Search
Program Calendar
Browse By Day
Search Tips
Virtual Exhibit Hall
Personal Schedule
Sign In
We make use of machine learning techniques to determine whether financial analysts’ questions during earnings conference calls are informative about the quality of firms’ internal control over financial reporting, and whether this information can help auditors assess internal control quality. Using a large sample of quarterly conference calls of listed U.S. firms from 2010 to 2018, we use a topic modeling approach that represents both analysts’ question phrasing and thematic content. This approach allows us to capture variance in what analysts ask and how they ask it. We find that the resulting topic model score significantly predicts internal control material weaknesses beyond determinants identified by prior research, and that auditors make use of the information discussed by analysts when determining audit fees. We further suggest that auditors rely on this information more in years where the firm’s information environment is more volatile. Overall, our results suggest that analysts may serve as external monitor of internal control systems, thereby safeguarding financial reporting quality.
Christian Hofmann, Ludwig-Maximilians-Universitaet
Sebastian Kuhn, LMU Munich
Nina Schwaiger, Ludwig-Maximilians-Universitaet