Assessing whether reported financial statements are intentionally misstated (or manipulated) is of considerable interest to researchers, creditors, equity investors, and governmental regulators. While there is a lot of information out there on deception detection, there is not a lot of useful information. This recent study, where the authors analyze linguistic features present in answers of CEOs and CFOs during quarterly earnings conference calls, caught my eye. Investors might want to pay attention
The economist summarizes the study (below).
Deceptive bosses, it transpires, tend to make more references to general knowledge (“as you know…”), and refer less to shareholder value (perhaps to minimise the risk of a lawsuit, the authors hypothesise). They also use fewer “non-extreme positive emotion words”. That is, instead of describing something as “good”, they call it “fantastic”. The aim is to “sound more persuasive” while talking horsefeathers.
When they are lying, bosses avoid the word “I”, opting instead for the third person. They use fewer “hesitation words”, such as “um” and “er”, suggesting that they may have been coached in their deception. As with Mr Skilling’s “asshole”, more frequent use of swear words indicates deception. These results were significant, and arguably would have been even stronger had the authors been able to distinguish between executives who knowingly misled and those who did so unwittingly. They had to assume that every restatement was the result of deliberate deception; but the psychological traits they tested for would only appear in a person who knew he was lying.
Abstract From the Study
We estimate classification models of deceptive discussions during quarterly earnings conference calls. Using data on subsequent financial restatements (and a set of criteria to identify especially serious accounting problems), we label the Question and Answer section of each call as “truthful” or “deceptive”. Our models are developed with the word categories that have been shown by previous psychological and linguistic research to be related to deception. Using conservative statistical tests, we find that the out- of-sample performance of the models that are based on CEO or CFO narratives is significantly better than random by 4%- 6% (with 50% – 65% accuracy) and provides a significant improvement to a model based on discretionary accruals and traditional controls. We find that answers of deceptive executives have more references to general knowledge, fewer non-extreme positive emotions, and fewer references to shareholders value and value creation. In addition, deceptive CEOs use significantly fewer self- references, more third person plural and impersonal pronouns, more extreme positive emotions, fewer extreme negative emotions, and fewer certainty and hesitation words.