Environmental Performance Evaluation Under the Frontier Analysis Framework: A Farm-Level Systematic Review and Meta-Analysis.

Quantifying and improving the environmental performance (EP) of agriculture has become an urgent research and policy priority. Over the past three decades, frontier-based analytical frameworks have been widely applied to evaluate farm-level EP. We present a global systematic review and meta-analysis of 121 studies (comprising 800 unique environmental performance estimates) that use a frontier-based approach for farm-level EP evaluation. Adhering to PRISMA guidelines, we extract and synthesise information on modelling approaches, estimation methods, pollutant types, as well as other data and study characteristics. We also consider the determinants for EP. We find that there is inertia in the shift towards more recent and methodologically better environmental performance evaluation techniques, such as the by-production approach and material balance methods, at the farm level. A random effects meta-regression reveals that the methodological approach and estimation methods explain the variation in EP. Farm type and inclusion of determinants are also important. We identify multiple significant determinants of environmental performance across agricultural systems. The findings advance academic understanding of how modelling frameworks shape EP estimates. They also offer practical insights to help policymakers understand specific policy variables and farm-specific factors that could be leveraged to improve environmental performance. For example, fostering pro-environmental attitudes and encouraging the adoption of sustainable agricultural practices through extension services can help enhance environmental performance in farming.