Full Length ArticleHow to prevent leadership hubris? Comparing competitive selections, lotteries, and their combination
Introduction
As Aristotle famously noted, power and glorification often lead to hubris in leaders, which is defined as overconfidence in one's own abilities and the abuse of power (Aristotle Aristotle, 2003: 1378b 23–30). Hubris results in neglecting the limitations and precariousness of one's human condition (e.g. Cairns, 1996). For example, CEOs affected by hubris pay high premiums for unprofitable corporate acquisitions (Billett & Qian, 2008; Hayward & Hambrick, 1997a; Malmendier & Tate, 2008), invest in pet projects funded by internal cash flows (Malmendier & Tate, 2005), compensate themselves with salaries that the firm's performance does not justify (Billett & Qian, 2008), and demand rewards based on luck or other factors beyond their control (Liu & De Rond, 2016). Another example is the tax evasion revealed in the so-called Panama Papers, which were leaked in 2015. It became apparent that 143 politicians from all over the world, including many former and current democratic countries' heads of state and government were corrupted by their power and used offshore shell corporations to hide illegal financial transactions (Obermayer & Obermaier, 2016).
Previous research has extensively analyzed the detrimental consequences of hubris, in particular the consequences of CEO hubris for corporate outcomes. Governance mechanisms that may limit hubris of leaders have often been discussed, but mostly under the perspective of board control. For example, Hayward and Hambrick (1997a) show that vigilant boards, characterized by an independent board chair, outsider directors, and share ownership, restrain CEOs from paying large premiums for corporate acquisitions, which they might otherwise feel hubristically entitled to do.
To our knowledge, this is the first study that proposes a governance mechanism used successfully in history to tackle the problem of leadership hubris today: competitive selection combined with lotteries (in the following, termed partly random selection). Historical evidence suggests that being chosen randomly prevents hubris, which is overconfidence and the abuse of power (Buchstein, 2010; Dowlen, 2017; Duxbury, 2002; Manin, 1997; McCormick, 2006; Sintomer, 2014; Van Reybrouck, 2016). It has also been shown that successful people who recognize that randomness or luck has played an important role more often express humility and a pro-social focus (Bartlett & DeSteno, 2006; Frank, 2016). However, lotteries may result in incompetent candidates being chosen. Therefore, they have usually been combined with conventional selection methods. In this vein, we develop our suggestion. Including randomness in the process through which leaders are appointed raises awareness among leaders themselves, board members, journalists, and others that “great man” theories may be unrealistic.
Lotteries today are rarely used to appoint leaders in real-life organizations, in contrast to what we know from history (see e.g. Buchstein, 2009; Manin, 1997; Sintomer, 2014; Van Reybrouck, 2016). Therefore, to test our theory, we devised a laboratory experiment in which subjects are appointed as group leaders in three treatment conditions: a competitive selection treatment, a random selection treatment, and partly random selection, in which competitive selection is combined with lottery. The design of this experiment is inspired by a historical example at the University of Basel, called Wahl zu Dreyen or selection from three (Burckhardt, 1916: 34; Stolz, 1986: 670). We measure hubris by using an incentivized behavioral measure, which has been applied in former experimental research on the abuse of power (Bendahan, Zehnder, Pralong, & Antonakis, 2015) and which has been suggested in observational studies to be a good proxy for leadership hubris (Hayward & Hambrick, 1997a). This design allows us to test whether the historically approved partly random selection of leaders helps to reduce leadership hubris more than the current practice of selecting leaders on the basis of competitive performance evaluations.
In contrast to former studies on leadership hubris, our study focusses on how features of organizational design shape the incentives of leaders (Zehnder, Herz, & Bonardi, 2017). The strength of this economic approach is that it determines the exogeneous conditions under which leaders with hubris can and will emerge. Former studies analyze the psychological traits of leaders or observe the performance outcomes of leaders; the empirical results are thus confounded by selection effects and omitted variables and are prone to endogeneity bias. Former research on leadership hubris does also not view organizational choices as trade-offs-decisions requiring a cost-benefit-analysis in which one option is compared to another. In our theory and experimental design, we consider that the cost and benefits of alternative methods of leader selection depend on the situation at hand. It implies that there is no globally best alternative of leader selection; but a best alternative to prevent leadership hubris. To study the exogenous conditions of the emergence of leadership hubris and to compare the different alternative selection methods against each other, we set up a microeconomic structure, where individuals are randomized to treatment, with clear and consequential links between decisions and payouts (Zizzo, 2010).
The remainder of the paper is structured into four sections. In the next section, we define leadership hubris and discuss how a random component in competitive selections can help to reduce leadership hubris. We present our main arguments with a formal model comparing competitive selection methods with partly random selection methods. We model the tradeoffs between the positive competence effects of competitive selections and the positive effects of partly random selections in preventing power abuse. Then, we introduce the empirical strategy we applied to test our hypothesis in a laboratory experiment and present our empirical findings. We close our paper with a discussion of our findings in the context of the existing literature on leadership hubris and their implications for the practice of selecting leaders.
Section snippets
What is leadership hubris?
The concept of hubris has attracted growing interest in business and management research, yet it is poorly defined in this literature (Hollow, 2014). Richard Roll (Roll, 1986) was the first to use the term to show that hubristic CEOs are often responsible for shareholder losses during mergers and acquisitions. Subsequently, researchers in management and finance have used the hubris concept to demonstrate the negative consequences of “exaggerated self-confidence” in CEOs (Chatterjee & Hambrick,
Experimental method
We test our theory in a laboratory experiment. In practice, conducting such an empirical test is not possible, because today modern organizations do not use lotteries to appoint their leaders. Our study complements the historical case studies, which derive from real-world settings but are not causally interpretable.
Discussion
Leadership hubris is a severe problem in many organizations (Hayward & Hambrick, 1997a; Hayward et al., 2006; Hiller & Hambrick, 2005; Roll, 1986). Leaders affected by hubris tend to overlook their limitations and take decisions that are harmful to the community (Billett & Qian, 2008; Hayward & Hambrick, 1997a; Malmendier & Tate, 2008). Previous studies on how governance mechanisms can prevent leadership hubris have mainly focused on board vigilance. These studies found that weak boards create
References (85)
Inequity in social-exchange
Advances in Experimental Social Psychology
(1965)- et al.
Leader corruption depends on power and testosterone
The Leadership Quarterly
(2015) - et al.
Destructive leadership behaviour: A definition and conceptual model
Leadership Quarterly
(2007) - et al.
Power and morality
Current Opinion in Psychology
(2015) - et al.
Who makes acquisitions? CEO overconfidence and the market’s reaction
Journal of Financial Economics
(2008) - et al.
Norms of 300 general-information questions: Accuracy of recall, latency of recall, and feeling-of-knowing ratings
Journal of Verbal Learning and Verbal Behavior
(1980) - et al.
Charismatic leaders and destructiveness: An historiometric study
The Leadership Quarterly
(1996) - et al.
Narcissism and emergent leadership in military cadets
The Leadership Quarterly
(2006) - et al.
A productive clash of cultures: Injecting economics into leadership research
The Leadership Quarterly
(2017) - et al.
CEO humility, narcissism and firm innovation: A paradox perspective on CEO traits
The Leadership Quarterly
(2017)