As every US presidential race shows, politicians work tirelessly to convince voters that they have what it takes to run the country — the mother of all institutions! In a way, this is also the plight of top-level leaders who never interact with those individuals on whose support they depend. Think of corporate heads; just like politicians, they too need to convince others — stockholders, analysts, top management, employees, and stakeholders at large — of their competence and ability to steer an institution, whether a private or public organization, in the right direction.

What exactly determines whether a top-level leader is perceived to have what it takes? This is the question we asked ourselves in a recent paper published in the August issue of the Academy of Management Journal. We know that when it comes to forming impressions of leaders, individuals rely on performance signals — that is, they look at how well the economy is doing or at company results. But we also know that observers tend to look for certain characteristics in leaders — for example, having an aura of charisma. Both of these mechanisms, which are often perceived to be at loggerheads, have been separately shown to predict the evaluation and selection of leaders and this in a variety of contexts. What is unclear is how these mechanisms work in combination, particularly when they convey different information.

boss cartoonConsider the case of the late Steve Jobs. One of the reasons he is often hailed as an exemplary CEO is that he was both considered to be charismatic and presided over a high-performing company. But what would we have thought of him if Apple were performing well but he were not charismatic? Or conversely, if he were charismatic but Apple were not performing well? Perhaps counterintuitively, we found an answer to this question by studying US presidential elections. Such elections serve as an ideal context to test how top-level leaders, whether in the corner or oval office are evaluated and selected. Indeed, every election year, we can observe the performance of the same institution (the country) which serves as a comparable signal of the incumbent’s performance, and we can observe the outcome of the election in terms of votes received. Also, importantly for modeling purposes, elections are held at known cycles (thus the decision to select a leader is “exogenous” and does not depend on unobservable factors).

We examined all US presidential elections from 1916 to 2008, looking at both the effect of the economy and of incumbency following a well-established econometric model of voting, developed by Ray C. Fair from Yale University. This model assumes that the characteristics of presidential candidates do not matter — an assumption that we question and use to improve the model, particularly so when the model is unclear as to whether the incumbent or his party will be rewarded or punished at the next election.

Therefore, we extended this model by adding in a measure of how charismatic the candidates are. To measure charisma we coded candidates’ nomination speeches for specific rhetorical techniques that have been shown, both in the lab and in the field, to influence how charismatic one is perceived to be; these techniques have to do with providing strong frames of reference and a vision by using storytelling, metaphors, and so forth. The tactics also focus on the substance of what is said including the leader’s values and the goals that are articulated. We ignored message “delivery” (use of voice, nonverbal behavior) because we only had data in written form for the elections held in the distant past; this decision did not matter because rich verbal rhetoric correlates with rich gestures as we have discovered previously. Thus verbal charisma serves as a good proxy for charisma overall.

What we found is that the effect of charisma “kicks in” when the economy and the incumbency record alone do not predict a clear winner. In this sense, it appears that voters are very pragmatic because:
Based on how well they perceive the economy to be doing, they either reward or punish the incumbent candidate (or party) by voting for or against them.
If it is unclear how well the economy is doing—a situation we refer to as “attributional ambiguity” — they then closely scrutinize the candidates, which is when they are most likely to be swayed by the allure of charisma.
We also replicated these results in an experiment in which participants saw one of six versions of a TV business news report we produced for this study. Here, we manipulated CEO charisma (high or low) and performance signals (clearly good/bad or ambiguous). Again, we found that charisma mattered most when performance signals were unclear.

Taken together, what these results show is that charisma really makes a big difference in times of attributional uncertainty — that is, when it is unclear whether the organizational performance we tend to attribute to leaders is good or bad. Interestingly, in the experiment, we also found that charisma even mattered, albeit less so, when performance signals were clear. Thus, charisma seems to help across-the-board and it can give leaders the benefit of the doubt across an array of situations, even when the performance of their organization is poor. For those interested, for a more technical treatment of this topic, see our recent article titled “Charisma: An ill-defined and ill-measured gift.”

What can we say about the current U.S. elections? Presently, econometric forecasts seem to communicate a clear negative signal which should, in and of itself, be sufficient to drive voters’ decisions against the incumbent party — in other words, the Republicans currently have the upper hand. However, the situation can still change in the coming months. If the state of the economy improves substantially the Democrats will have a clear chance. Or, if the signals become ambiguous, charisma will kick in and the presidency will be up for grabs. We will post our own forecast shortly after both parties nominate their candidates (here).

Stay tuned for updates!


Note:  This article gives the views of the author, and not the position of LSE Business Review or the London School of Economics.

Featured image credit: Joi Ito CC-BY-2.0 Inside photo credit: valerly osipov CC-BY-2.0

Philippe Jacquart is an associate professor of leadership at EMLYON Business School in France. His current research focuses on leadership at the upper echelons. Twitter: @PhilippeJCQRT


AntonakisJohn Antonakis is full professor of organizational behavior at the Faculty of Business and Economics of the University of Lausanne. He is currently working on leadership, individual differences, and power. Twitter: @JohnAntonakis