1 Copyright 2005 by the Educational Publishing Foundation Review of General Psychology 1089-2680/05/$12.00 DOI: 10.1037/1089-2618.104.22.168 2005, Vol. 9, No. 2, 169 –180 What We Know About Leadership Robert B. Kaiser Robert Hogan Kaplan DeVries Inc. Hogan Assessment Systems This article reviews the empirical literature on personality, leadership, and organiza- tional effectiveness to make 3 major points. First, leadership is a real and vastly consequential phenomenon, perhaps the single most important issue in the human sciences. Second, leadership is about the performance of teams, groups, and organiza- tions. Good leadership promotes effective team and group performance, which in turn enhances the well-being of the incumbents; bad leadership degrades the quality of life for everyone associated with it. Third, personality predicts leadership—who we are is how we lead—and this information can be used to select future leaders or improve the performance of current incumbents. nean army but was sent to prison for 7 years in A very smart political scientist friend used to say, “The fundamental question in human af- 1971 for taking part in an attempted coup. After fairs is, who shall rule?” We think the funda- his release, he went to Libya to train with other mental question is, “who should rule?” Leader- West African revolutionaries; there he met ship is one of the most important topics in the Charles Taylor (the recently deposed dictator of human sciences and historically one of the more Liberia), who became Sankoh’s major ally. poorly understood; it is important for two rea- Sankoh founded the Revolutionary United sons. First, leadership solves the problem of Front to overthrow the Sierra Leonean govern- how to organize collective effort; consequently, ment and take over the country’s diamond it is the key to organizational effectiveness. mines. With good leadership, organizations (govern- Sankoh was bright, charming, and charis- ments, corporations, universities, hospitals, matic, and he immediately attracted a large pop- armies) thrive and prosper. When organizations ular following, especially among the teenage succeed, the financial and psychological well- underclass. He promised to reform education, being of the incumbents is enhanced. health care, and other public services and to Second, and more important from a moral distribute the diamond revenues. Instead, he perspective, bad leaders perpetrate terrible mis- used the revenues to buy arms (from Charles ery on those subject to their domain. Consider Taylor) and political support. He paid his sol- the career of Foday Sankoh, the former dictator diers irregularly because he expected them to of Sierra Leone, who died in July 2003. Sankoh live by looting and even by cannibalizing vic- was born in 1937 and grew up in a Sierra Leone tims of the army. New recruits were sometimes dominated by a small, corrupt urban elite whom required to murder their own parents, which he deeply resented. He joined the Sierra Leo- toughened them and made it hard to return home. His young recruits, deprived of parenting and raised in chaos, were notoriously savage and specialized in amputating appendages, Robert Hogan, Hogan Assessment Systems, Tulsa, Okla- homa; Robert B. Kaiser, Kaplan DeVries Inc., Greensboro, which they kept in bags. Those with the most North Carolina. body parts were rewarded. By the end of the We are grateful for the helpful comments of Roy 1990s, Sierra Leone was, according to the Baumeister and John Antonakis on earlier versions of this United Nations, the poorest country on earth. To article. Correspondence concerning this article should be ad- stop the slaughter and ameliorate the misery, the dressed to Robert Hogan, Hogan Assessment Systems, 2622 United Nations, after several false starts, inter- East 21st Street, Tulsa, OK 74114, or Robert B. Kaiser, vened in 2000. Sankoh was taken captive by an Kaplan DeVries Inc., 1903 G Ashwood Court, Greensboro, emboldened mob that had been fired upon by NC 27455. E-mail: [email protected] or [email protected]landevries.com his bodyguards. He was subsequently indicted 169
2 HOGAN AND KAISER 170 by an international court for crimes against hu- Germany? In this Marxist view, people are manity. While in prison, he “lost his mind,” had merely the creatures of their circumstances. The second view is represented by Sigmund a stroke, and died of a pulmonary embolism, Freud, Thomas Carlyle, and Max Weber, who leaving his impoverished country and its muti- lated citizenry finally in peace. Sadly, the moral argued that, from time to time, shrewd, talented, and charismatic figures emerge in society, cap- to this story—that bad leaders cause much mis- ery—is all too common. tivate and energize a significant following, and then change history. Although writers such as This article tries to make three points. The Herbert Marcuse (1969) have tried to integrate first is that leadership matters; it is hugely con- the views of Marx and Freud, the history of sequential for the success of organizations and social theory over the past 100 years has been the well-being of employees and citizens. Sec- ond, when conceptualized in the context of hu- the dialectic exchange between these two perspectives. man origins, it becomes clear that leadership is We adopt the currently out-of-vogue view an adaptive tool for individual and group sur- that history is the history of social movements vival. We believe that, in essence, leadership primarily concerns building and maintaining ef- led by individuals, for better or worse (as de- fective teams: persuading people to give up, for scribed in the preceding). That is, we favor explanations based on concrete personalities a while, their selfish pursuits and pursue a com- rather than abstract social forces. mon goal. Our final point is that the personality of a leader affects the performance of a team: Who we are determines how we lead. Defining Personality Personality concerns two major elements: (a) Conceptualizing Leadership generalizations about human nature (what peo- ple are like way down deep) and (b) systematic We first began studying leadership in the accounts of individual differences (which dif- mid-1980s, and we quickly discovered that the ferences are important and how they arise). literature contained few defensible generaliza- With regard to generalizations, the pioneers of tions other than such nuggets as leaders seem to personality psychology (e.g., Freud, Jung, be somewhat taller and a little bit brighter than Adler, Horney, and Erikson) argued that the their subordinates (Stogdill, 1948). Since then most important generalization we can make is we have been assembling a perspective on lead- that everyone is somewhat neurotic, which ership that makes sense to us. The following is means that the most important problem in life is a review of our perspective. to overcome one’s neurosis. However, that gen- eralization is contradicted by the data; for ex- ample, the base rate of neuroticism is too low to Conceptualizing History be a generalized characteristic (Renaud & Estes, 1961). Moreover, the “good life” involves more There are two major viewpoints regarding the principal dynamic in history and human affairs, than the absence of pathology (Seligman & Csikszentmihalyi, 2000). and they derive from two distinct causal per- On the other hand, a review of sociology, spectives. The first is the tradition represented anthropology, and evolutionary psychology by Hegel, Marx, Durkheim, and modern-day sociologists (and social psychologists, although suggests an alternative generalization that, in fact, is two related generalizations. First, people they do not realize it); this tradition assumes that there is a tide running in human affairs, a always live in groups; we evolved as group- tide defined by history or the economy— by living animals. Second, every group has a status large impersonal forces outside human con- hierarchy; there are people at the bottom, in the trol—and individuals are merely floating on the middle, and at the top, and everyone knows who tide. Many of us have the illusion that we con- is where. This suggests that the most important problems in life concern getting along with trol our own destiny, but what individual ac- tions brought about the worldwide depression other people and achieving some measure of status. We refer to these concerns as “getting of the 1930s that swept the Nazis into power in
3 SPECIAL ISSUE: WHAT WE KNOW ABOUT LEADERSHIP 171 along” and “getting ahead,” and individual dif- dition is by far the larger and more popular Leader- literature. It consists of such works as ferences in these capabilities predict a wide range of occupational outcomes (see J. Hogan (Roberts, 1990) ship Secrets of Attila the Hun & Holland, 2003). It is also worth noting that and the self-serving and account-settling mem- effective leaders are skilled at building relation- oirs of former CEOs and politicians. Despite its ships and acquiring status. popularity, the troubadour tradition is a vast collection of opinions with very little support- To understand personality, the concept should be defined from two perspectives: (a) ing evidence; it is entertaining but unreliable. how a person thinks about him- or herself (i.e., In contrast, the academic tradition is a col- lection of dependable empirical nuggets, but it a person’s identity) and (b) how others think is also a collection of decontextualized facts that about that person (i.e., a person’s reputation). A do not add up to a persuasive account of lead- person’s identity concerns his or her most ership. This is the result of two unfortunate deeply held beliefs, whereas a person’s reputa- trends in earlier leadership research. The first tion is an index of his or her success in life. Identity is hard to study, and we do not know a concerns the fact that leadership researchers great deal about it. In contrast, reputation is easy have historically ignored personality (Bass, 1990), and they have done so despite evidence to study and vastly consequential. Our research indicates that it is important to that personality has effects on leadership (see, distinguish two aspects of reputation, which we for example, Mann’s, 1959, conclusions as call “the bright side” and “the dark side.” The compared with the reanalysis of his data by bright side concerns the initial impression we Lord, DeVader, & Alliger, 1986). Second, re- searchers have routinely defined leadership ei- make on others—it reflects our social perfor- mance when we are at our best—for example, in ther as standing out in a crowd or as occupying a job interview or on a first date. The five-factor a senior position in an organization. Both defi- nitions overlook the fundamental essence of model (Wiggins, 1996) is a taxonomy of the bright side; it reflects how observers perceive leadership. and describe others in the early stages of a relationship (McAdams, 1995). The dark side Leadership Effectiveness reflects the impression we make on others when Leadership is usually defined in terms of the we let our guard down or when we are at our people who are in charge of organizations and worst, such as when we are stressed, ill, or their units; by definition, such people are lead- intoxicated. The bright side concerns the person ers. But reflect for a moment on the skills you meet in an interview; the dark side concerns needed to successfully negotiate the status hier- the person who actually comes to work. Dark archy of a large bureaucratic organization. side tendencies typically coexist with well-de- Think about the people who are in charge of the veloped social skills that mask or compensate organization where you work and try to find for them in the short run. Over time, however, examples of real leadership. The people who dark side tendencies erode trust and undermine rise to the tops of large organizations are dis- relationships. Both the bright side and the dark tinguished by hard work, intelligence, ambition, side of reputation can be studied through ob- political skill, and luck but not necessarily by server descriptions, and most of the major out- talent for leadership. comes in life (jobs, promotions, relationships) As an alternative way to conceptualize lead- reputation. Moreover, effective leaders depend on ership, think for a moment about human origins. described reputations (as distinctive have People evolved as group-living animals, be- subsequently). cause there is safety in numbers. Over the 2 million years of human prehistory, the various The Leadership Literature hominid groups were in competition for the control of resources, and the competition was Although the leadership literature is im- typically quite savage. For example, when mense, it can be effectively sorted into two Genghis Khan invaded Persia, he killed every the troubadour tradition categories that we call inhabitant (de Hartog, 2000). People are natu- . The troubadour tra- and the academic tradition
4 HOGAN AND KAISER 172 rally selfish and inclined to pursue their short- analysis as applied to managerial work. The term self-interest. Leadership involves persuad- competency movement spread rapidly and ing people to set aside, for a time, their selfish quickly became chaotic and idiosyncratic. Our pursuits and work in support of the communal first point is that every existing competency interest. In the context of the violent tribal war- model can be captured with the domain model fare that characterized most of human history, proposed by Hogan and Warrenfeltz (2003). leadership was a solution for group survival; The model appears in Table 1. leadership is a collective phenomenon (Avolio, In brief, this model identifies four broad Sosik, Jung, & Berson, 2003, p. 287). classes of managerial competencies: (a) intra- In our view, then, leadership should be de- personal skills (regulating one’s emotions and fined in terms of the ability to build and main- easily accommodating to authority), (b) inter- tain a group that performs well relative to its personal skills (building and maintaining rela- competition. It follows that leadership should be tionships), (c) business skills (planning, budget- evaluated in terms of the performance of the ing, coordinating, and monitoring business ac- group over time. Our view is a radical departure tivities), and (d) leadership skills (building and from the conventional wisdom of leadership motivating a high-performance team). We research. Most studies define leadership in would like to highlight three points about this terms of emergence—the person in a group of domain model. First, it is developmental: In- strangers who exerts the most influence— or in trapersonal skills develop first, probably in the terms of ratings of an individual “leader” by preteen years; interpersonal skills develop next, more senior “leaders.” Although very few stud- probably during the teenage years; business ies have used indices of group performance as skills develop when a person enters the work- 1 the criterion for leadership, we believe this is force; and leadership skills develop last. Sec- the most appropriate way to define and evaluate ond, the model is a hierarchy of increasing leadership. With this definition in mind, we turn trainability, with intrapersonal skills being hard to a discussion of what we know about to train and leadership skills being the easiest to leadership. train. Third, the model is comprehensive; every existing competency model can be organized in What We Know About Leadership terms of these four domains. In addition to having a taxonomy of compe- The foregoing is the framework in terms of tencies, we also have very good measures of the which we conceptualize leadership. The re- key elements in these domains. There is solid mainder of the article concerns the dependable meta-analytic evidence showing that measures facts, what we know about leadership that is of core self-esteem and measures of integrity empirically true. We think we can summarize predict occupational performance in the .30 – what we know in terms of seven points. .50 range (Judge & Bono, 2001; Ones, Viswes- varan, & Schmidt, 1993). Similarly, measures Competencies of interpersonal skill correlate in the .50 region with performance in customer service and sales Our first point concerns competency models. jobs (Frei & McDaniel, 1998; Vinchur, Schipp- The competency movement began with the mann, Switzer, & Roth, 1998). We can also work of David McClelland (1973), a personality predict business skills using measures of cogni- psychologist with practical interests. McClel- tive ability with equally good results (Schmidt land’s model was designed to identify compe- tencies that were specific to a particular job in a 1 particular organization, with no intention of - For example, in one of the first meta-analyses of lead generalizing. The modern enthusiasm for com- ership, Lord et al. (1986) remarked that most leadership researchers “have over generalized results from leadership petencies seems to have taken off after the pub- perceptions to the topic of leadership effectiveness” (p. lication in 1982 of a book by McClelland’s 407). Although researchers are beginning to realize the colleague, Boyatzis, partly as a result of the importance of defining leader effectiveness in terms of team book’s appeal and partly as a result of wide- or unit performance, much work remains to be done on this topic. spread dislike of traditional methods of job
5 SPECIAL ISSUE: WHAT WE KNOW ABOUT LEADERSHIP 173 Table 1 The Domain Model of Competencies Definition and sample competencies Domain Intrapersonal Internalized standards of performance; able to control emotions and behavior (courage and willingness to take a stand; career ambition and perseverance; integrity, ethics, and values; core self-esteem and emotional stability; patience; tolerance of ambiguity) Interpersonal Social skill role-taking and role-playing ability; talent for building and maintaining relationships (political savoir faire, peer and boss relations, self-presentation and impression management, listening and negotiating, oral and written communications, customer focus, approachability) Business Abilities and technical knowledge needed to plan, budget, coordinate, and monitor organizational activity (business acumen, quality decision making, intellectual horsepower, functional/technical skills, organizing ability, priority setting, developing effective business strategy) Leadership Influence and team-building skills (providing direction, support, and standards for accomplishment; communicating a compelling vision; caring about, developing, and challenging direct reports; hiring and staffing strategically; motivating others; building effective teams; managing diversity) Credibility as a leader depends vitally on & Hunter, 2004). Finally, we can predict vari- ous aspects of leadership performance with va- perceived integrity: keeping one’s word, fulfill- lidities as high as .50 using multivariate regres- ing one’s promises, not playing favorites, and sion equations of normal personality (e.g., not taking advantage of one’s situation. The Hogan & Hogan, 2002; Judge, Bono, Ilies, & most important question we ask of potential Gerhardt, 2002). All of this means that we have leaders is, “Can we trust you not to abuse the the assessment tools needed to identify potential privilege of authority?” A meta-analysis con- leaders; regrettably, these tools are rarely used ducted by Dirks and Ferrin (2002) showed re- in selecting corporate executives (DeVries, liable correlations between trust in one’s super- 1993). visor and a range of positive leadership out- comes, including improved job performance, job satisfaction, and organizational commit- Implicit Models of Leadership ment. Like Caesar’s wife, people in leadership positions must avoid even the appearance of Earlier we stated that discussions of person- impropriety. ality should distinguish between identity and In addition, good leaders make good deci- reputation. Our second point is that we now sions in a timely way. In times of crisis and have a very clear view of the reputational ele- uncertainty, the most effective leaders make ments of leadership. Specifically, the literature prompt decisions (Vroom & Jago, 1988; Yukl, on implicit leadership theories suggests the 1998, chap. 11). Naval historians are astonished characteristics people look for in their leaders; at the quality of Horatio Nelson’s decision mak- this research also tells us which of the positive ing under the almost unimaginably difficult and attributes listed by C. Peterson and Seligman confusing conditions of a sea battle (Pocock, (2004) define effective leaders in the eyes of the 1987). But decisiveness is also important under led. In order of importance, the four themes that normal conditions. Mintzberg (1973) observed appear regularly in this literature—the leader- that managers are involved in decision making ship virtues—are integrity, decisiveness, com- all day long, and the quality of their decisions petence, and vision (e.g., Kouzes & Posner, accumulates. 2002; Lord, Foti, & DeVader, 1984).
6 HOGAN AND KAISER 174 Good leaders are also competent; they are a ethnographic studies of leadership (Boehm, 1999). In hunter-gatherer groups, the head man contributing resource for their groups. In hunter- gatherer tribes—which are ferociously demo- is modest, self-effacing, competent, and com- mitted to the collective good. And if he is not, cratic—the head man is usually distinguished from the group by superior hunting ability and a he gets removed, sometimes quite violently. broader moral perspective (see Boehm, 1999). Expertise is needed for legitimacy and respect Personality and Leadership from the team (French & Raven, 1959); the fact In the best study yet published on the links that colleges and universities are typically led between personality and leadership, Judge et al. by failed academics partially explains problems (2002) conducted a meta-analysis in which they with faculty morale. examined 78 studies of the relationship between Finally, good leaders are able to project a vision, to explain to the group the purpose, personality and leadership. They organized per- meaning, and significance of its key undertak- sonality in terms of the generally accepted tax- onomy of reputation, called the five-factor ings. Napoleon noted that “leaders are dealers in model (Wiggins, 1996); this is a taxonomy of hope”; we would add that vision is their cur- the bright side of personality. The dimensions rency. In addition, vision facilitates team per- of the five-factor model are Extraversion, formance by clarifying roles, goals, and the way Agreeableness, Conscientiousness, Emotional forward (House, 1971). George H. W. Bush is Stability, and Openness. (Emotional Stability by all accounts a decent and likable man, but he and Conscientiousness reflect the first element is utterly pragmatic in his thinking; before the of the domain model shown in Table 1, intrap- 1992 election, he complained to his staff that he did not understand “this vision thing,” which, of ersonal skills; Extraversion and Agreeableness course, is not what people want to hear from concern the second domain, interpersonal skills; and Openness, which is related to vision, an- potential leaders. chors the fourth domain, leadership skills.) Judge et al. (2002) classified their leadership Good to Great criteria in terms of both emergence and effec- Most business books are empirical nonsense, tiveness. Their results showed that all five di- Good to Great , but Collins’s (2001a) book, mensions were related to overall leadership seems to be an exception. He and his staff (emergence and effectiveness combined), with searched databases for the Fortune 1000 com- true correlations of .24 or greater for each, ex- panies to identify companies that had 15 years R cept for Agreeableness (.08). The multiple of performance below the average of their busi- value for all five dimensions predicting emer- ness sector and then 15 years of sustained per- gence was .53, and it was .39 for predicting formance significantly above the average of their criterion of effectiveness (see Hogan & their sector. They found 11 companies that fit Hogan, 2002, and Lord et al., 1986, for simi- this profile. The next question was, what distin- larly strong relationships between leadership guished these 11 companies? Their somewhat and personality). reluctant conclusion was that the distinguishing feature was a new CEO who took charge of the Does Leadership Matter? organization and then improved its performance. It is useful to know that personality predicts These 11 CEOs all shared the same two char- indices of leadership effectiveness, but does acteristics (above and beyond the four elements leadership actually matter in terms of the per- described earlier; Collins, 2001b). First, they formance of an organization? And, if it does, were modest and humble, as opposed to self- then what are the mechanisms? The answer to dramatizing and self-promoting. Second, they were phenomenally, almost preternaturally, per- the first question is yes; the relevant data come sistent. These findings were a jolt to the busi- from studies of the economic utility of senior ness literature (which had been promoting the managers. For example, Joyce, Nohria, and Roberson (2003) reported that CEOs account cult of the charismatic CEO), but we think they for about 14% of the variance in firm perfor- make sense in terms of the data provided by
7 SPECIAL ISSUE: WHAT WE KNOW ABOUT LEADERSHIP 175 mance. To put this number in perspective, in- 198,514 employees from 7,939 business units, they showed that employee engagement and dustry sector accounts for about 19% of that variance (McGahan & Porter, 1997). In addi- satisfaction, at the business-unit level, corre- lated .37 and .38, respectively, with a composite tion, Barrick, Day, Lord, and Alexander (1991) showed that, relative to executives with average index of business-unit performance that in- cluded turnover, customer loyalty, and financial performance, high performers provide an addi- tional $25 million in value to an organization performance. Putting these various studies together, we see during their tenure (see also Day & Lord, 1988, that (a) personality predicts leadership style and Thomas, 1988, for evidence regarding the (who we are determines how we lead), (b) lead- financial impact of leaders on organizations). ership style predicts employee attitudes and Concerning the question of how leaders in- team functioning, and (c) attitudes and team fluence the performance of their organizations, the general model is that leader personality in- functioning predict organizational performance. This model linking leader personality to orga- fluences the dynamics and culture of the top nizational performance is portrayed in Figure 1. management team, and the characteristics of the top management team influence the perfor- mance of the organization. Two very interesting Managerial Incompetence articles provide data to support these themes. In Although the literature on managerial com- the first, R. S. Peterson, Smith, Martorana, and petence is sparse and fragmented (but growing), Owens (2003) used data from CEOs of 17 very the literature on managerial incompetence is large corporations (e.g., IBM, Coca-Cola, Dis- remarkably coherent. The problem is very im- ney, Xerox, CBS, Chrysler, and General Mo- portant; survey after survey shows that 65%– tors) to show that CEO personality powerfully 75% of the employees in any given organization affects the dynamics and culture of the top report that the worst aspect of their job is their management team, with correlations in the .50 immediate boss. Estimates of the base rate for range for most hypothesized relationships be- managerial incompetence in corporate life tween personality and various aspects of team range from 30% to 75%; a recent review re- functioning (e.g., cohesiveness, corruption, and ported the average estimate to be 50% (DeVries risk tolerance). Moreover, the characteristics of & Kaiser, 2003). Historically, managerial in- the top management team were substantially competence has been conceptualized in terms of correlated with business outcomes such as in- not having the characteristics needed for suc- come and sales growth, return on investment, cess, that is, too little of the right stuff. We and return on assets. believe that failure is related more to having In the second article, Harter, Schmidt, and undesirable qualities than to lacking desirable Hayes (2002) reviewed the literature on em- ones, that is, having the wrong stuff. ployee satisfaction and showed that satisfaction Bentz (1985) pioneered the study of manage- means, in essence, satisfaction with supervisors. rial incompetence with an interview study of That is, how employees view their supervisors failed managers at Sears; he noted that virtually is the primary determinant of their overall sat- all of them had a “personality defect” of some isfaction. Then, in a meta-analysis, including Figure 1. How leader personality affects organizational performance.
8 HOGAN AND KAISER 176 sort. Bentz’s findings were then replicated by Scher (1988) reported that the distinguishing researchers at the Center for Creative Leader- feature of most forms of self-defeating behavior ship (McCall & Lombardo, 1983) and others. is the pursuit of short-term gains that carry Leslie and Van Velsor (1996) summarized the significant long-term costs (see Table 2 for literature on failed managers in terms of four other examples of this dynamic). themes: (a) poor interpersonal skills (being in- Second, although high scores on the 11 dark sensitive, arrogant, cold, aloof, and overly am- side dimensions shown in Table 2 are associated bitious), (b) unable to get work done (betraying with negative consequences in the long run, low trust, not following through, and being overly scores are not necessarily desirable either; this ambitious), (c) unable to build a team, and (d) is what makes personality psychology so inter- unable to make the transition after a promotion. esting. Low levels of dutifulness suggest prob- After reviewing this literature, Hogan and lems with authority; low levels of imaginative- Hogan (1997) proposed that the standard per- ness suggest lack of vision; low levels of bold- sonality disorders, as described in the Diagnos- ness suggest indecisiveness; and so on. tic and Statistical Manual of Mental Disorders Optimum performance is associated with more 4th edition ( ; American Psychiatric As- DSM–IV moderate scores. Kaplan and Kaiser have ap- sociation, 1994), provide a taxonomy of the plied this reasoning to executive assessment; most important causes of managerial failure. their data clearly show that there is an optimal Personality disorders are not forms of mental level for most managerial behaviors (e.g., illness; they are dysfunctional interpersonal dis- Kaplan & Kaiser, 2003). positions that (a) coexist with talent, ambition, The third point concerns how executive se- and good social skills and (b) prevent people lection decisions are made (Sessa, Kaiser, Tay- from completing the essential task of leader- lor, & Campbell, 1998). Most formal selection ship: building a team. These dysfunctional dis- tools are rarely used. Former subordinates— positions are what we described earlier as the those who are best able to report on a person’s dark side of personality. Hogan and Hogan de- talent for leadership—are almost never con- veloped an inventory of the 11 key dimensions sulted. Often new executives are recruited from of the dark side using the DSM–IV Axis II outside the organization, making it even more personality disorders as a guide. The inventory difficult to evaluate the candidate appropriately. is intended to predict managerial failure, and The most common selection tool is an inter- subsequent research shows that it does (Hogan view, and the dark side tendencies are designed & Hogan, 2001). This taxonomy is presented in to create favorable immediate impressions; nar- Table 2. cissists and psychopaths excel during inter- There are three points to note about these views. We speculate that many executives are dark side characteristics. First, they are hard to hired for the very characteristics that ultimately detect, for two reasons. On the one hand, they lead them to fail. coexist with well-developed social skills (Hogan & Hogan, 1997, 2001). On the other hand, these tendencies, although flawed, are in- Organizational Effectiveness tended to make a positive impression on others, The professional literature in psychology has and they do in the short run. For example, very little to say about the determinants of or- people with high scores on the Bold scale (nar- ganizational effectiveness. Perhaps the best- cissism) initially seem confident and charis- known treatment of the subject is provided by matic. Over time, however, these features turn Katz and Kahn (1978). After noting how com- into a sense of entitlement and an inability to plicated the subject is, Katz and Kahn suggested learn from mistakes. Paulhus (1998) reported idio- effectiveness organizational defining that, in an unstructured group task in which the graphically, in terms of how efficiently an or- participants are strangers, narcissism predicts ganization converts its resource inputs into out- making a strong initial impression and being puts. This definition is internally consistent but nominated as a leader but subsequently being ignores the fact that organizations are in com- rejected by the group as a result of arrogance petition with one another. and high-handedness. Indeed, Baumeister and
9 177 SPECIAL ISSUE: WHAT WE KNOW ABOUT LEADERSHIP Long-term weakness(es) meanness manages by crisis insensitive to morale issues exploits others of entitlement vision, erratic decision making volatility micromanaging about pleasing superiors Unable to admit mistakes; sense Displays outbursts and emotional Impulsive, attention seeking; Uncommunicative and Indecisive and risk averse Lies, defies rules and authority, Displays passive–aggressive Indecisive, overly concerned Fanciful; displays over-the-top Overcontrolling, rigid, Mistrustful, vindictive, litigious Short-term strength(s) politics boss informed thinking self-sacrificing Team player; considerate; keeps Willing to take risks, charming Makes few mistakes Charming with good social skills Displays visionary outside-the-box Tough and resolute under pressure Entertaining, flirtatious, engaging Energetic and enthusiastic Courageous, confident, charismatic Hard working; has high standards; Insightful about organizational Definition and projects; fails to follow through grandiosity and entitlement; unable to odd or eccentric ways attention wants to be noticed and the center of inflexible, intolerant of ambiguity learn from mistakes to others’ feelings procrastinating, stubborn, resentful of requests for increased performance being criticized argumentative and combative and exploitive bright, manipulative, deceitful, cunning, Acts and thinks in creative but sometimes Overtly cooperative, privately Excessively takes risks and tests limits; Cynical, mistrusts others’ intentions, Moody, intense, easily annoyed by people Reluctant to take risks as a result of Expressive, animated, and dramatic; Excessively self-confident; exhibits Meticulous, precise, perfectionistic, Conforms and is eager to please superiors Aloof and uncommunicative, insensitive Axis II disorder Borderline Paranoid Schizotypal Antisocial/psychopathic Passive–aggressive Avoidant Histrionic Schizoid Dependent Obsessive–compulsive Narcissistic Dimension Table 2 Dimensions of Managerial Incompetence Excitable Imaginative Dutiful Diligent Leisurely Cautious Bold Skeptical Colorful Mischievous Reserved
10 HOGAN AND KAISER 178 Our final point does not concern a reliable leadership is the key to organizational effective- empirical leadership; ness. Consequently, every organization makes generalization about hiring mistakes, every organization alienates at rather, it proposes a model for conceptualizing least part of its workforce, every organization organizational effectiveness. However, organi- zational effectiveness is an organic part of any has its share of bad managers, many organiza- discussion of leadership when leadership is seen tions pay only lip service to strategy formula- as a collective phenomenon, a resource for the tion, and many organizations fail to monitor their own performance in these key areas. Thus, performance and survival of a collectivity. In every organization has its inefficiencies. As our view, organizational effectiveness can be Pericles said to the elders of Athens on the eve conceptualized in terms of five components. The first component of organizational effec- of their cataclysmic war with Sparta, “I care less about the Spartans’ strategy than I do about our tiveness is talented personnel. Other things be- ing equal, a more talented team will outperform mistakes.” a less talented team. Talented personnel are identified through good selection methods and References recruited through good leadership. The second component of organizational effectiveness is Diagnos- American Psychiatric Association. (1994). motivated personnel: people who are willing to tic and statistical manual of mental disorders (4th perform to the limits of their ability. Other ed.). Washington, DC: Author. things being equal, a motivated team will out- Avolio, B. J., Sosik, J. J., Jung, D. I., & Berson, Y. perform a demoralized team. The level of mo- (2003). Leadership models, methods, and applica- tions. In W. C. Borman, D. R. Ilgen, & R. J. tivation in a team or organization is directly Handbook of psychology Klimoski (Eds.), related to the performance of management (Har- (Vol. 12, pp. 277–307). Hoboken, NJ: Wiley. ter et al., 2002). Barrick, M. R., Day, D. V., Lord, R. G., & Alex- The third component of organizational effec- ander, R. A. (1991). Assessing the utility of exec- tiveness is a talented management team, with Leadership Quarterly, 2, utive leadership. 9 –22. talent defined in terms of the domain model Bass and Stogdill’s handbook of Bass, B. M. (1990). presented in Table 1 (and incompetence defined (3rd ed.). New York: Free Press. leadership in terms of the taxonomy presented in Table 2). Baumeister, R. F., & Scher, S. J. (1988). Self-defeat- The fourth component is an effective strategy ing behavior patterns among normal individuals. for outperforming the competition. This is Psychological Bulletin, 104, 3–22. Bentz, V. J. (1985, August). A view from the top: A where many organizations have problems. An thirty year perspective of research devoted to the effective strategy depends on systematic re- discovery, description, and prediction of executive search and a deep knowledge of industry trends. Paper presented at the 93rd Annual Con- behavior. But business managers do not enjoy research vention of the American Psychological Associa- (otherwise, they would be in the research busi- tion, Los Angeles, CA. ness), and people who enjoy research do not Boehm, C. (1999). Cam- Hierarchy in the forest. talk frequently with business managers. As a bridge, MA: Harvard University Press. result, business strategy is often developed on Boyatzis, R. E. (1982). The competent manager. New an ad hoc basis by top management teams (think York: Wiley. about the strategy that has been instituted at New York: Har- Collins, J. (2001a). Good to great. perCollins. your place of employment and how it was Collins, J. (2001b). Level 5 leadership: The triumph developed). Harvard Business of humility and fierce resolve. The final component of organizational effec- 66 –76. Review, 79, tiveness is a set of monitoring systems that will Day, D. V., & Lord, R. G. (1988). Executive leader- allow senior leadership to keep track of the ship and organizational performance. Journal of talent level of the staff, the motivational level of Management, 14, 453– 464. the staff, the performance of the management de Hartog, L. (2000). Genghis Khan: Conqueror of group, and the effectiveness of the business London: Tauris Academic. the world. strategy. It is the responsibility of the senior DeVries, D. L. (1993). Executive selection: A look at leadership in an organization to put these five what we know and what we need to know. Greens- boro, NC: Center for Creative Leadership. components in place. Ultimately, then, good
11 SPECIAL ISSUE: WHAT WE KNOW ABOUT LEADERSHIP 179 DeVries, D. L., & Kaiser, R. B. (2003, November). Kaplan, R. E., & Kaiser, R. B. (2003). Developing . Paper presented at the Going sour in the suite MIT Sloan Management Re- versatile leadership. Maximizing Executive Effectiveness workshop, view, 44 (4), 19 –26. Miami, FL. Katz, D., & Kahn, R. L. (1978). The social psychol- Dirks, K. T., & Ferrin, D. L. (2002). Trust in lead- (2nd ed.). New York: Wiley. ogy of organizations ership: Meta-analytic findings and implications for The leader- Kouzes, J. M., & Posner, B. Z. (2002). research and practice. Journal of Applied Psychol- (3rd ed.). San Francisco: Jossey- ship challenge ogy, 87, 611– 628. Bass. Frei, R. I., & McDaniel, M. A. (1998). Validity of Leslie, J. B., & Van Velsor, E. (1996). A look at customer service measures in personnel selection. derailment today. Greensboro, NC: Center for Cre- 1–27. Human Performance, 11, ative Leadership. French, J. R. P., Jr., & Raven, B. H. (1959). The Lord, R. G., DeVader, C. L., & Alliger, G. (1986). A bases of social power. In D. Cartwright (Ed.), meta-analysis of the relation between personality (pp. 150 –167). Ann Arbor, Studies in social power Journal of Applied traits and leader perceptions. MI: Institute for Social Research. 402– 410. Psychology, 71, Harter, J. K., Schmidt, F. L., & Hayes, T. L. (2002). Lord, R. G., Foti, R. J., & DeVader, C. L. (1984). A Business-unit-level relationship between employee Organiza- test of leadership categorization theory. satisfaction, employee engagement, and business tional Behavior and Human Performance, 34, outcomes: A meta-analysis. Journal of Applied 343–378. Psychology, 87, 268 –279. Mann, R. D. (1959). A review of the relationship Hogan, J., & Holland, B. (2003). Using theory to between personality and performance in small evaluate personality and job performance rela- 241–270. groups. Psychological Bulletin, 66, tions: A socioanalytic perspective. Journal of Ap- Marcuse, H. (1969). Eros and civilization. London: plied Psychology, 88, 100 –112. Penguin. Hogan development Hogan, R., & Hogan, J. (1997). McAdams, D. (1995). What do we know when we survey manual. Tulsa, OK: Hogan Assessment 365– know a person? Journal of Personality, 63, Systems. 395. Hogan, R., & Hogan, J. (2001). Assessing leadership: McCall, W. M., & Lombardo, M. M. (1983). Off the A view from the dark side. International Journal track: Why and how successful executives get de- 40 –51. of Selection and Assessment, 9, Greensboro, NC: Center for Creative Lead- railed. Hogan, R., & Hogan, J. (2002). Leadership and so- ership. ciopolitical intelligence. In R. E. Riggio, S. E. McClelland, D. C. (1973). Testing for competence Murphy, & F. J. Pirozzolo (Eds.), Multiple intelli- rather than intelligence. American Psycholo- (pp. 75– 88). San Francisco: gences and leadership gist, 28, 1–14. Jossey-Bass. McGahan, A. M., & Porter, M. E. (1997). How much Hogan, R., & Warrenfeltz, R. (2003). Educating the does industry matter, really? Strategic Manage- modern manager. Academy of Management Learn- 15–30. ment Journal, 18, ing and Education, 2, 74 – 84. Mintzberg, H. (1973). The nature of managerial House, R. J. (1971). A path-goal theory of leader New York: Harper & Row. work. effectiveness. Administrative Science Quar- Ones, D. S., Viswesvaran, C., & Schmidt, F. L. terly, 16, 321–338. (1993). Comprehensive meta-analysis of integrity Joyce, W. F., Nohria, N., & Roberson, B. (2003). Journal of Applied Psychology, 78, test validities. 2 formula for sus- What really works: The 4 679 –703. tained business success. New York: Harper Busi- Paulhus, D. L. (1998). Interpersonal and intrapsychic ness. adaptiveness of trait self-enhancement. Journal of Judge, T. A., & Bono, J. E. (2001). Relationship of 197–208. Personality and Social Psychology, 74, core self-evaluation traits—self-esteem, general- Peterson, C., & Seligman, M. E. P. (2004). Character ized self-efficacy, locus of control, and emotional strengths and virtues: A handbook and classifica- stability—with job satisfaction and job perfor- Washington, DC: American Psychological tion. mance: A meta-analysis. Journal of Applied Psy- Association. chology, 86, 80 –92. Peterson, R. S., Smith, D. B., Martorana, P. V., & Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. Owens, P. D. (2003). The impact of chief execu- (2002). Personality and leadership: A qualitative tive officer personality on top management team and quantitative review. Journal of Applied Psy- dynamics. Journal of Applied Psychology, 88, 765–780. chology, 87, 795– 808.
12 HOGAN AND KAISER 180 London: Bodley Pocock, T. (1987). Horatio Nelson. with leadership: A survey of the literature. Journal Head. 35–71. of Psychology, 25, Renaud, H., & Estes, F. (1961). Life history inter- Thomas, A. (1988). Does leadership make a differ- views with one hundred normal American males: ence to organizational performance? Administra- American Journal of Pathogenicity of childhood. tive Science Quarterly, 33, 388 – 400. Orthopsychiatry, 31, 786 – 802. Vinchur, A. J., Schippmann, J. S., Switzer, F. S., III, Leadership secrets of Attila the Roberts, W. (1990). & Roth, P. L. (1998). A meta-analytic review of Hun . New York: Warner Books. predictors of job performance for salespeople. Schmidt, F. L., & Hunter, J. (2004). General mental Journal of Applied Psychology, 83, 586 –597. ability in the world of work: Occupational attain- The new lead- Vroom, V. H., & Jago, A. G. (1988). Journal of Personality ment and job performance. ership. Englewood Cliffs, NJ: Prentice Hall. 162–173. and Social Psychology, 86, The five-factor model of per- Wiggins, J. S. (1996). Seligman, M. E. P., & Csikszentmihalyi, M. (2000). New York: Guilford Press. sonality. Positive psychology: An introduction. American Leadership in organizations (4th Yukl, G. (1998). Psychologist, 55, 5–14. ed.). Upper Saddle River, NJ: Prentice Hall. Sessa, V. I., Kaiser, R. B., Taylor, J. K., & Campbell, Executive selection. Greensboro, NC: R. J. (1998). Received June 24, 2004 Center for Creative Leadership. Accepted September 25, 2004 Stogdill, R. M. (1948). Personal factors associated