Decomposition of CO2 Emission Drivers and Energy Transition Dynamics Across Australian States: A Kaya–Shapley Approach

Abstract

This study employs the Kaya identity and Shapley value decomposition to analyse the main drivers of CO2 emissions per capita in Australia and its states/territories from 2008–2009 to 2023–2024. This study extends the traditional three-factor Kaya model to a five-factor framework that explicitly incorporates renewable energy dynamics. Within this environment, we introduce an index for evaluating energy transition in each state over time, captured through the combined effects of renewable energy share and the fossil fuel-to-renewable ratio. The results indicate that energy intensity improvements are the dominant driver of emission reductions, contributing well over 100% of the total decline in most jurisdictions. However, this pattern does not hold universally: in the Northern Territory, energy intensity increased and contributed positively to emissions. In contrast, economic growth continues to exert upward pressure on emissions, partially offsetting these reductions and highlighting the ongoing challenge of decoupling. The Energy Transition Index (ETI) rose more than fivefold nationally (0.0018 to 0.0096), with Tasmania the highest among all states (0.185–0.291) and Western Australia and the Northern Territory the lowest. The Energy Transition Effect (ETE), defined as the natural logarithm of the ETI, follows the same regional pattern, while the Net Energy Transition Contribution (NETC) derived from the Shapley decomposition shows that the transition’s effect on emissions has been episodic rather than linear, turning negative in years when growth in fossil-fuel use outpaced renewable displacement. Marked regional variation is observed, with resource-intensive states such as Western Australia, Queensland, and the Northern Territory exhibiting distinct emissions patterns driven by mining and LNG activities, while more service-oriented states show more stable reductions.