Financial reporting frameworks and distress prediction models in SME auditing: Evidence from the Visegrad Four countries

  • 13 Views
  • 1 Downloads

Creative Commons License DMCA.com Protection Status
This work is licensed under a Creative Commons Attribution 4.0 International License

Type of the article: Research Article

Although financial reporting and auditing standards are substantially harmonized across the European Union, important differences remain in national accounting regulations, audit thresholds, and the practical application of the going concern principle. These differences are particularly relevant for small and medium-sized enterprises (SMEs), which constitute the dominant segment of the Visegrad Four (V4) economies. This paper examines differences in financial reporting and auditing frameworks among the Czech Republic, Hungary, Poland, and Slovakia and develops sector-specific distress prediction models to support going-concern assessments in SME auditing.

The empirical analysis is based on financial statement data for 66,988 active firms obtained from the Orbis database. After data cleaning and consistency checks, a modelling sample of approximately 41,000 SMEs was constructed. Financial distress is defined as a persistent inability to cover interest obligations, represented by two consecutive years in which earnings before interest and taxes (EBIT) are lower than interest expenses. Distress status is modelled using financial ratios, firm-size indicators, industry characteristics, and selected variables inspired by ISA 570. Separate binomial logistic regression models are estimated for country–industry groups derived from NACE classifications and evaluated using hold-out samples.

The results confirm that country- and sector-specific models achieve satisfactory predictive performance and provide useful support for assessing going-concern risks. The results also show that the most informative predictors differ across countries and industries, reflecting differences in regulatory environments and economic structures. The study highlights the importance of local calibration when developing distress prediction models and demonstrates that a universal approach may lead to reduced predictive accuracy. The proposed models provide a practical screening tool for auditors, lenders, and SME managers, complementing professional judgement and broader audit procedures.

Acknowledgments

This study is co-financed by the governments of Czechia, Hungary, Poland, and Slovakia through Visegrad Grants from the International Visegrad Fund. Visegrad Grant No. 22420285, Title of the project: “Distress prediction models in V4 countries and their audit applicability”. The mission of the Fund is to advance ideas for sustainable regional cooperation in Central Europe. The authors gratefully acknowledge the support of their home institutions and thank the anonymous reviewers for their insightful comments and suggestions.

view full abstract hide full abstract
    • Table 1. List of potential predictors of distress of SMEs operating in V4 countries applied in own research
    • Table 2. Grouping the industry categories using the CART approach
    • Table 3. Cox-Snell pseudo R2 results
    • Table 4. Regression results – CZ models
    • Table 5. Regression results – SK models
    • Table 6. Regression results – PL models
    • Table 7. Regression results – HU models
    • Table 8. ROC testing results (AUC values)
    • Table 9. Youden’s J statistic and optimal cutoff
    • Table 10. True positive and true negative rates
    • Conceptualization
      Michal Karas, Błażej Prusak
    • Data curation
      Michal Karas, Eva Gulyas
    • Formal Analysis
      Michal Karas, Błażej Prusak, Milos Tumpach
    • Methodology
      Michal Karas, Jiri Lunacek
    • Project administration
      Michal Karas
    • Writing – original draft
      Michal Karas, Błażej Prusak, Eva Gulyas, Milos Tumpach
    • Writing – review & editing
      Michal Karas, Błażej Prusak, Jiri Lunacek
    • Investigation
      Błażej Prusak, Eva Gulyas, Milos Tumpach
    • Supervision
      Błażej Prusak, Jiri Lunacek
    • Validation
      Błażej Prusak
    • Visualization
      Błażej Prusak
    • Resources
      Eva Gulyas, Milos Tumpach, Jiri Lunacek