Issue #46 - June 2007

Completion evaluation 

Following the 1992 peace settlement that ended 16 years of civil war in Mozambique, IFAD aligned its support with government efforts to develop the family-farm sector as a means of reintegrating returning refugees and displaced persons into the economy, and ensuring food security for them. IFAD decided to intervene in Niassa Province in the northwest of the country, an area characterized by very limited human and physical capacity, and poverty levels far exceeding the national average. Given the predominance of agriculture and poor state of economic and social infrastructure in the province, the Niassa Agricultural Development Project (NADP) aimed at supporting agricultural development and the construction/rehabilitation of roads and community infrastructures such as schools, health posts and water points.

Key recommendations

• More frequent supervision and implementation support missions should be fielded when interventions are implemented in conditions of extreme uncertainty (such as post-conflict and post-emergency situations) in order to assess the validity of design assumptions and introduce requisite modifications early in implementation.

• Projects in remote areas should be managed and coordinated at the local level. The feasibility of, and problems encountered in, such action will need to be assessed carefully at the design stage.

• In remote areas with limited human and institutional capital, training/refresher courses should be provided for new staff.

• Every effort should be made to identify an effective in-country approach to raising agricultural productivity, production and incomes in sparsely populated remote areas.

Targeting

Many persons left the area during the war and it was not known how many would return. Current data that would have provided an indication of the size of the population were not available at the design stage, and thus the population of the project area was overestimated by 100 per cent. The appraisal report estimated that about 45,000 households would be farm-dependent and poor, thereby constituting the initial target group, and would increase to 52,600 by the end of the project period. This was a serious discrepancy: when IFAD revisited the project design late in the implementation period, the target group had been reduced to 30,000 households. This not only had an impact on the targets set for components/sub-components but also on the economic rate of return, which at appraisal was optimistically estimated at 16 per cent.

Closer is better

Although centralized management/coordination was not foreseen in the original project design, for practical reasons these functions were transferred from Maputo to Lichinga only in 2001. Thus, for a considerable period of time, project management and coordination were directed from the capital, far from the project area. This had a negative impact on implementation performance.

Training is important

Despite an adequate training budget under NADP, this was not sufficiently utilized and little was done at the outset to define implementation rules and procedures or to provide training. In post-conflict and post-emergency situations, there is a natural urge to deliver on-the-ground as quickly as possible and therefore jump into implementation without first establishing a solid foundation. Investments in staff training may be considered as ‘delivery costs': for a multi-sector project in a remote area, working with numerous institutions with limited capacity, these costs are considerable but necessary for achieving impact.

Technology improvements may be difficult to introduce

The evaluation analysed the sequencing and balancing of different sub-components with regard to support to agriculture under the project. Input and produce markets hardly existed before the project; therefore, it may have been more relevant to concentrate on developing markets rather than research and extension when there were no markets to supply the recommended inputs and absorb incremental produce. This raises a general strategic issue: in areas with limited market development, should major investments in agricultural extension be postponed until such time as the market can supply the produce that it is recommended the farmers use? Therefore, it is not a question of the relevance of research and extension per se but of the most relevant sequence when prioritizing investments. Project design anticipated that input markets would develop with the assistance of NADP, among others. However, while markets for agricultural produce have been developed to some extent, in the rural areas of Niassa there is still a lack of improved seed, fertilizer, and agricultural tools and equipment. As such, given the Government's limited resources for research and extension, it may have been more relevant (and cost-effective) to accord priority to commercial farming areas where there was a demand and where inputs and technologies were available for improving agricultural productivity and income.

Another explanation may be that there was no market failure but rather a negligible demand for inputs and new technologies. Even with perfect markets, it was far from certain that farmers would move towards more intensive land use and apply more advanced technologies. It may well be that, given the abundance of land in Niassa, the farmers believed they were making the best choice: an external project coupled with research/extension could do little to influence such choices.