Contributions

Past research and lessons learned

To guide GLDC development, past research investment in GLDC crops was critically assessed based on published material and available evidence, including from Phase I CRPs³⁶. A framework of markets, institutions and policies was used to analyze case studies of high and low adoption and draw strategic lessons to guide GLDC design and implementation.

Current adoption rates of improved varieties provide the benchmark. On average, improved varieties cover about one-third of the area planted to GLDC crops in SSA and about two-thirds in India. This level of adoption has taken 30-40 years of Research and Development (R&D). Rates of adoption show large differences between regions – the average area planted to improved varieties of the nine crops in West and Central Africa (WCA) (27%) is considerably lower than in Eastern and Southern Africa (ESA) (40%). Of the area planted to improved varieties, 26% is occupied by varieties released since 2000. This suggests that, in some locations, seed systems have performed well and that models of seed delivery exist that can be replicable elsewhere. Adoption of improved varieties is accelerating for some crops such as hybrid millet in India and WCA, and for cowpea, groundnut, chickpea and pigeonpea and soybean in WCA and ESA. Rapid adoption for these crops in target countries suggests that significant spillover benefits exist for non-target countries. Since 2000, a total of 541 improved varieties have been released for the 9 crops, a median of 26 per crop, or 2 varieties per year across 5 countries in WCA, and 6 countries in ESA and India, indicating a strong pipeline for plant breeding programs. However, for sorghum and groundnuts in India, the high number of releases is not reflected in higher adoption. The reasons for this require careful scrutiny and corrective action by current breeding programs.

A conclusion is that crop improvement R&D for the semi-arid tropics can give high returns³⁷. An analysis for 10 ICRISAT interventions gave an average Internal Rate of Return (IRR) of 42% and a return of US$43 for each dollar invested. Impact assessments of microdosing in SSA and hybrid pearl millet in SA showed significant increases in yield, improving household food security³⁸. These figures indicate that adoption and impacts can be expected from continued investment through GLDC, although its research needs to identify and address the reasons for differences, learn from these experiences and test their replicability.

The performance of agronomic research has been mixed. While the full package of Conservation Agriculture or integrated soil fertility management options are not so widely adopted³⁹, the fertilizer component has been successful, with most farmers in SSA who adopt microdosing, using fertilizer for the first time. Given widespread low soil fertility, the impact of microdosing on yields is visible and immediate which explains why microdosing is popular, with adoption rates of 30% in Zimbabwe and 18% in Niger⁴⁰. Since improved varieties are fertilizer-responsive, crop management in GLDC needs to focus on increasing the adoption of fertilizer and optimizing the fertilizer-use efficiency among smallholders with limited cash resources. Integrated watershed management has increased household income by 32%⁴¹.

The commissioned report argues that the Markets-Institutions-Policy rubric is useful because the enabling environment can make or break a new technology. Success stories like pearl millet in India and cowpea in Nigeria show pre-conditions played an important role in achieving high adoption. However, while they are necessary for high adoption, they are not sufficient. Case studies of low adoption like improved pigeonpea in Malawi and conservation agriculture in Zimbabwe reinforce the importance of having the right product along with targeted uptake efforts. In addition, many lessons have already been incorporated in crop improvement programs, resulting in adjustments to breeding objectives. These include: