The Effects of Language Power and Communication Mode on Investor Judgments

In this study, I examined how, in the context of the question and answer (Q&A) segment of earnings calls, language power contained in a manager’s speech (powerful or powerless) interacts with the mode by which the manager’s speech is communicated to investors (audio recording or written transcript) to influence investor judgments of the attractiveness of a target company as an investment opportunity (i.e. persuasion). I used a controlled experiment to provide insights into this research question. I found that that language power in a manager’s speech had no influence on investor judgments of a target company when earnings call information is transmitted via written transcripts. In contrast, when earnings call information was transmitted via audio recordings, investor judgments were more negative when the manager speaking on the call used powerless language than when he used powerful language. In addition, when earnings calls were transmitted via audio recordings, the valence of participants’ speaker-related thoughts were more negative when the manager used powerless language in his speech than when he used powerful language. The results also showed that the effects of language power on investor judgments were mediated by the valence of participants’ speaker-related thoughts when earnings calls were transmitted via audio recordings. Further analysis that I conducted provided evidence to suggest that non-verbal hesitations might have compensated for the negative effects of other powerless language cues (i.e. verbal hesitations and hedges) on investor judgments when earnings calls were transmitted via audio recordings. My study extends the literature on linguistic analysis by using an experiment to hold constant other aspects of a manager’s speech, and showing that language power plays an incremental role in influencing investor judgments. It is also the first to examine the relative effects of audio and text based disclosure on investor judgments.