Sustainable Energy Management in Vietnamese Firms: Evidence from Logit Regression and Random Forest

Vietnam is an energy-intensive economy that has a rapidly growing industry. This fact has increased the urgency for firm-level sustainable energy management practices, especially given the country’s low ranking in global environmental performance. In this study, we investigate the firm-level determinants of sustainable energy management adoption in Vietnam, focusing on structural characteristics, innovation capabilities, and external linkages. Using data from the 2023 World Bank Enterprise Survey, we apply both a logit regression model and a Random Forest algorithm, a novel combination in the study of determinants of sustainable energy management adoption. The logit model identifies significant positive relationships between sustainable energy management adoption and factors such as firm size, R&D investment, international quality certification, and export orientation. Managerial experience shows a non-linear relationship with sustainable energy management adoption, whereas foreign ownership influences it only when combined with R&D investment. The Random Forest model complements these findings by revealing nonlinear relationships and highlighting the predictive importance of variables like international certification, managerial expe-rience, and manufacturing sector affiliation. Together, the models show that internal capabilities and external pressures drive the adoption of sustainable energy management. Our results suggest that policy interventions should be designed with sector- and firm-specific contexts in mind to foster more sustainable firms.