National ESG Performance and Economic Implications for the Global Construction Industry: Evidence from Panel, Cross-Sectional, and Machine Learning Analyses

The global construction industry's significant economic and environmental footprint drives the need for sustainability, making Environmental, Social, and Governance (ESG) criteria pivotal. This study empirically investigates national-level ESG impacts on construction value-added and housing costs, addressing knowledge gaps. A multi-method approach analyzed global data (2009–2023 panel, N=588; 2022 cross-section, N=33). Panel analysis employed a Two-Way Fixed Effects (LSDV) model. The cross-sectional 2022 housing cost analysis used ordinary least squares (OLS) regression and correlation. Complementary machine learning (Random Forest, XGBoost) explored predictive relationships for value added. The panel model identified GDP Growth (Annual %) as a significant positive predictor of construction value added (Coef. 0.01, p=0.02). Cross-sectionally, higher Control of Corruption correlated with increased 2022 housing costs (Coef. 1445.33, p<0.001), while higher Internet Users (% Pop.) correlated with lower costs (Coef. -376.95, p<0.001). Machine learning showed high predictive accuracy for value added, its lag being dominant; without it, political stability and social factors were prominent. Economic conditions and industry persistence strongly drive construction value added. However, specific governance (corruption control) and social digitalization aspects significantly correlate with housing costs, offering nuanced insights into the ESG-economic nexus in construction, differentiating predictors for aggregate industry performance versus project-related cost indicators.