DOMESTIC AND FOREIGN VALUE ADDED IN GLOBAL VALUE CHAINS: EVIDENCE FROM FIVE KEY INDUSTRIES IN VIETNAM’S SOUTHEAST REGION BY MACHINE LEARNING APPROACH
Huynh Thi Thuy Giang1, Nguyen Thi Hong Nham2, Le Mai Hai3, Nguyen Van Nen4, Nguyen Thanh Huyen4, Nguyen Anh Phong4, Tran Thi Huong4
1Office of Student Affairs, Vietnam National University-Ho Chi Minh City
2University of Economics Ho Chi Minh City (UEH)
3Ho Chi Minh City University of Economics and Finance
4University of Economics and Law, Vietnam National University Ho Chi Minh City
Abstract: This study examines the impact of global value chain (GVC) participation and firm characteristics on firm performance (ROA) in the Southeast region of Vietnam. Using panel data of listed firms from 2020-2024, the study applies machine learning models, including Ridge Regression, Random Forest, and Gradient Boosting. The results show that Ridge Regression provides the best predictive performance, indicating that the relationship between GVC indicators and firm performance is predominantly linear. Firm performance also exhibits strong persistence over time. Among GVC variables, domestic value added (DVA) has a stronger impact than foreign value added (FVA), suggesting that internal value creation is more important than reliance on imported inputs. In addition, financial leverage, firm size, and supply chain efficiency significantly affect performance, while technological capability shows a moderate effect. Overall, the findings highlight that firm performance depends not only on GVC participation but also on the quality of participation and internal capabilities, particularly in the context of a developing regional economy.
Keywords: Global value chains (GVC); Firm performance; Southeast region of Vietnam; Domestic value added (DVA); Foreign value added (FVA); Machine learning.
VOLUME 10 ISSUE 04 2026: 218 – 233