What are you saying? Using topic to detect financial misreporting*

Detection models of financial misreporting have evolved beyond basic quantitative or financial measures to include textual or linguistic characteristics of firms’ disclosures. While these textual analysis methods provide incremental power in identifying misreporting, they examine how content is being disclosed as opposed to what is being disclosed. This study introduces a novel fraud-detection measure, labeled as “topic,” that quantifies the thematic content of financial statements. We derive our measure from a Bayesian topic modeling methodology called Latent Dirichlet Allocation (LDA). We then demonstrate the incremental predictive power of our topic measure in detecting intentional financial misreporting. We identify occurrences of financial misreporting using SEC enforcement actions (AAERs) and restatements arising from intentional misapplications of GAAP (i.e., irregularities). We find strong evidence that topic predicts intentional misreporting beyond financial and textual style characteristics. Furthermore, our results indicate that the detection power of financial metrics is subsumed by our topic measure in prediction models for both AAERs and restatements arising from irregularities.