Leveraging Machine Learning to Understand Environmental Tax Opposition in Different Regions and Periods

High environmental taxes face global opposition despite their broad recognition as essential for environmental protection. This study employs the random forest algorithm and the ISSP Environment surveys from 2020 (N=43,678) and 2010 (N=27,032) across 28 countries to identify factors associated with opposition to environmental tax increases. Simultaneously analyzing 41 individual-level factors and tax system characteristics, we find individual values and environmental evaluations generally outweigh demographic factors for predicting tax opposition. The importance of factors varies over regions and time. In high-income countries, concerns about environmental issues and prioritizing jobs and prices emerge as influential, gaining prominence over the previous decade. Conversely, an individual’s lack of commitment to pro-environmental behavior is the most important predictor in emerging economies. Our results highlight the dynamic nature of environmental attitudes. Policymakers and advocates can leverage these insights to tailor and target the communication of environmental tax increases in different contexts, for instance, by emphasizing job creation.