CEOs' charismatic language leads analysts to less accurate projections

CEOs' charismatic language leads analysts to less accurate projections

University Park, Pa. -- Securities analysts tend to offer more favorable, yet often inaccurate, forecasts to companies whose new CEOs use charismatic language in their vision statements, according to new research coauthored by a professor at Penn State's Smeal College of Business. The research results suggest that analysts should be more cautious when considering CEO charisma in their projections.

Vilmos Misangyi, assistant professor of management at Smeal, and his colleagues Angelo Fanelli of HEC School of Management and Henry Tosi of the University of Florida examine the effects of CEO charisma on analyst forecasts in their paper, "In Charisma We Trust: The Effects of CEO Charismatic Visions on Securities Analysts," forthcoming in Organization Science.

The researchers looked at a random sampling of 367 CEO transitions that occurred between 1990 and 1999 and measured the use of charismatic language in new CEOs' vision statements through a text analysis of the CEOs' first letters to shareholders following their appointments.

Charismatic visions tend to be very critical of the past; use moral, ideological and emotional overtones in their formulation of future goals; and use empowering language in describing their means to achieve their vision. Thus, to measure charismatic language in each letter to shareholders, the researchers divided each of the sentences in each letter into three categories: "assessment of the past," "plans for the future," "shareholders, employees, and organizational capabilities." Sentences that did not fit into any category were disregarded.

Next, the researchers examined each of the categories for the occurrence of particular words that were preselected as being reflective of a charismatic vision of the past, future, and organizational capabilities. Among the most frequently occurring of the preselected terms in CEO assessments of their new companies' pasts were the words "difficult," "disappointing," and "problems." "Believe," "important," and "commitment" were some of the most frequently occurring terms in their visions for the future. And terms like "our," "we," and "team" were frequently used in sentences dealing with their assessment of their organizations' capabilities.

After assigning each of the CEO letters a CCV (CEO charismatic visions) score based on the intensity (i.e., number of charismatic words/total words in letter) of such charismatic language contained in the letter, the researchers used statistical regression techniques to examine the relationship between the CCV scores and analyst recommendations during two time periods: six months and 12 months after each letter's publication. Their results show that the more charismatic language that appeared in the letters, the more likely analysts were to recommend the company's stock. The professors also found that analysts across the board offered more uniformly positive recommendations when the letters included more charismatic language.

In the final portion of their analysis, Misangyi and his colleagues compared analyst earnings-per-share forecasts with each company's actual earnings. Their results show that the use of charismatic language in CEO vision statements is associated with increased analyst forecast error, both positive and negative.

"The results suggest that CCV are related to the individual and collective judgments of securities analysts because they are associated with favorable analyst stock recommendations and uniformity across analysts," the researchers conclude. "Furthermore, it appears that CCV, as a form of symbolic action, adversely affect analysts' forecasts of future firm EPS."

Given that securities analysts' livelihoods depend so heavily on their reputations and, therefore, on the accuracy of their forecasts, this new research may cause analysts to be more mindful of charismatic language in corporate communications.

"Our results imply that CCV may have negative consequences for analysts' judgments and ultimately for their careers," the researchers write. "It appears that analysts should be wary of incorporating CCV into their evaluations."