Constructing Prospective Multiregional Input-Output Tables via Scenario-Constrained Optimization for Future Environmental Footprint Assessment

Abstract Prospective multiregional input-output (MRIO) tables that integrate scenarios from integrated assessment models (IAMs) support forward-looking environmental footprint analysis, but the trade-off between historical structural fidelity and scenario compliance has not been quantified. We constructed 12 prospective MRIO tables from EXIOBASE 2019 and the IMAGE SSP2 scenario for 2035, including a Leontief-based projection and 11 optimization-based projections under different combinations of scenario constraints on GDP, value added, sector outputs, as well as energy and agriculture consumption. Among the constraints, those on energy and agriculture consumption were satisfied at the highest cost in coefficient deviations from the Leontief-based projection and shifted the corresponding transaction flows most significantly. Regional shares of the global carbon footprint remained stable, but sectoral attribution within each region was substantially reshuffled, and the identification of high-growth consumption hotspots was sensitive to scenario compliance level. Some hotspots persisted at all levels of compliance. The consumption-based carbon footprint driven by China’s final energy demand grew the fastest and emerged as a major new hotspot in 2035. The footprint of livestock consumption showed persistent high growth in multiple regions and India and the Rest of Asia rose in environmental significance through their industrial sectors.