Words That Mean Change: Paradoxical Thinking and Eco‐Efficiency in the Automotive Sector

This study uses the automotive industry to investigate the link between paradoxical thinking language in corporate sustainability reports and companies' eco‐efficiency. We hypothesized that different types of paradoxical thinking would be associated with varying eco‐efficiency outcomes. Using ChatGPT to generate a lexicon of paradoxical terms, we analyzed sustainability reports from the automotive industry spanning 2002 to 2022, testing associations with four eco‐efficiency indicators. Our analysis reveals that paradoxical thinking matters, but different forms have varying degrees and forms of impact. Specifically, we find that cautious, conditional, and risk‐aware language correlates with higher eco‐efficiency, while growth‐focused and innovation‐oriented language correlates with lower eco‐efficiency. Companies using deliberative, uncertain language achieve superior environmental and economic balance, while those emphasizing expansion and bold innovation show weaker eco‐efficiency performance. This research suggests that specific language patterns in sustainability reports both reflect and potentially influence practical environmental outcomes, suggesting that paradoxical thinking is not uniform but varies significantly in its environmental consequences.