Bankruptcy Prediction from 10-K Narratives: Evidence from Interpretable Text Scores and Accounting Baselines
A new study finds that analyzing the narrative sections of annual 10-K filings, especially Items 1, 1A, and 7, can greatly improve bankruptcy prediction compared to traditional financial ratios. The Pre-Bankruptcy Stress (PB Stress) Score, a transparent text-based measure, boosts model accuracy, raising the area under the curve (AUC) from 0.83 to 0.90. This method captures early distress signals missed by standard models, though it relies on company disclosures and rare bankruptcy events.
What it examines
This study examines if analyzing the narrative sections of annual 10-K filings, using a new Pre-Bankruptcy Stress (PB Stress) Score, can improve bankruptcy prediction beyond traditional accounting ratios. The approach uses dictionary-based text analysis to capture distress signals in company disclosures, aiming for more timely risk identification.
What it concludes
The PB Stress Score adds significant predictive value to bankruptcy models, capturing distress signals not found in financial ratios. This tool can help investors, creditors, and regulators better monitor financial risk. Limitations include rare bankruptcy events and reliance on disclosed information. Future research could expand to broader defaults and advanced text analysis.
Evidence objects
Analyzing narrative sections of 10-K filingsItems 1, 1A, and 7significantly improves bankruptcy prediction, outperforming traditional accounting ratios and offering early warning signals not found in financial data alone.
key_findings bullet 1 · key_findings · validation V0
The study introduces the Pre-Bankruptcy Stress (PB Stress) Score, a transparent, dictionary-based text measure capturing distress signals like liquidity issues, debt covenant stress, operating decline, restructuring, and business fragility in company disclosures.
key_findings bullet 2 · key_findings · validation V0
Adding the PB Stress Score to a five-factor financial model boosts the area under the curve (AUC) from $0.83$ to $0.90$, increasing bankruptcy capture in the riskiest decile from $44%$ to $65%$.
key_findings bullet 3 · key_findings · validation V0
This paper introduces a transparent, distress-specific Pre-Bankruptcy Stress Score using interpretable text features from 10-K narratives, surpassing traditional accounting ratios and sentiment dictionaries. Its alignment of prediction windows with disclosure dates and robust benchmarking demonstrate substantial predictive improvements, offering a novel, practically impactful approach to bankruptcy prediction in quantitative risk management.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
Bankruptcy is a low-frequency but high-impact corporate event, making early risk identification important for creditors, investors, regulators, and risk managers. Traditional bankruptcy-prediction models rely primarily on accounting ratios, but these measures may reflect financial deterioration only after it appears in reported financial statements. Narrative disclosures in annual 10-K filings may therefore provide incremental warning signals about emerging distress. This study examines whether 10-K narratives improve bankruptcy prediction beyond conventional accounting variables. Using firm-year observations matched to 10-K text, SEC financial statement data, and bankruptcy events from the Florida-UCLA-LoPucki Bankruptcy Research Database, the analysis evaluates bankruptcy risk over the year following the 10-K filing date. The paper develops a transparent Pre-Bankruptcy Stress (PB Stress) Score, a dictionary-based measure designed to capture distress-specific language related to liquidity and funding stress, debt covenant and refinancing stress, operating deterioration, restructuring and legal distress, and business fragility. The score is evaluated against a five-variable accounting baseline and a Loughran-McDonald dictionary benchmark. In the primary one-year holdout test, adding the PB Stress Score increases AUC from 0.8323 to 0.9019 and raises top-decile bankruptcy capture from 44.12% to 64.71%. The positive incremental pattern remains visible across bootstrap inference, alternative accounting benchmarks, alternative outcome definitions, and out-of-time validation. The findings indicate that distress-specific 10-K narratives provide interpretable incremental information for bankruptcy-risk monitoring beyond conventional accounting ratios.
Source row: 281 · abstract type: unknown