Updating yield growth rates in the IMPACT model to enhance global analysis of the future of agriculture
Changes in yields of food commodities are key drivers—and consequences—of changes in agriculture and food systems. Simulation models used to explore the future of agriculture require initial assumptions about baseline levels and changes in yields as a starting point and then can be used to analyze how those yields are affected by the dynamic interaction between changes in demand, markets, technology, climate and other factors. This paper reports on how critical initial yield assumptions have recently been reviewed and updated in the International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT). This paper is aimed at a technical audience interested in core details of the IMPACT modeling framework and is a follow-up to the latest full model documentation (Robinson et al 2024).
IMPACT was developed at the International Food Policy Research Institute (IFPRI) at the beginning of the 1990s to address a lack of long-term vision and consensus among policymakers and researchers about the actions necessary to feed the world in the future, reduce poverty, and protect the natural resource base. Over time, this economic model has been expanded and improved, and IMPACT is now a system of linked models around a core multimarket economic model of global production, trade, demand, and prices for agricultural commodities. IMPACT supports integrated analysis of the implications of physical, biophysical, and socioeconomic trends and phenomena, allowing for varied and in-depth analysis on a variety of key issues of interest to policymakers. As a flexible policy analysis tool, IMPACT has been used to research linkages between agricultural production and food security at the national and regional levels. IMPACT also has been used in commodity-level scenario analyses and has contributed to thematic and interdisciplinary scenario-based projects. IMPACT is one of major global models of the agriculture sector, but because of its high level of commodity and geographic disaggregation, it plays a particularly important role in analyzing patterns of agricultural productivity growth around the world.
Potential commodity yields are key drivers in the IMPACT baseline scenario and an important lever influenced by investment and policy scenarios. The baseline assumptions about exogenous growth in crop and livestock yields are referred to in IMPACT as “intrinsic productivity growth rates” (IPRs). This paper describes the approach developed to review and update the IPRs to reflect recent trends and incorporate expert1 opinion, making the yield projections more transparent, inclusive, and efficient, and more consistent across CGIAR centers, improving projections and enhancing policy relevance. The remaining sections include a description of the IMPACT Model, the methods for updating the IPRs using projections based on FAOSTAT data and expert consultations, the findings related to the IPRs from each of the expert consultations, and a description of the process used to incorporate the findings from the consultations into IMPACT and concludes with discussion. A description of the Power BI tool used to support the expert consultations is presented in Appendix A.
Authors
Hareau, G.; Rosegrant, Mark W.; Andrade, Robert; Petsakos, Athanasios; Sulser, Timothy B.; Chamberlin, Jordan; Sonder, Kai; Kihiu, Evelyne; Suarez, V.; Cenacchi, Nicola; Dunston, Shahnila; Hartley, Faaiqa; Mishra, Abhijeet; Thomas, Timothy S.; Alene, Arega D.; Gbegbelegbe, Sika; Mjuma, Andrea; Enahoro, Dolapo K.; Petnkeu, Jeronya Mbiatat; Antonio, Ronald Jeremy S.; Cabrera, Ellanie; Basnet, Shyam; Gimutao, Maureen; Pede, Velerien; Chan, Chin Yee; Chung, Ka Yi Karen; Sari, Evren Can
Citation
Hareau, G.; Rosegrant, Mark W.; Andrade, Robert; Petsakos, Athanasios; Sulser, Timothy B.; et al. 2025. Updating yield growth rates in the IMPACT model to enhance global analysis of the future of agriculture. IFPRI Modeling Systems Technical Paper 2. Washington, DC: International Food Policy Research Institute. https://hdl.handle.net/10568/178821
Keywords
Modelling; Yields; Growth Rate; Agriculture; Productivity; Crop Yield
