Revisiting Decomposition Analysis for Carbon Dioxide Emissions from Car Travel

Decomposition analyses are helpful to policy makers and analysts who aim to reduce carbon dioxide (CO 2 ) emissions from car travel. So far, a large number of decomposition methods have been proposed. However, there is still no consensus about the best decomposition method because each method has certain advantages and disadvantages. Which method is valid for the decomposition of the changes in CO 2 emissions from car travel? This study revisited the refined Laspeyres index (RLI), logarithmic mean Divisia Index I (LMDI I), and modified Laspeyres index (MLI) methods. After a discussion of theoretical issues, period, time-series, and cross-regional decompositions of the changes in CO 2 emissions from passenger cars in Japan were conducted with the three methods. Although the RLI and LMDI I methods have been most widely used by researchers and analysts, these methods contain theoretical problems with the attribution and distribution of inter action terms, particularly when some factors change positively but others change negatively. The recently proposed MLI method helps in resolving those issues by attributing and distributing the inter action terms to related factors according to the changes in each factor. The case studies showed that the differences in the attribution of the inter action terms to the related factors of the three methods affected the decomposition results significantly. The MLI method generates more valid decomposition results than do the RLI and LMDI I methods because of the reasonable attribution and distribution of the MLI method's interaction terms.