Forecasting financial spillovers in a green supply chain: a deep learning analysis of Tesla and its Taiwanese supplier

What the paper says

Purpose This study aims to investigate whether short-term financial spillovers exist within a green supply chain by analyzing the dynamic stock price linkage between Tesla Inc. and its Taiwanese upstream supplier, Innolux Corporation. The research examines how Tesla’s daily price movements, particularly sustainability-related negative shocks, affect the supplier’s short-term stock performance. Design/methodology/approach This study uses daily stock price data for Tesla and Innolux from 2014 to 2024. A long short-term memory (LSTM) deep learning model is used to forecast short-term price patterns, while a decision tree model identifies nonlinear threshold responses that characterize spillover transmission within a green supply chain. A regression model is further applied to statistically validate the cross-firm linkage and quantify the magnitude of Tesla’s influence on its supplier. Findings The results demonstrate that Tesla’s prior-day price movements have a significant predictive effect on Innolux’s next-day closing price. The LSTM model shows strong forecasting accuracy (RMSE = 0.4760; R² = 0.9121), while the decision tree identifies a critical threshold indicating that when Tesla’s daily decline exceeds 1.4%, Innolux is highly likely to experience a price drop. The regression analysis further confirms a strong linear relationship between the two firms’ stock price dynamics (R² = 0.992). Together, these findings provide robust evidence of short-term financial spillovers from a leading EV manufacturer to its upstream supplier within a green supply chain ecosystem. Research limitations/implications This study analyzes only a single upstream supplier in Tesla’s green supply chain, which may constrain the generalizability of its findings. In addition, external influences – such as macroeconomic fluctuations, policy shocks and geopolitical events – were not incorporated into the models. Future research may broaden the scope to include multiple firms across different tiers of the supply chain, integrate sustainability-related news or sentiment data, and apply nonlinear or regime-switching models to capture more complex spillover dynamics. Practical implications This study offers investors and portfolio managers actionable insights into short-term financial spillovers within green supply chains. The identified predictive linkage between Tesla and its upstream supplier, Innolux, provides a basis for timely risk assessment and threshold-based trading strategies, particularly during periods of sustainability-related negative shocks. The findings also support policymakers by underscoring the importance of improving data transparency, refining ESG disclosure mechanisms and promoting digital finance tools that enhance the resilience of sustainable industrial ecosystems. Social implications This study highlights the societal importance of transparency and resilience in sustainable supply chains. By revealing the financial interdependence between a leading global EV manufacturer and a Taiwanese upstream supplier, the findings emphasize the role of cross-border collaboration in advancing sustainability objectives. Enhanced data openness and the use of AI-based analytical tools can empower investors, firms, and policymakers to make informed decisions that support environmental responsibility and positive social impact. Originality/value To the best of the authors’ knowledge, this study is among the first to examine short-term financial spillovers within a green supply chain using deep learning techniques. By focusing on Tesla and its Taiwanese upstream supplier, Innolux, it presents a novel case-based analytical framework that integrates LSTM forecasting, decision tree threshold identification and regression validation. The research contributes to the green finance and sustainable supply chain literature by uncovering previously unobserved, short-horizon spillover mechanisms between leading EV firms and their suppliers within a sustainability-oriented industrial ecosystem.