Shadow pricing of carbon emissions from agriculture using spatially adaptive reference sets

Abstract We propose a novel spatially adaptive nonparametric framework for estimating shadow prices of carbon dioxide emissions (CSP) in the agriculture sector. Our approach addresses the frequently overlooked issue of spatial heterogeneity among production units. Using a city-level panel dataset from four major agricultural provinces in China during the period 2008–2023, we define spatially adaptive reference sets based on the geographical proximity among metropolitan areas rather than assuming a single production frontier. Our study highlights the importance of considering spatial heterogeneity in environmental performance analysis and proposes a policy-relevant approach for promoting sustainable development in agriculture.