Sensitivity and Uncertainty Analysis in MRIO Modelling; Some Empirical Results with regard to the Dutch Carbon Footprint

MRIO databases, augmented with environmental and social data of economic sectors, are an appropriate basis for the analysis of the whole life-cycle impacts of products and services across international supply chains. Environmental MRIO models require large amounts of data that all have their specific uncertainties. This paper presents a sensitivity and uncertainty analysis in order to gain an understanding of the directions in which efforts should be made to reduce the uncertainties in the outcomes of MRIO models.

Carbon footprint

The analyses were carried out for an MRIO model to calculate the Dutch carbon footprint. A sensitivity analysis of the technical coefficients showed that changes in the coefficients in the domestic blocks and in the Dutch import blocks had the largest effects on the calculated footprint. Therefore, it may be concluded that in improving input-output tables and data for a carbon footprint analysis, the data in the domestic tables and the data on imports into the region for which the carbon footprint is determined should receive most attention.

Uncertainty

The uncertainty analysis consisting of a Monte Carlo simulation based on probability distributions around the model coefficients showed a relatively low degree of uncertainty in the total Dutch carbon footprint; uncertainties in the carbon emissions allocated to regions, sectors and products were larger. In particular, sectors and products with large shares in non-CO2 greenhouse gas emissions showed great uncertainties in allocated emissions. So, future efforts in reducing uncertainties in data should focus on non-CO₂ greenhouse gas emissions.