Estimates of optimal supplies of animal-sourced foods differ by food system goal and socioeconomic context

Abstract

Reduced consumption of animal- sourced food (ASF) has been recommended for environmental and human health objectives; however, ASF can be important for food security and diet affordability. We explored country- level relationships among various metrics of food supply, socioeconomic context, food security, diet affordability, and agricultural greenhouse gas emissions (GHGe) to characterize how optimal inclusion ranges for ASF vary with socioeconomic factors and food system goals. Data from 2015 to 2022 for 153 countries were sourced to generate Bayesian Networks representing relationships among the studied food system metrics. Networks were used in simulations to characterize optimal ASF inclusion to achieve environmental, food security, or diet affordability goals based on individual country data. Results are most appropriately interpreted in aggregate rather than as representations of specific countries due in large part to data limitations. Across countries simulated, median total ASF inclusion in the food supply to support food security, GHGe, or affordability objectives was 18.2% +/- 12.1%, 11.9% +/- 6.8%, and 17.6% +/- 8.5%, respectively. Joint optimization for these goals resulted in median ASF inclusion of 15.1% +/- 7.2%, with notable regional differences. Although ASF increases were supported in most developing regions, decreases were supported in developed countries. The reported SD in optimal ASF inclusion were considerable, and represented between- country variation. Empirical relationships of food categories to goals consistently favored dairy and egg products over meats. These results support previous literature highlighting the environmental intensity of ASF, but also indicate that moderate ASF supplies contribute to multiple food system goals simultaneously.