Green assets are not so green: assessing environmental outcomes using machine learning and local projections

Abstract

This paper examines the environmental impact of green assets using machine learning and impulse responses by local projections. A series of 87 green assets from various classes are considered, namely firms providing renewable energy and carbon offset solutions, carbon and sustainable investing ETFs and green cryptocurrencies. The dataset spans the period from 2015 to 2022 and comprises globally sourced environmental and financial data. The current study examines whether asset prices, returns and trading volumes have an impact on environmental indicators such as temperature (global mean and anomalies) and greenhouse gas concentration. The results indicate that adoption of these green assets does not have a significant environmental impact, suggesting that they should not be used as substitutes for real climate action. This work serves as a cautionary tale on the nexus between green assets and environmental indicators and the results can be used by governments and corporations when formulating climate and ESG strategies.