community-Based Rural Land Development Project – Ministry of Lands, Housing and Urban Development – Lilongwe, Malawi Republic of Malawi, 2011 – Implementation Completion Results Report for the Community-Based Rural Land Development Project – Ministry of Lands, Housing and Urban Development – Lilongwe, Malawi Republic of Malawi, 2011 - The Agriculture Sector Wide Approach - Malawi’s prioritized and harmonised agricultural development agenda – Ministry of Agriculture, Irrigation and Water Development – Lilongwe, Malawi Republic of Malawi, 2010 – Malawi Demographic and Health Survey in 2010 – National Statistics Office – Zomba, Malawi World Bank, 2004-2011 – Various aide-memoires of supervision missions for the Community-Based Rural Land Development Project World Bank, 2004-2011 – Implementation Status and Results Reports for the Community-Based Rural Land Development Project World Bank, 2004 - Project Appraisal Document for the Community-Based Rural land Development Project - Report No 28188-MAI - Washington DC. World Bank, 2010 – Country Economic Memorandum – Seizing Opportunities for Growth through Regional Integration and Trade – Report No 47969-MW – Washington, DC - 56 - IBRD 39199 32°E 34°E MALAWI TANZ A NIA TAN ZA NI A COMMUNITY-BASED To Song we To Mbeya RURAL LAND DEVELOPMENT Tunduma Chitipa PROJECT AREAS Karonga THYOLO PROJECT DISTRICTS RIVERS 10°S Chisenga NTCHEU ADDITIONAL PROJECT DISTRICTS MAIN ROADS KA RO N KAR ON GA To C HI CHI T I PA CITIES AND TOWNS RAILROADS Muyombe Nykia Chilumba DISTRICT CAPITALS DISTRICT BOUNDARIES (2,606 m) REGION CAPITALS REGION BOUNDARIES Chelinda Mkondowe To NATIONAL CAPITAL INTERNATIONAL BOUNDARIES Muyombe Livingstonia Katumbe PH I R UM PHI Rumphi Ruarwe BA M ZI M B A s. Kafukule Mzuzu Mtn Euthini Nkhata NORTHERN NORT HERN Bay ZAMBIA a KH ATA N KHATA Viphy B AY Mzimba Chinteche (MALAWI) Lake Mala 12°S 12°S Luwawa To Lundazi Nkhunga wi Kaluluma KASUNGU Nkhotakota Kasungu NKHOTAKOTA MOZAMBIQUE CH IS N T CHI S I Ntchisi C E N T R A L MCHI NJ I MCHINJI DO WA DOWA Dowa S AL I M A MA Makanjila Bua To Mchinji Chipata Salima 14°S LILONGWE 14°S Namitete Monkey Bay LI L ON GW E LIL GW DE DZA D ZA To To Cuamba Furancungo MANGOCHI Dedza Mangochi To Ulongwe NTCHEU SOUTHERN To Cuamba Ntcheu Balaka ACH IN G A M ACHI N GA 0 20 40 60 80 100 Kilometers BALAKA Machinga ire Sh 0 20 40 60 Miles Lake ZO MB ZOM B A Chilwa M WAN ZA Zomba To Lirangwe Tete BLANTYRE PHALOMBE Mwanza Chiradzulu Blantyre CHIRADZULU Phalombe MULANJE 16°S Chikwawa Mulanje Sapitwa 16°S THYOLO (3,002 m) Thyolo CH I KWAWA CHI K WAWA To Liciro MALAWI N’gabu To Morire NSANJE Nsanje MOZAMBIQUE This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank To Vila de Sena Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 32°E 34°E 36°E APRIL 2012