Though Africa is a continent unrivalled in its amount of resources and capacity to feed not only the continent, but also the world, its food import bill amounts to US$36 billion every year, excluding fisheries. Furthermore, a key finding of the 2014 Africa Progress Report states that “African countries can reduce poverty and inequality by boosting agriculture, which affects two thirds of the continent’s population”.
With more than half the total employment of sub-Saharan Africa located in agriculture, it is an important sector for the medium to long-term development plans of governments. Growth in agriculture does not only lead to increased GDP, but also to poverty alleviation. Various country contexts have shown that agriculture growth leads to more poverty-alleviation than growth in any other sector.
There is a growing global trend towards evidence-based policy decisions. It highlights the overall understanding of a need to reduce the risk of continuing to spend excessively on an intervention (e.g. fertilizer subsidies, rural land development projects) that may not result in the expected outcomes. Strategic decisions need to be taken in order to invest in those projects which achieve the most bang for buck and impact positively on a country’s population.
This is where CABRI's work on value for money in sector financing comes in. Most recently, our work on agriculture has looked at tools to assess value for money in agriculture spending in Africa. Evaluating the impact of agriculture financing interventions strengthens the ability of governments in knowing where it is best to invest, supports in better motivating for funds (both internally and also externally), and a necessary input into policy-making. Primarily, the aim of agriculture investments by governments are threefold: (i) increase food security, (ii) reduce dependency on imports, and (iii) increase farmer incomes.
The 2nd agriculture dialogue brought together budget officials and representatives from agriculture ministries from 13 African countries to learn, discuss and share experiences on impact evaluation tools.
Impact evaluations are different from our understanding of Monitoring and Evaluation (M&E). Monitoring is a continuous process during which inputs, outputs and activities are tracked. Evaluation is the periodic review of a programme, which is either planned, ongoing or complete. Impact evaluation, on the other hand, focuses on asking and finding answers to questions around cause-and effect. This helps policy-makers evaluate and also correct interventions at different stages of a project cycle.
- Randomised controlled trials
- Difference in differences
- Instrumental variables estimation
- Propensity score matching
- Regression discontinuity design
An impact evaluation plan, detailing programme objectives helps to build a results chain, find the right performance indicators, and identify the correct evaluation methodology.
Impact evaluations in the agriculture sector sometimes seem to be more challenging than those in other sectors. Measuring agricultural impact is particularly challenging due to externalities or spill over effects of the agricultural intervention (e.g. subsidies). This needs to be taken into account at the evaluation level. A simple comparison of the users of an intervention versus the non-users would not reflect the programme impact, since spill-overs are easily achieved (though positive spill-over effects are desirable, they do complicate the process of measuring the impact of an evaluation). This can be demonstrated by a transfer of knowledge from one farmer in the evaluation region to another in a different region (control group). It can also be the effect a programme has on another crop that was not targeted. Though important to take into account, indirect effects are hard to extricate.
Additionally, being able to determine whether the results of a certain intervention are valid for a different context (external validity) is a challenge. Though this is a common fate for all impact evaluations, factors like geographic elevation, rainfall or soil play a big role in influencing agricultural outcomes that have nothing to do with the intervention.
The more favoured approach to a systematic approach within government to impact evaluations, involves a central government agency taking responsibility and integrating impact evaluations into the monitoring and evaluation processes. But it takes time and resources to set up such a system, with clear guidelines on procedures and methodologies. Covering for the costs of (sometimes quite elaborate) impact evaluations are difficult to justify, especially when introducing these in a sector where the attribution gap can be quite significant. Here, it is advisable to design an evaluation around a pilot project, before undertaking a whole-of-government exercise. If the budget permits, it would be a good idea to do a mid-term evaluation (intermediate outcomes) in addition to a final evaluation, which will determine whether the programme achieved its ultimate goal.
Though self-funded impact evaluation projects are preferable for reason of ownership, countries can explore support from external (donor) agencies. In fact, involving donors for ad hoc evaluations is one way that can lead to the institutionalisation of impact evaluations on the government level, with country institutions getting more and more involved per assessment and standardising procedures and methodologies along the way.
The Agriculture Sector Dialogue is the fourth in a series of sector dialogues. Previous CABRI dialogues have focused on the health, education and infrastructure sectors. Policy officials from governments and international experts from various organisations across the globe engage in roundtable panel discussions and group work to examine specific challenges and agriculture financing mechanisms. The discussions are based on a country case-studies undertaken by CABRI. The 3rd Agriculture financing dialogue will focus on a thematic discussion of the various issues that inhibit or promote good budgeting in the agriculture sector.