Analyzing carbon dividends’ impact on financial security via ML & metaheuristic search

The question of a carbon tax and dividend has long intrigued policymakers and researchers. If enacted, this policy would levy a tax on carbon emissions, with the resulting revenue being distributed to American citizens through an annual dividend. Although current literature extensively covers the climate impacts of a carbon dividend, its economic effects in the short term remain unclear. We hypothesized that the modeled implementation of a carbon dividend would increase predicted financial security and stability. To analyze the tax’s effect on financial security across various income levels, we used the US Federal Reserve Board’s Survey of Household Economics and Decisionmaking (SHED) as a representative sample for the population of the United States. We then developed the Financial Security and Resilience Index (FSRI) using the response data and applied machine learning algorithms to predict changes in financial security based on each individual’s net income from a carbon dividend. Under an equal dividend scenario, we observed a 0.19% mean increase in financial security. After introducing a
genetic algorithm that allocated varying amounts of the carbon dividend’s total revenue to different income ranges, we observed an increase of ~0.5% in mean financial security. The varied allocation of funds achieved through metaheuristic search algorithms as a concept can certainly become an effective instrument, helping policymakers better understand possible improvements in policy design. These findings can aid in generating a strong movement in favor of the dividend, as our results indicate that a substantial majority of the population would benefit from the policy.