The Language of Transparent Disclosures

Universtity of 黑料传送门 Booth 

Most of the research on voluntary disclosure relies on the aggregate measures of disclosure such as management forecasts, tone and readability of textual disclosures. This project develops a new linguistic measure of transparency for textual disclosures that captures how forthcoming the management is with disclosures in the real time. Using the question and answer (Q&A) section of earnings conference calls, we measure the extent to which management answers each question. The broad category of “non-answers” includes instance when the management says “we don’t know”, “we don’t provide this information”, “we don’t disclose this”, etc. We further plan to use this measure to study product market competition and voluntary disclosure. Our goal is to make the code that extracts “non-answers” available to researchers. In contrast to manual coding of conference calls as in Hollander et al. [2010], this measure will offer new opportunities in voluntary disclosure research.