Purpose This study argues that the mixed evidence on how analyst coverage influences earnings management can be reconciled by accounting for analysts' in-house network centrality, a source of heterogeneity in information production that aggregate coverage measures overlook. Design/methodology/approach We use a large sample of US-listed firms to examine whether coverage by central analysts affects the extent and type of earnings management. Following recent studies, we measure analysts' centrality by their in-house connections and construct two firm-level variables to capture heterogeneity in network positions. Findings We find that although analyst coverage generally reduces both accrual-based and real earnings management, firms followed by more central analysts tend to substitute real activities manipulation for accrual-based methods. This pattern is consistent with the greater transparency effect rather than stronger monitoring or reduced pressure. Further analysis shows that the substitution effect weakens in firms with high information asymmetry or strong corporate governance and remains unchanged after the Sarbanes-Oxley Act. Originality/value These findings clarify the mechanisms through which analyst heterogeneity influences financial reporting behavior and help reconcile conflicting results in prior research.
Jiang et al. (Mon,) studied this question.