Textual Analysis of Sustainability Reports: Topics, Firm Value, and the Moderating Role of Assurance
Bibliographic record
Abstract
This study investigated how specific sustainability topics disclosed in standalone sustainability reports influence firm value and whether third-party assurance moderates this relationship. Drawing on signaling, agency, stakeholder, and legitimacy theories, we applied latent Dirichlet allocation (LDA) to extract latent topics from U.S. corporate sustainability reports. We analyzed their impact on Tobin’s Q using panel regressions and supplement our findings with discrete Bayesian networks (DBNs) and Shapley additive explanations (SHAP) to capture non-linear patterns. We identified six core topics: environmental impact, sustainable consumption, daily necessities, socio-economic impact, healthcare, and operations. The results revealed that topics of healthcare and daily necessities have immediate and sustained positive effects on firm value, while environmental and socio-economic impact topics demonstrate lagged effects, primarily two years after disclosure. The presence of assurance, however, produces mixed outcomes: it enhances credibility in some cases, but reduces firm value in others, especially when applied to environmental and socio-economic disclosures. This suggests a dual signaling effect of assurance, potentially increasing investor scrutiny when gaps in performance are highlighted. Our findings underscore the importance of topic selection, consistency in reporting, and strategic application of assurance in ESG communications to maintain stakeholder trust and market value.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.130 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".