Robust determinants of green finance

Abstract

Green finance is instrumental in helping emerging and developing economies transition to environmentally sustainable practices and address climate change challenges. Nevertheless, robust empirical evidence on its determinants remains limited. This study analyzes the potential and robust determinants of green finance in 93 developing countries from 2000 to 2020. First, we provide an extensive review of existing research on green finance and identify its potential determinants. Second, we determine the robust determinants of green finance by applying a range of estimation techniques, including Bayesian Model Averaging (BMA), Lasso, Elastic Net, Weighted Least Squares, and non-parametric methods (Random Forest and Gradient Boosting). The robust predictors of green finance are globalization, GDP per capita, trade, multilateral flows, climate vulnerability, government effectiveness, rule of law, official development assistance, unemployment, and population. Using these predictors improves the accuracy of out-of-sample forecasts compared to models that include all potential predictors of green finance identified by the literature. By focusing on the identified determinants, policymakers can mobilize green finance more effectively to support climate-resilient development.