A STRUCTURAL DECOMPOSITION APPROACH TO COMPARING MRIO DATABASES

Abstract

The construction of multi-regional input–output tables is complex, and databases produced using different approaches lead to different analytical outcomes. We outline a decomposition methodology for investigating the variations that exist when using different multiregional input–output (MRIO) systems to calculate a region's consumption-based account. Structural decomposition analysis attributes the change in emissions to a set of dependent determinants, such as technical coefficients, the Leontief inverse and final demands. We apply our methodology to three MRIO databases – Eora, GTAP and WIOD. Findings reveal that the variation between Eora and GTAP can be attributed to differences in the Leontief inverse and emissions’ data, whereas the variation between Eora and WIOD is due to differences in final demand and the Leontief inverse. For the majority of regions, GTAP and WIOD produce similar results. The approach in this study could help move MRIO databases from the academic arena to a useful policy instrument.

Keywords:

• Multiregional input–output databases
• Variation
• Uncertainty
• Structural decomposition analysis
• Consumption-based accounting
• Model comparison

Acknowledgements

This work was performed while AO and KSO were visiting the Centre for Integrated Sustainability Analysis at the University of Sydney, Australia. We thank two anonymous reviewers for their insightful and constructive comments on this paper.

Funding

The work was supported by the Australian Research Council (ARC) under its Discovery Projects [DP0985522] and [DP130101293]. The contribution from AO and JB formed part of the programme of the UK Energy Research Centre and was supported by the UK Research Councils under Natural Environment Research Council award NE/G007748/1.

SUPPLEMENTAL MATERIAL

Supplemental material for this article is available via the supplemental tab on the article's online page at http://dx.doi.org/10.1080/09535314.2014.935299

Notes

denotes matrix diagonalisation.

Version 199.74.

Dietzenbacher and Los (2000) warn that analyses that decompose total value added need to be treated with care due to the dependency problem. The equation used is similar to Equation 4 with q replaced by total value-added v, and e^′c representing sectoral value added per unit of output. A decomposition equation containing three terms, e^′c, L and y assumes that Δ e^′c, and Δ L are independent. The authors point out that “changes in intermediate input coefficient and in value added coefficient affect each other” (Dietzenbacher and Los, 2000, p. 4). SDA applied to measures of consumption-based emissions require the calculation of the emissions per unit of output and this dependency issue will need to be considered. It is not appropriate to assume that a change in emissions efficiency can occur independently of the technology matrix used to calculate L.