Internal Credit and External Blame: Self-Attribution in Operations and Supply Chain Performance
Внутренняя заслуга и внешнее обвинение: самоатрибуция в эффективности операций и цепочки поставок
2026-06-19
SCID: 54.1/rwga3968
Discuss with AI
LLaMA-3.1-8B-Instruct fine-tuningOM self-attributionearnings call Q&A transcriptsexternal blameinternal credit
Figures from the paper
Abstract (AI)
Problem definition: This study examines self-attribution tendencies and their associated outcomes in operations management (OM) performance evaluation, in which managers are not always objective when giving credit to outcomes. They may attribute strong performance to internal factors (“internal credit”) while assigning poor performance to external factors (“external blame”). Methodology: Large language models based on LLaMA-3.1-8B-Instruct are fine-tuned to analyze earnings call Q&A transcripts and construct a firm-level measure of OM self-attribution. The superiority of LLM's performance is validated against traditional NLP approaches (Gradient Boosting, K-Nearest Neighbors, and Random Forest) and alternative LLMs (e.g., GPT-4o). The LLM-based measures are then used in Multinomial Logit models and fixed-effect models to assess the existence of self-attribution and its associations with subsequent outcomes. Results: Empirical analysis reveals a consistent pattern of OM self-attribution: managers claim internal credit for intra-organizational actions while assigning external blame to supply chain partners or macroeconomic conditions. This pattern persists after controlling for managers' experiences, Q&A sequencing, leading questions, contextual factors, and uncertainty in LLM outputs. OM self-attribution is more prevalent among larger firms, firms led by more experienced managers, and firms with lower operational efficiency and supply chain risk. Whereas non-OM financial self-attribution has been documented as a psychological bias, OM self-attribution exhibits a distinctive diplomatic and constructive nature. In contrast to the temporary effect of non-OM financial self-attribution, OM self-attribution is associated with real changes to financial and operational outcomes. Firms with stronger OM self-attribution tendencies receive more positive market and media responses and tend to have better future positioning in production and demand management and supply chain optimization. Theoretical and Managerial Implications: This study complements existing work on financial self-attribution by identifying a unique pattern of self-attribution in the OM context that functions as a strategic illusion of managerial control. OM self-attribution can be interpreted as a strategy where managers strategically shape attributions to signal active engagement in an uncertain environment, ultimately associated with positive future firm performance.
Key Findings
1
An LLaMA-3.1-8B-Instruct–based LLM fine-tuned on earnings call Q&A transcripts produces a firm-level OM self-attribution measure that outperforms traditional NLP methods (gradient boosting, k-NN, random forest) and alternative LLMs (e.g., GPT-4o).
2
Firms exhibiting stronger OM self-attribution receive more positive market and media responses and show better future positioning in production, demand management, and supply chain optimization.
3
Managers display consistent OM self-attribution: crediting internal actions for strong performance and blaming external supply chain or macroeconomic factors for poor performance.
4
OM self-attribution differs from non-OM financial self-attribution by being diplomatic and constructive and is associated with real subsequent improvements in financial and operational outcomes.
5
OM self-attribution is more prevalent in larger firms, firms led by more experienced managers, and firms with lower operational efficiency and lower observed supply chain risk.
Research Object
Operations-management (OM) self-attribution behavior by firm managers as expressed in earnings call Q&A transcripts
Research Subject
The tendency to assign internal credit and external blame (self-attribution) and its associations with subsequent financial, operational, market, media, and supply-chain performance outcomes
Publication Details
Publication Date
2026-06-19
Journal
Publisher
ISSN
Cited by
1
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai4
AI-Assisted Pipeline for Dynamic Generation of Trustworthy Health Supplement Content at Scale2018
Large language models encode clinical knowledge2023
FinBERT : A Large Language Model for Extracting Information from Financial Text*2022
Investor Psychology and Security Market Under‐ and Overreactions1998