Foreign direct investments and CO2 emissions in Asia: A data-driven optimal transport perspective

The environmental implications of Foreign Direct Investment (FDI) remain a topic of debate, with traditional models facing challenges in establishing a clear nexus between FDI and emissions. Against this background, we propose a novel, non-parametric approach based on Optimal Transport theory to examine the evolution of joint and conditional distributions of FDI and carbon emissions (CO2) across Asia. Leveraging satellite data, our method estimates optimal transport maps that represent the least costly way to transition between FDI and emissions distributions over time. We further construct one-period-ahead distributional forecasts using Wasserstein barycentric interpolation, ensuring minimal divergence from past observations. Our data-driven framework captures complex, nonlinear dependencies without imposing rigid parametric assumptions. The analysis reveals a positive relationship between FDI and emissions, with aggregate forecasts indicating a generally increasing trend for both variables. However, distinct patterns emerge across selected countries, suggesting heterogeneous environmental impacts of FDI.