Consolidated Financial Statements, Industry Classification, and Discretionary Accruals

This study analyzes the effect of the difference between consolidated and separate financial statement revenue on the appropriateness of discretionary accruals, focusing on Korean listed companies for which the consolidated financial statement became the primary financial statement following the adoption of K-IFRS. Discretionary accruals are typically estimated using accrual models by industry and year. A large revenue discrepancy between consolidated and separate financial statements may indicate that the performance of subsidiaries in various industries could influence the estimation of discretionary accruals. In this study, we assess the appropriateness of discretionary accruals by examining whether they meet statistical assumptions, specifically that the residuals from the accrual model’s regression have an expected value of zero and no serial correlation. The analysis reveals that, across the full sample, the expected value of discretionary accruals was zero, but significant serial correlation was present. However, when dividing the sample into quintiles based on the revenue discrepancy between consolidated and separate financial statements, we found that for firms with no revenue discrepancy, discretionary accruals had an expected value of zero and showed no serial correlation. Conversely, for firms with the largest revenue discrepancy, while the expected value of discretionary accruals remained zero, the serial correlation was highest and statistically significant. These findings suggest that when estimating discretionary accruals by industry, it is important to first examine whether the classification and impact of industries are accurately reflected in the consolidated financial statements that companies report.