Auditor size and analysts’ earnings forecast accuracy: The moderating effect of earnings manipulation

This study offers an alternative explanation for the positive relation between auditor size and accuracy of analysts’ earnings forecasts. Utilizing a U.S. dataset, this study reaffirms the robustness of this well-established relation, even when employing a more recent dataset and a staggered difference-in-difference approach. Specifically, analysts’ earnings forecast accuracy improves when auditors switch from non-Big N to Big N, and vice versa. More crucially, this paper introduces novel evidence demonstrating that this relation is consistently and positively moderated by conventional proxies for earnings management. These proxies encompass firms’ management of real activities and the subsequent restatement of inflated earnings. Further analyses reveal that this relation is negatively moderated by external scrutiny applied by both analysts and auditors. In contrast, I do not find evidence indicating that the relation is similarly moderated by proxies related to downward management guidance. These results suggest that, at least within a subset of Big N clients, there appears to be a tendency for engaging in upward earnings management to meet earnings benchmarks, while downward management guidance does not significantly contribute to the accuracy of analysts’ earnings forecasts.