Case Study: Finer-grained footprinting to account for subnational variation
This case study examined subnational sourcing, and deforestation risk, of soy production in Brazil for animal feed associated with different components of Swedish consumption.
Overview
Traditional multiregional input-output (MRIO) based models for consumption-based accounting – including the PRINCE model – are limited to national-scale assessments of production and related environmental impacts. However, the types of environmental and health risks posed by many pressures can be very different depending on where they occur: local conditions such as water scarcity, landscape, biodiversity and population density. For such pressures, national-scale results may be of limited usefulness.
Work conducted within the PRINCE project coupled subnational-scale production, trade and land-use data with a global MRIO model to provide proof-of-concept for linking localized production and impacts to final consumption. This was achieved by advancing development of SEI’s Input-Output Trade Analysis (IOTA) model and applying it to the case of Brazilian soy production.
Using the underlying datasets from the Trase platform developed by Stockholm Environment Institute and Global Canopy, state-level Brazilian soy production and exports were linked into a global trade matrix and incorporated within IOTA to provide full production-to-consumption mapping. Brazilian soy is a relevant case to study, because the country comprises several very distinct biomes and because soy production is one of the major remaining drivers of deforestation, responding in large part to growing international demand for animal feed.
This approach combined the production-end detail offered by data-driven material flow approaches such as Trase, with the full global supply chain coverage provided by MRIO-based modelling, such as that used in PRINCE. Brazilian soy production was broken down to the level of 27 Brazilian states, while providing full sector-of-purchase estimates in final consuming countries, including Sweden.
Among other things, the results demonstrate the great variability between which parts of Brazil Sweden and EU source their soy from. They also show the variability of land-use requirements for subnational Brazilian soy production within Sweden’s sourcing profile for different sectors associated with the use of Brazilian soy for animal feed.
A better understanding the subnational origins of natural resources and emissions embedded in national consumption could provide more informative, nuanced consumption-based environmental accounts, especially if applied to more commodities and industries.