Analyzing Speech to Detect Financial Misreporting

In our paper, Analyzing Speech to Detect Financial Misreporting, forthcoming in the Journal of Accounting Research, we examine whether nonverbal vocal cues elicited from speech are useful in detecting intentional deception in financial reporting. Detecting deceptive financial reporting is an increasingly important concern for auditors, regulators, investors, and the various constituents that interact with corporations. High profile accounting scandals such as Enron, WorldCom, Tyco, and Satyam have cost market participants several billions of dollars and eroded confidence in published financial statements. These events call into question the ability to uncover financial misstatements by auditors who review and provide an opinion on the financial statements (PCAOB [2007], [2010]). Even sophisticated market participants such as institutional investors and analysts have been remarkably unsuccessful at detecting financial fraud (Dyck et al. [2010]).

We empirically document that vocal dissonance markers are useful for identifying financial misreporting. In a laboratory setting, we generate a speech sample of misreporters and truth-tellers to provide construct validity for the vocal dissonance marker using automated vocal emotion analysis software based on Layered Voice Analysis (LVA) technology. We find vocal dissonance markers from the early part of the speech samples—the precise time when dissonance should be most pronounced—are positively associated with four measures of dissonance from misreporting. This lends support for the LVA-based cognitive dissonance measure. In an archival setting, we find that cognitive dissonance in CEO speech can predict whether a firm’s quarterly financial reports will be adversely restated at better than chance levels. The predictive ability of vocal dissonance markers is incremental to accounting based predictors of adverse irregularity restatements.

This paper provides some of the first archival evidence to suggest that important nonverbal clues to detect financial misreporting are present in earnings conference calls. These results should be informative to investors, analysts and auditors who attempt to use earnings conference calls as an information source for assessing the risk of misreporting. However, we caution the reader of the following limitations. First, while we attribute our findings to cognitive dissonance felt by subjects, it is possible that some unknown emotional factor(s) correlated with this construct accounts for our results. Second, LVA is an emerging technology and, as with most commercial products, its inner workings are proprietary. While our laboratory results suggest the LVA dissonance metrics capture aspects of the construct of cognitive dissonance, we are unable to document the mechanisms by which LVA is able to do so. Finally, we only investigate CEO speech. Our laboratory results suggest the possibility that detection of dissonance from misreporting is not specific to CEOs, thus, there may be important information in CFO speech as well. We leave an examination of relative dissonance in CFO speech, or speech from other corporate officials, for future research.

The full paper is available for download here.