Mapping global value chains using customs data – the case of energy storage technologies

Licentiate thesis, 2025

This thesis examines global value chains for energy storage technologies by addressing three research questions that focus on different parts of the chain: (i) the diffusion of end-use technologies, (ii) the supply of intermediate components, and (iii) the origin of raw material supply.
First, an indicator is developed to quantify the diffusion of energy storage technologies based on the temporal flexibility they provide, with the aim of enabling cross-country comparisons of ESS diffusion. Due to data limitations, the indicator was initially applied to a small sample of regions. The results
indicate a general expansion of ESS in these regions and reveal technological variations across them.
Second, a methodology is developed to use customs data to map the global diffusion of a focal product and the supply of its components. The methodology is applied to lithium-ion battery (LIB) energy storage systems (ESS) as illustrative case. The results reveal the geographical diffusion of ESS, global patterns of concentration and specialisation of LIB supply, and their transformations over time.
Finally, a methodology is introduced to identify the geographic origin of raw materials using customs data. This methodology is applied to two minerals of importance to LIBs, cobalt and lithium. The results show that, compared to current practice, exemplified by the market processes in the life-cycle
assessment database Ecoinvent, the proposed methodology can improve the spatial and temporal resolution of raw materials sourcing and hence the precision of environmental impact assessments.
Some general data problems are identified and addressed in this thesis. Current national databases do not support global monitoring of technological diffusion, which motivates the development of the customs-data-based methodology. However, customs data introduces its own data challenges,
including inconsistencies in data reporting, variations between databases and difficulties to distinguish producing and consuming countries from intermediate countries merely involved in re-trading of goods. These issues are identified and mitigated in the proposed methodologies.
Overall, this thesis demonstrates how customs data can be leveraged to quantitatively study global value chains and applies the developed methodologies to energy storage technologies. Designed to be generalisable, these methodologies are transferable to other technologies and expected to provide spatially explicit insights into global value dynamics and enhance the spatial resolution of environmental impact evaluations.