Analysis of world trade data with machine learning to enhance policies of mineral supply chain transparency

The increasing integration of supply chains worldwide and the establishment of resilient material flows emphasize the significance of transparency. As regulations and policies around mineral supply become more stringent, organizations are actively seeking effective tools to assess the transparency of their supply chains. Ensuring supply chain transparency plays a vital role in international trade data since it addresses the issue of inconsistent reporting by two parties involved in a transaction, sometimes referred to as bilateral asymmetries. Nevertheless, bilateral asymmetries might be utilized as a proxy to examine discrepancies in the transparency of supply chains. This paper presents a methodology to evaluate supply chain transparency using bilateral asymmetries as a proxy and provide insights into policy changes. We used a machine learning-based methodology on UN Comtrade data to study asymmetry trends in 116 million trade transactions over 30 years. The analysis demonstrates different levels of asymmetry among commodities and countries, suggesting differences in the transparency of supply chains. We exemplified the implementation of the methodology by analyzing 14 commodities associated with lithium batteries and their primary resources. The findings identify seven commodities that exhibit good (reliable) reporting practices, while two commodities (Cobalt and Lithium Primary Batteries) demonstrate bad (unreliable) reporting patterns. This indicates specific areas that should be examined and improved through policy changes. The paper presents a succinct and practical approach to measuring and strengthening supply chain transparency, giving actionable insights for policymakers and stakeholders for future actions.