Book Description

The purpose of this study is to identify important SEC comment letters and examine the mechanisms by which they affect firm value. The SEC periodically reviews public-company financial statements, issuing comment letters in response to disclosure deficiencies, to ensure that investors are provided with material information, and to prevent fraud. Given that comment letters consist of unstructured text, statistical text classification may be an effective technique to identify comment letter importance. The information in comment letters is distributed over several separate filings and they are not widely cited by the press or analysts as information sources, which may result in investor inattention and underreaction to their disclosure. I utilize negative abnormal returns following comment letter disclosure as the primary indicator of comment letter importance, and develop a Naive Bayesian classification model that signals important comment letters from their text features that are associated with the indicator. In a holdout sample, the text classification model correctly identifies important comment letters between 10 and 40 percent better than chance. The average out-of-sample abnormal return for firms with signaled comment letters is -5.8 percent during the 90 days post-disclosure, but only when the comment letters were viewed on EDGAR. Signaled comment letters are associated with lower persistence of profits and increased material restatements in the year following comment letter disclosure.