Foretelling or foresight? Accounting-based bankruptcy prediction models and earnings quality in the case of Polish listed companies
Objective: The article aims to examine the relationship between bankruptcy risk and earnings quality in designing accounting-based bankruptcy prediction models. The models classify companies (firm-year observations) into two groups with high or low (no) bankruptcy risk. We investigated the differen...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Krakow University of Economics
2025-06-01
|
| Series: | International Entrepreneurship Review |
| Subjects: | |
| Online Access: | https://ier.uek.krakow.pl/index.php/pm/article/view/2307 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Objective: The article aims to examine the relationship between bankruptcy risk and earnings quality in designing accounting-based bankruptcy prediction models. The models classify companies (firm-year observations) into two groups with high or low (no) bankruptcy risk. We investigated the difference in earnings quality between those two groups.
Research Design & Methods: We used quantitative research methods, such as descriptive statistics, correlation analysis (Pearson’s and Spearman’s rank correlation), and Welch ANOVA. The study sample consisted of firm-year observations of companies listed in the Warsaw Stock Exchange for 17 years (2007-2023) ranging from 5 004 up to 5 688 firm-year observations. We employed five accounting-based bankruptcy prediction models specific to the Polish context and two metrics of earnings quality: accrual and real earnings management. We estimated the proxy of accrual earnings management using the modified Jones model and real earnings management with the Roychowdhury model. We estimated the bankruptcy risk using five prediction models and then analyzed as a continuous (in correlation analysis) and dichotomous variable (Welch ANOVA).
Findings: The research results demonstrate that companies classified by company failure models as high bankruptcy risk are associated with lower earnings quality. The results of the Welch ANOVA analysis are consistent across all combinations of accounting-based prediction models and earnings quality proxies used in the study research. The findings imply that a high bankruptcy risk is associated with managers engaging in more intensive accrual and real earnings management. The results suggest managers are more inclined to influence reporting numbers and operational activities to achieve desired goals.
Implications & Recommendations: Scholars should consider diminishing the quality of earnings associated with higher bankruptcy risk in designing and developing future accounting-based bankruptcy prediction models. Financial statement users like investors, financial analysts, financial auditors, and other stakeholders should also consider earnings quality. The study provides an avenue for future research by calling for research across earnings quality and bankruptcy prediction models.
Contribution & Value Added: The study contributes to a better understanding of the relationship between accounting-based bankruptcy prediction models and how they estimate bankruptcy risk and earnings quality. As far as we know, earnings quality has not been considered a factor in the design of models for bankruptcy prediction.
|
|---|---|
| ISSN: | 2658-1841 |