Inte tillgängligt på svenska
Nicolas Dierick
- 7 April 2026
- OCCASIONAL PAPER SERIES - No. 385Details
- Abstract
- This article provides a novel insight into whether earnings manipulation signals are reflected in banks’ internal credit risk estimates, as measured by the probability of default (PD) estimates, and whether such manipulation has an impact on credit risk (point in time or deferred). The hypothesis is that firms engaging in manipulation may be exposed to increased credit risk over time, which should be reflected in higher PD values. Using AnaCredit – a granular dataset covering credit exposures from European banks between 2019 and 2022 – and financial statement data from Orbis, we constructed a sample of 4,649 publicly traded corporations, for which we computed the Beneish M-Scores that are used to detect potential earnings manipulation. This allowed us to determine the interrelation with PDs. Our results reveal a weak and negative correlation between M-Scores and PDs, suggesting that earnings manipulation signals are not fully absorbed by banks’ internal models. Further analysis shows that these results are driven by the high prevalence of firms with no earnings manipulation signals. Firms for which the M-Score effectively indicates potential earnings manipulation (8.9% of the sample) are observed to have higher PDs, which also increase further as the M-Score worsens. These findings support the hypothesis that earnings manipulation signals are not fully reflected in credit risk estimates over time, indicating that their impact – when it occurs – is deferred instead of being captured immediately in internal models. Our results indicate that the relationship between potential earnings manipulation and banks’ internal credit risk estimates is highly context-dependent and non-linear. Cross-sectional analyses by country and industry show consistent patterns linking default risk to M-Scores in selected countries and sectors. [...]
- JEL Code
- G32 : Financial Economics→Corporate Finance and Governance→Financing Policy, Financial Risk and Risk Management, Capital and Ownership Structure, Value of Firms, Goodwill
M41 : Business Administration and Business Economics, Marketing, Accounting→Accounting and Auditing→Accounting
M42 : Business Administration and Business Economics, Marketing, Accounting→Accounting and Auditing→Auditing
C33 : Mathematical and Quantitative Methods→Multiple or Simultaneous Equation Models, Multiple Variables→Panel Data Models, Spatio-temporal Models