Purpose This study aims to construct an improved global value chain (GVC) embedding mode indicator system in the context of new trends in GVC changes under deglobalization. It seeks to reveal the relationship between GVC embedding modes and carbon emission efficiency (CEE) based on internal driving factors and spatial spillover effects, and explore new pathways for enhancing the CEE of the transportation sector in Belt and Road (B&R) countries. Design/methodology/approach An econometric analysis was conducted based on the transportation industry data of B&R countries from 2007 to 2022. Findings GVC forward, upstream and downstream embedding modes have a positive relationship with the CEE, with the internal driving factor being the technology effect. GVC backward embedding mode has a negative relationship with the CEE, with the internal driving factor being the scale effect. These relationships vary across countries, time, and industries. The spatial spillover effects of GVC embedding were concentrated in the upstream embedding mode. However, after emergencies, represented by the COVID-19 pandemic, the spatial spillover effect disappeared. Originality/value (1) An improved GVC embedding mode indicator system is proposed, expanding the quantification theory of GVC embedding modes. (2) A STIRPAT-internal driving factor testing model is constructed that is not limited to specific industries, regions or time periods, achieving a transition from impact effect analysis to the identification of internal driving factors. (3) A spatial spillover perspective is introduced, extending the study of the relationship between GVC embedding modes and the CEE to the spatial level. Graphical abstract Graphical abstract A conceptual diagram showing research aim, framework, and innovation linked with arrows and labeled effects. The three-section conceptual diagram is arranged vertically with headings “Research aim”, “Research framework”, and “Innovation”, each enclosed within rectangular boundaries. At the top, under “Research aim”, a single text block reads: “Explore new approaches to improve the C E E of the transportation sector in B and R countries from the perspective of GVC embedding modes”. In the middle section labeled “Research framework”, a dashed rectangular boundary encloses a flow structure. On the left, a text reads, “G V C embedding mode” and connects to two items: “Analysis of Internal Driving Factors” and “Analysis of Spatial Effects”. From these, arrows extend toward the right. A red arrow labeled “Technical effect (plus)” points toward the right. Below it, a blue arrow labeled “Scale effect (minus)” also points toward the right. A black arrow labeled “Impact of unexpected events” points horizontally toward the right. All arrows converge toward a label on the right reading “Transport C E E”. Three vertical shaded downward arrows appear behind the horizontal arrows. At the bottom, the “Innovation” section contains three items arranged horizontally. The first reads, “Propose an improved G V C embedding mode indicator system”. The second reads, “Construct the S T I R P A T-internal driving factor testing model”. The third reads “Introduce a spatial spillover perspective”. Each item is preceded by a small circular marker and aligned beneath the framework section. A shaded downward arrow extends from “G V C embedding mode” to the first section in innovation. The next shaded downward arrow extends from the start of the horizontal arrows in the research framework to the second section in innovation. Another shaded downward arrow extends from the end of the bottom horizontal arrow in the research framework to the third section in innovation.
Li et al. (Fri,) studied this question.