Project Background and Description
Malawi is one of the poorest countries in Africa, and as the dominant economic activity in rural areas, agriculture plays a crucial role in poverty reduction. With high population growth rates, increasing land degradation, and high susceptibility to climate shocks, Malawi will need to increase its agricultural production substantially and sustainably to feed the growing population on a shrinking per capita farm size. Yields of major staple foods are low because of the limited adoption of modern inputs, dependence on rain-fed agriculture, declining soil fertility, and inadequate agricultural extension and research.
Moreover, market opportunities are limited, and farmers are poorly connected to existing markets. Most farmers in Malawi depend on subsistence, maize-focused production systems.
The World Bank has been supporting the government of Malawi in its effort to promote sustainable growth in agricultural productivity. The Irrigation, Rural Livelihoods and Agricultural Development Project (IRLADP) supported irrigation farming through the integrated provision of hardware, mainly irrigation infrastructure, and software, mainly local and institutional capacity building. The project introduced water user associations (WUAs) for the local management of water and financial resources in the irrigation schemes. The IRLADP also supported the Input for Asset (IFA) public works program to compensate poor rural households for their labor with inputs. The Agricultural Development Program Support Project (ADPSP) addressed the efficiency of decisionmaking at the institutional agricultural policy and farm input–productivity level. At the farm level, the project supported the reform of the Farm Input Subsidy Program (FISP) and provided training and agricultural extension to members of farmer organizations.
Both projects thus supported farm productivity and the government’s ability to create an enabling environment for agribusiness.
The objective of the Project Performance Assessment Report is to assess how the farmlevel support of both projects contributed to sustainable increases in agricultural productivity among smallholder farmers (SHFs). Both projects fostered an integrated approach to increases in agricultural productivity by promoting the uptake of traditional measures to support supply (irrigation, modern inputs, and agronomic knowledge) together with complementary practices of improved land and water management. The Project Performance Assessment Report discusses two fundamental assumptions inherent in the two projects: (i) that their beneficiaries have sustained the high returns to rain-fed and irrigated farming; and (ii) that the integrated approach can be transferred to nonbeneficiaries and to areas outside the projects’ intervention sites. For continuity and scalability to occur, the government—with assistance from the private sector—was expected to take over the coordinating and implementing role of service provision to farmers, especially for public goods such as irrigation.
The projects document the successful delivery of agricultural services and the improvement of project beneficiaries’ productivity, but the evidence underpinning the productivity effects and the sustainability of those effects is weak. The IRLADP overachieved its targets for the delivery of irrigation infrastructure and capacitybuilding activities. The ADPSP achieved some but not all targets for service delivery.
Both projects reported achievements for increases in the agricultural yields of their beneficiaries, with the IRLADP substantially overachieving its targets. However, the evidence presented in the projects’ self-evaluations is inconclusive as to the sustainability of the projects’ activities. This is especially problematic for the ADPSP because no counterfactual for project activities—that is, what would have happened to the intervention area if the project was not rolled out—was established.
The Independent Evaluation Group’s assessment of the government’s production estimates does not detect sustainable increases in agricultural productivity over time or across districts in Malawi. The productivity of maize and rice was low at the start of both projects, and the productivity of both crops increased at the beginning of the projects’ implementation. Midway through the projects’ implementation, productivity flattened. Productivity gains became more volatile at the end of the projects and afterward. Overall, there have been negligible systematic improvements in maize and rice yields over the past two decades. The main shortcomings in performance are linked to the limited uptake of project activities beyond beneficiaries, the limited capacity of the government of Malawi to implement service delivery, and the limited effort to shift the mind-set of SHFs from subsistence to market orientation.
Project beneficiaries are still using and receiving advice on technologies supported by the projects, but these technologies were not scaled to nonbeneficiaries or areas outside of project sites. Beneficiaries in supported irrigation schemes continued to receive agricultural extension and use improved technologies after the projects ended.
Nationally representative data suggest that most farmers in Malawi have no access to irrigation and remain highly dependent on rain-fed farming. Moreover, despite the high coverage of extension services in Malawi, few households receive effective agricultural advice from the lead farmers supported by the projects to provide extension services.
Thus, this mode is of questionable utility. Similarly, the effective use of the complementary land and water management practices that increase the profitability of modern technologies remains low.
Neither project beneficiaries nor non-beneficiaries have been able to diversify away from subsistence-oriented maize production toward more rewarding income-generating farming. Diversification of farm systems is occurring in Malawi but arises from the necessity of food security instead of improved market opportunities. Population pressures force many SHFs to intercrop maize with legumes. Beneficiaries in irrigation sites showed a diverse production pattern of staples, legumes, and vegetables. The cash received from selling irrigated crops or legumes is typically reinvested into maize production to assure food security at home. Moreover, farmers show little progress in commercializing farming: only a limited share of the major food crops—even the higher value ones—is sold. The marketed share does not change appreciably over time.
Beneficiaries struggle to get their produce to the market when rural, unpaved feeder roads are not rehabilitated or are impassible during the rainy season. As a result, farmers commonly have no choice except to use on-the-spot transactions to sell their crops to middlemen and thus become price takers.
Design and Preparation
The design of both projects to provide productivity support through an integrated approach was innovative. Access to modern inputs, agronomic knowledge, and irrigation increases cropping intensities and productivity when yields are below agronomic potential. Complementary agricultural services maximize the profitability of agricultural inputs in maize production (ADPSP) and irrigated rice and maize production (IRLADP). The IRLADP introduced an innovative approach for capacity building and the institutionalization of irrigation management through WUAs, which continue to operate.
It is difficult to achieve sustainable growth in agricultural productivity when some project activities are primarily designed to address the food security of poor households.
Through support for the FISP and IFA, the projects targeted poor farmers to improve their food security. However, sustainable growth in agricultural productivity requires more than meeting the basic food needs of poor households. Instead, it requires continued and substantial investments to intensify farming. The poorest SHFs are less able to make these investments, as they have historically been dependent on external support to improve their agricultural activities. Moreover, the size of their irrigated plots is typically too small to allow the economies of scale required to make irrigation profitable. Both projects thus supported a continuation of subsistence-oriented farming with limited possibilities for commercialization and failed to sufficiently scope the sustainable impact of the projects in the Malawian context.
A supply-side approach to stimulate agricultural productivity did not create the right agribusiness mind-set or incentives for farmers to invest in agricultural production or to
commercialize agricultural systems. The marketing aspect of both projects was weak and limited to support of farmer organizations and selective rehabilitation of rural roads. The projects did not create an agribusiness mind-set for farming nor did they develop farmers’ nonphysical connection to markets. The projects’ design thus did not pay attention to the demand side of the agricultural sector or to how farmers should create additional value after production to tap into growing urban markets. Without access to higher-value markets, farmers lacked incentives and motivation to keep investing in crop production to optimize revenue generation. Shifting the entrenched subsistence mind-set in Malawi requires an extensive and persistent cultural engagement, which was absent from the projects’ designs.
The assumption that the government’s system for service delivery was sufficiently resourced and staffed to provide the services was unrealistic. The projects used intensive approaches to local capacity building through demand-driven extension and bottom-up WUAs for irrigation management that require sustained follow-up. The projects’ design implicitly assumed that after their completion, the government would have the capacity to continue and increase service delivery to farmers. This was not the case. For example, WUAs have not been sufficiently supported in the registration process nor in their efforts to repair irrigation infrastructure. The projects did little to break rural communities’ historical dependence on donor or government support to sustain their agricultural production.
The focus on the intensive support of specific irrigation schemes came at the expense of a comprehensive catchment or landscape approach to irrigation development and contribution to higher-level resilience. The projects promoted land and water management activities suitable for small and well-managed irrigation or demonstration plots. Although such efforts have resulted in scattered pockets of success, appropriate land and water management practices have not been expanded from project sites. The projects did not provide incentives for catchment conservation in the upper-stream parts of the water source, which could compromise the steady and sustained availability of water for irrigation. The projects also did not pay enough attention to the risks of climate variability or build resilience to these risks.
Implementation and Supervision
Both projects contributed to the provision of modern inputs and agricultural extension to poor farmers and local capacity building. The ADPSP introduced reforms to the FISP and the IRLADP used IFA work schemes to improve the access of poor SHFs to modern inputs. Moreover, activities were implemented bottom-up and used a participatory approach that contributed to building local capacity and community ownership. The
IRLADP was further successful in providing irrigation infrastructure and in strengthening WUAs for the operation of irrigation schemes.
The government’s limited capacity and resources to continue and sustain project activities resulted in limited effectiveness. The extension approach did not result in an effective sharing of agricultural information from lead farmers. The lack of government support for the repair of larger irrigation parts hampered the effectiveness of WUAs.
Consequently, some of the supported irrigation sites suffer from water leakages, breakages, or the destruction of large pipes, which the WUAs cannot afford to repair themselves. Finally, the IFA stopped when the IRLADP ended. This suggests that the government of Malawi has limited interest in sustaining a safety-net approach to irrigation development or limited capacity to do so.
Although both projects made important contributions to improving decision-making about agricultural policy, the institutionalization of these changes remains limited. Both the IRLADP and ADPSP contributed to improvements in the capacity of the Ministry of Agriculture, Irrigation and Water Development for planning and managing the agricultural sector. The ADPSP improved coordination among the multiple ministries and their departments involved in the agricultural sector. The ADPSP also harmonized donor support through a recipient-executed trust fund that increased the dialogue between the government and the donor community. Nonetheless, the institutionalization of the government’s capacity remains an issue. Government agencies do not provide adequate incentives to sustain the human and technical capacity built by projects. Thus, much of the well-intentioned support dissipated without new donor funding.
Independent Evaluation Group project ratings are described in appendix A. The evaluation methodology and evidence sources are described in appendix C.
This assessment offers the following lessons:
• An integrated and participatory approach to agricultural development can initiate sustainable productivity growth among SHFs. In the context of a SHF dominated agricultural sector and low productivity, traditional support measures of input supply are needed to close agronomic yield gaps. However, the adoption of such support measures will not be profitable for SHFs unless complementary training and extension support on proper input and land management are provided to reap the synergistic benefits.
• Agricultural projects with a supply-side focus on productivity growth that ignore market linkages are unlikely to provide the right agribusiness mind-set or incentives for farmers to sustainably invest in longer-term agricultural productivity. The projects’ integrated approach initially boosted the productivity of staple crops. However, continuous increases in agricultural productivity require a change in mind-set from semisubsistence food production to farming as a business. Sustainable productivity increases require the right economic incentives by explicitly considering the demand side of agricultural production.
• A government’s insufficient capacity and resources for agricultural sector development make it difficult to maintain an innovative but intensive demanddriven approach to service delivery in agriculture. In this case, the IRLADP introduced the WUAs for capacity building and institutionalization of irrigation management. Both projects provided extension services and organized activities in participation with local communities, but these advances need to be sustained to survive.
• Sustainable land and water management practices require a comprehensive approach that goes beyond irrigation or demonstration plots. The projects’ activities resulted in pockets of success, but the full returns to the adoption of small-scale approaches depend on what happens in upstream or surrounding parts of the agricultural landscape. Hence, a comprehensive catchment or landscape approach, especially to irrigation development, is needed to achieve higher-level sustainability.
• For projects preparing an Agriculture Sector-Wide Approach, monitoring production outcomes without a counterfactual does not allow an understanding of what is driving the anticipated productivity increases. Even if the ADPSP had sustainably increased agricultural productivity, the results framework and monitoring and evaluation system as designed would not have allowed the identification of the mechanisms driving the sustainable increases or the revisiting of project activities along the way.
José C. Carbajo
Director, Financial, Private Sector, and Sustainable Development Independent Evaluation Group