The unquantified cost of delay in African maize systems: an evidence gap for policy

Résumé fourni par la source

Timely planting and fertiliser application are established principles of African maize agronomy, but their acceptance has encouraged a mistaken sense that the underlying research question is closed. In rainfed African systems, operational delays reflect rainfall uncertainty, labour bottlenecks, land readiness, liquidity constraints and late input delivery. Although studies show that delay reduces yield, inconsistent benchmarks, intervals and contextual reporting prevent comparison of penalties and identification of avoidable losses. Credible estimates could make the opportunity cost of delay tangible, helping farmers prioritise competing operations while enabling programmes to distinguish late execution from other causes of poor performance. Here, we argue that the established direction of effect must be separated from unresolved questions about response shape, critical thresholds, heterogeneity, uncertainty and economic consequence, and propose a practical experimental and reporting framework that specifies: (i) the benchmark operation and reference point used to define delay; (ii) environmental context, including rainfall patterns and crop growth stages; (iii) observed absolute and relative yield contrasts, with time-normalised measures reported only where study design permits; and (iv) economic translation where agronomic outcomes have been measured. Addressing these questions requires coordinated multi-location and multi-season experiments using defined delay gradients, controlled co-treatments and links to the operational constraints experienced by farmers. The framework supports generation of context-specific evidence that can strengthen farmer decisions where timely action is feasible, identify constraints requiring institutional response and inform public investment appraisal grounded in measured and preventable loss.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.