Climate-risk language and corporate carbon emissions

Climate Language and Emissions

Companies face growing pressure to recognize and report climate-related risks, but more discussion of climate change does not necessarily mean that a company has reduced its emissions. Annual-report language can reflect attention, reporting requirements, compliance, or pressure from investors and regulators; it is not a direct measure of managers’ private beliefs or of action taken to cut carbon.

Authors Mohammad Haseeb, Miguel Angel Esquivias, and colleagues from Bahçeşehir Cyprus University, Chitkara University, Universitas Airlangga, Baku Engineering University, European University of Lefke, Al-Ahliyya Amman University, Termez University of Economics and Service, and Tashkent State University of Economics examined Chinese A-share listed firms from 2011 to 2024. They used climate- and carbon-related language in annual reports as an indicator of expressed climate-risk attention and analyzed its association with estimated firm-level CO₂e using firm and year fixed effects, additional fixed effects, and double machine learning.

A Positive Association

The analysis found a positive association between climate-risk narrative attention and estimated corporate CO₂e: firms with denser climate- and carbon-related language in annual reports also had higher estimated emissions, conditional on the models’ controls. The authors describe this as a relationship in the observed data, not evidence that increased attention causes emissions to rise. The measure captures the share of annual-report language devoted to climate and carbon terms, rather than a direct assessment of managerial perceptions.

The authors interpret the pattern through a principal–agent lens, in which managers’ near-term performance pressures could be consistent with maintaining operations rather than prioritizing costly abatement. However, the study does not measure short-termism, incentive horizons, executive compensation, or continuity-preserving expenditure. The proposed explanation is therefore an organizing interpretation of the statistical result, not a mechanism tested by the analysis.

Governance, Innovation and Coal

Stronger internal controls were associated with a flatter positive relationship between narrative attention and emissions. The authors treat internal control as a moderator, while noting that the index may also reflect reporting quality, organizational capacity, firm maturity, or general management quality. The result does not establish that internal controls reduce emissions by resolving principal–agent conflicts.

In coal-relevant and energy-intensive industries, assigned province–industry coal productivity was associated with a flatter relationship between climate-risk language and CO₂e. This measure represents an assignment to a broader province–industry setting, not the listed firm’s own energy-conversion technology. Green patenting was positively associated with climate-risk narrative attention and negatively associated with emissions in the same year; the authors characterize these as contemporaneous associations, not a causal pathway or evidence of causal mediation.

Limits and Next Steps

The positive association was more pronounced among firms with lower digital transformation, state-owned enterprises, non-high-tech firms, and firms in less market-oriented regions. These are descriptive differences across groups, not evidence that directing an intervention toward any group would change emissions. The authors conclude that climate-related wording alone is a weak proxy for carbon performance and that screening on narrative attention alone would be a poor stand-in for estimated emissions in this sample.

The authors emphasize that the study does not identify a causal effect of climate-risk attention on emissions. Reverse causality and unobserved, time-varying factors remain possible; fixed effects and double machine learning do not resolve those limitations, and the instruments discussed in the paper do not establish a credible exclusion restriction. Future work should use instruments or quasi-experiments whose only plausible route to emissions is through managerial climate attention, examine direct measures of incentive horizons and operational choices, and test whether the findings extend beyond Chinese A-share firms.

Corresponding Author: Mohammad Haseeb or Miguel Angel Esquivias or Md. Emran Hossain or Lu’ay Al-Mu’ani or Khurshid Khudoykulov or Wisnu Wibowo

Original Source: https://doi.org/10.1007/s44246-026-00306-z

Contributions: All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Mohammad Haseeb, Miguel Angel Esquivias, and Md. Emran Hossain. The methodology and formal analysis were primarily developed by Mohammad Haseeb. Miguel Angel Esquivias supervised the study and contributed to the investigation and overall research coordination. Md. Emran Hossain, Lu’ay Al-Mu’ani, and Khurshid Khudoykulov contributed to validation, resources, and software support. Wisnu Wibowo contributed to visualization, project administration, and manuscript review and editing. The first draft of the manuscript was written by Mohammad Haseeb, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.