Managers' Overconfidence and Earnings Management through Classification Shifting: The Moderating Role of Managerial Ability

Objective In recent years, accounting researchers have primarily focused on the management of discretionary accruals and the manipulation of real business activities, paying less attention to earnings management through classification shifting of income statement items. Although the topic of earnings management through classification shifting has recently gained attention from researchers, and some domestic studies have addressed it, there is limited empirical evidence on how classification shifting can be influenced by managerial beliefs. Therefore, this study aims to investigate the relationship between managers' overconfidence and classification shifting and examine the moderating role of managerial ability in this relationship. Methods To test the research hypotheses, data from 138 firms listed on the Tehran Stock Exchange over the period 2012–2023 (1,656 firm-years) were utilized. The Generalized Least Squares (GLS) method was applied for model estimation, with controls for both year and industry fixed effects. To address potential issues of heteroscedasticity and autocorrelation in the errors, cluster-robust standard errors at the firm level were employed. The models for measuring unexpected core earnings and unexpected changes in core earnings were estimated cross-sectionally at the industry level (120 industry-years). Results The research results indicate that a reduction in non-core earnings leads to an increase in unexpected core earnings and a decrease in unexpected changes in core earnings. These findings suggest the prevalence of the classification shifting phenomenon in Iranian firms. Additionally, the results show that managers' overconfidence strengthens the positive (negative) relationship between the reduction in non-core earnings and unexpected core earnings (unexpected changes in core earnings). With an increase in managerial ability, the impact of managers' overconfidence on the relationship between the reduction in non-core earnings and the two variables of unexpected core earnings and unexpected changes in core earnings diminishes. Furthermore, the results are not sensitive to the use of an alternative definition for measuring managers' overconfidence. Conclusion Based on the research results, it can be concluded that an increase in managers' overconfidence intensifies earnings management through classification shifting of income statement items. Overconfident managers have an exaggerated trust in their own abilities, and therefore, they intentionally classify some core expenses as non-core expenses to report higher core earnings than actual. This behavior not only skews the true financial performance of the firm but also misleads stakeholders relying on these statements. Furthermore, the research findings indicate that as managerial ability increases, the impact of managers' overconfidence on classification shifting diminishes. Compared to overconfident managers with weak managerial ability, overconfident managers with higher abilities have a more realistic assessment of their capabilities and sufficient capacity to meet their forecasts about the firm, thus having less motivation and need to reclassify income statement items. This suggests that improved managerial skills can mitigate the negative effects of overconfidence, ensuring more accurate and reliable financial reporting.