Can information on social media be used to help auditors judge whether a client will fail?
Over the last five years, audit firms have publicly commented on their increased focus on big data, including information found on social media. Likewise, regulators have suggested expanding the scope of audit evidence to consider nontraditional data sources. However, to date, less is known about how this information could be utilized by auditors.
To fill this information gap, Georgia Tech researchers Eric Condie and James “Robbie” Moon studied whether information on the investment-focused social media site Stocktwits could be useful for auditors’ decision making.
“There is growing research on how investors, firms and even analysts utilize social media,” said Moon. “However, there is little to no research on how social media could be relevant to auditors. We focus on one aspect of the auditor’s opinion—a firm’s ability to continue as a going concern (GC), i.e., whether or not the client will likely fail in the next year or so.”
In their paper, #Fail: Social Media, Firm Distress, and Going Concern Opinions, Moon and Condie study a sample of more than 15,000 audit opinions pertaining to more than 4,000 financially distressed companies and nearly 40 million social media posts related to those companies.
Their study asks two research questions:
- Is the information available on social media relevant to auditors?
- To what extent do auditors’ decisions reflect information available on social media?
“Using both a measure provided by Stocktwits and one we developed using a machine learning method, which could easily be replicated by auditors, we did find strong evidence that information on social media is associated with future firm failure,” said Condie.
They also found that although investors on social media do provide information relevant to evaluating the likelihood of firm failure, most auditor’s GC opinions fail to reflect this information.
Finally, the study provides analyses directly relevant to practicing auditors.
“We conducted a basic simulation where we assigned GC opinion to firms where investors express the most negative sentiment on social media,” said Moon. “Our naïve approach suggests that the quality of GC decisions increases. While we do not suggest firms should follow our exact approach, it provides a baseline for fostering innovative developments in using social media opinions and other types of information in the audit process.”
To find out more about their results and research methods, read the entire paper: #Fail: Social Media, Firm Distress, and Going Concern Opinions.
