The Leadership Chain

  • Uploaded by: victor flores
  • 0
  • 0
  • May 2020
  • PDF

This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA


Overview

Download & View The Leadership Chain as PDF for free.

More details

  • Words: 18,919
  • Pages: 58
The Leadership Chain: McClelland and his Legacy By Bradford F. Spencer, Ph.D.

INTRODUCTION Leadership is a much written about topic. The popular business magazines and indeed, some rather well selling books are often focused on specific leaders. In America, we often deify or crucify our leaders, and our business leaders are no exception to this foible. However, today’s research into the true causes of the patterns of what molds leaders does not focus on nearly enough. Certainly the scientific community looks down on the “research methodology.” The fact is the research often consists of interviews by business professors asking prominent captains of industry to speculate on why they are so different. To give credit where it is due, most executives admit they have no clue -- before they continue the dialogue to try to help the struggling researcher find the childhood incident which led to the drive that has set him or her apart. No wonder the conclusions are both suspect and not helpful to those asking what they do to improve themselves. This is where the approach of Dr. David McClelland differed so dramatically. This series of articles is designed as a pre-reading for the workshop (Leadership Process: Motivating Achievement) designed to both explain and cover how to apply the extensive work of Dr. McClelland and his students. For that reason much on detail “how to” implement the findings will be left for the lectures and interactions of the workshop. Our experience is that an overview of this extensive body of knowledge will give the group a running start, thus allowing much more territory to be covered in our limited time. Interestingly enough, the initial research was not about the systemic and predictable causes of leadership at all. Nor was his interest in business, and certainly not business applications. More will be said later as to why this social scientist toiled away in relative obscurity and to this day his name is not recognized by many -- but most of today’s works of substance cite his contributions. Certainly he was recognized as a giant among his peers and his position as head of the School of Social Psychology at Harvard University provided him with a unique platform. He used this platform to forward his work rather than to build fame. The sheer differentiation of businesses and the demands on their leaders in differing times and marketplaces makes comparison a daunting task. The unpredictability of the battlefields they find themselves on compounded by the changing competitive landscape expands the difficulty. Fortunately, we have generally accepted definitions of success as defined by Wall Street and its counterparts in every major free market nation which make it possible to compare results across continents, and indeed, oceans. When people of good will do their best and succeed beyond their peers it is only natural to seek to find if a pattern exists. A primary reason is to create an edge by standing on their shoulders in developing the next generation of successful leaders. It seems the general public has in a rather cynical (but perhaps understandable) fashion bought the myth that the drive is to dominate one’s fellow man -- or simple greed -- is the explanation behind what drives executives to positions of power.

The retailing of this perspective yields no help to the true student. Explaining the common obsessions that brought them to the position where they are surprised by the acclaim and often reluctantly find themselves on the covers of magazines, is the question we are setting out to answer with the implications for every managerial and executive level. Successful leaders recognize at a profound level the article is truly not about them, but rather a way to sell magazines while covering a business story. Pictures of the corporate headquarters do not jump off the newsstands. What this series of articles does not attempt to review is the root causes of the numerous reasons for failure once the pinnacle seems to have been reached. Conflicts so varied and deep as to defy a research process seem to be at work. But we think we have hints as to why the great often fall. However, the reality is that those driven to make businesses succeed tend to stay successful much more often than not. We do think we have a modicum of insight into why that is. Frankly, it does not make for great headlines or even at times for interesting reading. I apologize in advance if this series proves to be a bit dry and suffers from a few too many footnotes and charts. If you seek a study in the “management du jour” techniques, you are reading the wrong paper. What makes “the few” able to consistently elicit more discretionary effort, to hire better people and have systematically more motivated people with better results is what we are attempting to document. We have labeled the phenomenon “The Leadership Chain.”

INITIAL PREMISE Right out of the gate, the initial premise is simple. Without question, the key to beating the odds against today’s challenges has to do with creating an environment that produces an “achievement aroused,” high performing workforce. So what is the job of an effective leader? Is it to provide direction? To motivate? To challenge? “All of the above” may be correct, but they are limited views of what a leader is. Leaders do all these things. But what is the essential aim of leadership? It is to create a climate wherein people want to -- are aroused to -- produce great results, and seem to be driven to do it over and over. This is the job of the leader -- to create conditions wherein others produce great results, and they do it repeatedly. Reproducing these results over time is what produces greatness, in leaders, in organizations, and in people. The job of an effective leader is to stimulate employees to push themselves and accept personal responsibility to meet or exceed tough performance targets. It is to create organizations where people do things above and beyond the call of duty; where people are highly committed to delivering extraordinary results. In other words, great leaders systematically create highly motivated, high achievement aroused organizations.

RESEARCH No body of research and practical application has come close to matching the profound impact of the work of Dr. McClelland, his students, and his many associates. McClelland is footnoted in literally every major work on leadership from Drucker to Peters to Kotter, and is recognized as the foundation upon which most motivational theories are based. Yet he is not well known. The reason his name is not a household word is not from lack of quality, but a genuine humility and commitment to high quality, high impact research, rather than the lecture circuit or publishing best sellers. To explore the implications of this research and practice, we have summarized the research findings and their impact in this article. The end result is not intended to be a formula per se, but hopefully a clear and insightful look at a chain of cause and effect between individual motives, leadership behavior, and organizational climate, and how these dynamics drive organizational effectiveness. Leadership style is the critical link in the chain, connecting climate to motives to achieve organizational results. The findings presented in this article, and the research and practice contained within, form a chain of logic, the results of which provide a powerful guide for developing one’s own leadership and for creating and sustaining high performing organizations. You will learn to recognize what works and what doesn’t, and why.

“great leaders systematically create highly motivated, high achievement aroused organizations”

THE LEADERSHIP CHAIN Consider for a moment the following scenario. You have just accepted the role of CEO of a company that has been struggling for many years. The prior CEO founded the company and, after many years and numerous waves of venture capital investment, was ousted due to an inability to meet the financial projections and commitments. The business concept is a sound one, and the people hired are quite talented. The market is growing and needs your product. Yet the company stands on the brink of extinction. Something needs to be done and you were hired to do the job. The first thing you do is assess the business environment and discover that indeed the company has a good product and the market is ripe. The company has a clear strategy, one which will likely bear fruit. Unfortunately the organization’s climate is in shambles. Good people are leaving, morale has hit rock bottom and the people feel that the future is bleak. Your job is clear--get the company aligned around the strategy and get the people working together toward a shared cause. In other words, create a high performing work climate. The measure will be whether or not you meet the financial projections and commitments.

REVERSE ENGINEERING THE LEADERSHIP CHAIN McClelland and his associates, fortunately, have a lot to say on the subject, and their research provides a clear guide for your leadership mandate. By connecting the dots of his research findings, a compelling chain of reasoning unfolds. We call this the leadership chain. There are four links in the leadership chain. To understand the logic, we will reverse engineer it by looking at the end of the chain, the organization’s results, and then move backward to identify the primary causal logic.

The End of the Chain: Climate and Organizational Performance In the above scenario, the leadership chain is first evidenced by the organization’s performance-how is the company doing in the marketplace? If the company is not producing effectively, one contributing cause is often the result of a culture that is misaligned to the strategy manifested by a depressed organizational climate. By climate, we mean those factors in the culture that affect people’s, motivation, desire and commitment. Because our focus is on the motivation role in leadership we will further specify the definition of climate to “a subset of measurable variables in the culture that directly affect motivation.” As we heard above, morale is low, people are concerned about the future, and we can predict that the organization is likely to feel unfulfilling and unrewarding to be a part of. In other words, the organization’s performance is directly linked to the organization’s climate. We will discuss the quantitative proof later, but anyone who has worked under these circumstances will attest to the relationship between the work environment (which we are narrowing to “climate”) and results. Organizational climates can range from a structured, conforming, impersonal orientation to a more unstructured, collaborative, and personal orientation. Some climates emphasize group accomplishments, while others reward individual achievements. Climate, together with organizational strategy, dictate competitive advantage. While your strategy may be copied, it is extremely difficult to duplicate a culture and the climate. So we have the last link in the chain: organizational climate to organizational performance. A misaligned climate resulting in depressed discretionary effort and collaboration towards goal attainment indisputably affects organizational performance.

Figure 1. The End of the Chain: Climate and

Organizational Performance Climate, Business Strategy and Culture. It is important to understand the link between climate and the “bigger picture” concept of organizational culture. (I have included an appendix -- with greater detail on culture as defined by the significant contemporary research of Dr. William Schneider. The reason it is not included in the body of this series is that while it is extremely important, I am limiting this discussion to the topic of motivation. Thus we can cover the issues we need to under the subset of culture we call climate. I am also leaving out an extensive discussion explaining the strategic “value disciplines” as outlined by Michael Treacy and Fred Wiersema. (For a deeper discussion of these concepts, I refer you to their book Discipline of the Market Leaders.) We will revisit the concept in greater depth, but for now suffice it to say, the theory states that ideally the culture lines up with the strategic imperative of the organization. Strategy for our purposes simply refers to the marketing decisions designed to give the products competitive advantage. If you choose “product leadership” (Intel, Sony) you have no choice but to adopt a “competence culture” to maintain alignment. If you choose to adopt an “operational excellence” strategy (Wal-Mart, WellPoint), your options of culture to remain in alignment is limited to “Control.” If you choose “customer intimacy” as the best strategy, (Southwest Airlines, Avon Products) the culture that will be aligned is “collaboration.” The last strategic option comes from Geoffrey Moore, the high tech marketing guru, and is referred to as “discontinuous innovation.” If that is your option, the only aligned culture is “cultivation.” The model looks like this, but the detail is the topic of another book entirely:

The Middle Of The Chain: Leadership Behavior and Climate While other factors certainly impact an organization’s climate, the most potent impact is the behavioral style of leadership of the CEO and/or the leadership team. When organizations, like the one above, find themselves struggling, missing product deadlines, not responding to market changes, missing opportunities, etc., the most likely leadership response is fear and frustration. Out of fear and frustration, leaders resort to styles such as command and control, or blame. These coping styles, by their very nature, make the situation much worse. A strongly dominating or governing leadership style may not work to foster innovation, and a placating or involving style may not be best for producing timely results. In short, leadership style directly impacts the organization’s climate.

Figure 2. The Middle of the Chain: Behavior and Climate

Leadership

The Beginning of the Chain: Motives and Leadership Behavior Finally, while many factors affect a leader’s style, the most significant variable is the thinking patterns, assumptions, world view, and personality of the leader. In other words, how a leader thinks and what drives his or her behavior directly impacts his or her style. (In these articles, to minimize gender bias, we will use “he” and “she” interchangeably.) We call this thinking pattern and drive--the leader’s motive pattern.

Figure 3. The Beginning of the Chain: Motive Pattern and Leadership Behavior The Complete Leadership Chain The above premises represent, in effect, three connections between four links on a chain of causality that have profound implications for leadership and organization performance (figure 4). They can best be expressed simply as follows: Your motive pattern combined with your ability to express those motives maturely and effectively has a direct impact on your leadership style. Your leadership style in turn directly impacts the organization climate you create, which in turn affects the degree to which people desire to achieve and perform in your organization.

Figure 4. The Complete Leadership Chain

Motive Pattern

Leadership Behavior

Organizational Climate

Organizational Performance

In summary, effective leadership is the act of impacting clear and distinct organization factors, ones that are well within your control. Recognizing motive patterns and arousing motives towards accomplishing organizational results becomes easier when you understand the linkages between motive, leadership style, climate and behavior. Effective leadership can be learned through the direct application of a McClelland’s set of principles.

In the first part of this article, we will explore the first connection of this causal chain in detail and demonstrate the implication of this connection on leadership behavior. Then, each subsequent connection will be examined. Here is the outline for what follows: Part I.

The Relationship Between Motives and Leadership

Part II. The Leadership Chain Continued: The Relationship Between Leadership Style and Culture Part III. The Leadership Chain Completed: The Connection Between Culture and Performance Part IV. The Organizational Culture Alignment Model Part V.

Review and Summary of Conclusions

Appendix: A Brief History of McClelland and his Research Appendix: A brief Summary of the work of Schneider and Culture

Part I THE RELATIONSHIP BETWEEN MOTIVES AND LEADERSHIP Before looking at the relationship between motives and leadership style, let’s first explore in more detail what we mean by “motive.” For our purposes it is neither a dictionary definition nor a reason for a crime.

What Do We Mean by “Motive”? The term “motive” is a specific adjective in the behavioral sciences having to do with the repetitive underlying concerns or needs of a person. When you understand a person’s needs, you understand the recurring types of thinking this person engages in and the behavior that results. As an example, some people think about relationships, and tend to seek social and nurturing settings that follow that thinking pattern. Others are motivated primarily by their need to accomplish things of importance. They often would rather do it alone, if alone helps them meet their goals. To understand why a person acts a certain way, we need to discover their patterns of thinking. Patterns of thinking affect the person’s patterns of behavior. Think for a moment about your first professional job. You probably were nervous, and you probably rehearsed in your mind over and over how you would act early on in the job. You probably even thought a lot about your first day, well before you got there. Why did you think about it? Because you had a concern. You either wanted to do well in the job, that is, be efficient or you wanted to make an “impression,” or you wanted to connect with people, or you wanted to avoid failure, or some combination of the above. These are thinking patterns. They reflect some set of underlying concerns. These concerns drove your thinking pattern and therefore, the patterns in your behavior.

“to understand why a person acts a certain way, we need to discover his or her patterns of thinking”

When we are talking about a motive, then, we are not referring to wants or desires or preferences. We are referring to “psychological needs” which must be satisfied for a person to function well. McClelland, in fact, defines a motive as a “recurrent concern for a goal state (measured in fantasy) which drives, directs and selects a person’s behavior.”

A motive is a “recurrent concern for a goal state (measured in fantasy) which drives, directs and selects a person’s behavior.” (McClelland) Think about a time when you needed to be with someone you cared about. If the need was strong, you probably did whatever it took to be with that person. This was not a preference or a wish, but a strong need, which you took action to fill. It is needs and the motives which reflect the urge to meet those needs that was the focus of McClelland’s research.

How Motives Work McClelland’s initial focus of study was to determine whether or not he could identify within individuals a motive or need, which, if impacted, would significantly increase their productivity. He and his associates did in fact identify such a variable. They called the motive, the need for achievement (described in shorthand as nAch). Such a need, as in all psychological needs, affects the patterns of thinking which drives the patterns of behavior. The fact is, there are no behaviors without precedent thoughts and thus, if you know your patterns of thoughts, your patterns of behavior in a given set of circumstances are predictable. If you have a strong need for achievement, for example, you will likely seek ways of improving past or current performance, you will strive to better yourself, and you will seek to make significant and/or unique contributions to your work. Everyone has this need to some extent. This will occur in all parts of your life and it takes place unconsciously. What is interesting, however, is not that this need exists, but that different people have this need in varying amounts. It is also clear that different jobs require this to varying degrees and thus peoples “fit” varies dramatically. A night watchman with a high achievement motive will go stir crazy, drive others stir crazy with constant suggestions or get the meets met elsewhere. Moreover, one can, in fact, increase this need in others and in oneself. Drawing from a growing body of knowledge on the subject of motives, McClelland sought to understand the relationship between the need for achievement and a wide variety of effects, from the economic success of nations, to behavior of people under the influence of alcohol, to leadership effectiveness. In the latter arena (the subject of this series of articles) McClelland hypothesized that if you could understand and predict the motives that drive effective leaders, and then train people to alter or develop their drives, the result would be that any

organization, and ultimately any society can be significantly enhanced. (For a detailed summary of the history of McClelland’s research, see the insert at the end of this article.) In later studies of people and their motives, McClelland and his associates isolated two other specific motives, which, along with the achievement motive, account for as much as 95% of all social interaction. Since we are interested here in how people behave in organizations or at work, these are most relevant for our exploration. They are the need for affiliation (nAff), and the need for power (nPow). To understand the relationship between motives and leadership, let’s first explore these primary motives more fully.

The Three Primary Motives Achievement (nAch). The need for achievement is characterized by the need to outperform others, high internal standards of excellence, and a desire to innovate. People high in nAch prefer to take responsibility for their actions, set challenging goals, and need a lot of feedback. They tend to be most successful in entrepreneurial settings, and in jobs that require high standards, efficiency, responsibility, and independent work. Their daytime and nighttime dreams (fantasies) are actually about these topics. In Western society “achievement” is a value laden word which usually represents “good” as opposed to “bad.” Seeking the best and most effective way to rob a bank or commit an act of terror is lodged in nAch. In our usage it is the outcome the motive can be judged rather than the drive. Affiliation (nAff). The need for affiliation is characterized by having a strong desire to establish and maintain close relations with others. People high in nAff often tend to be highly cooperative, supportive, and communicative. They tend to be most successful in job settings that require teamwork and integration. Their daytime and nighttime dreams are about establishing relations and avoiding the disruption of these relationships. Again in Western society, “affiliation” is often viewed as a positive trait. For our purposes, the drive comes unattached to “good” or “bad.” Power (nPow). The need for power is characterized by the need to influence others. People whose nPow is high tend to get significant satisfaction in influencing others, are often concerned for their reputation and position. They typically act in ways that generate strong feelings in others, both positive and negative and they influence others naturally through their actions. They tend to prefer jobs where they can have a direct impact on others, and as a result, are often well suited for management positions where success depends on their ability to influence. Their daytime and nighttime dreams are about these topics. Again, in Western societies, “power” is considered a trait that comes loaded with baggage that makes it undesirable. It is the goals of this need to influence that determine the goals as positive or negative, but the “why” behind the behavior is nPow.

Because of our narrow definition, it is not the motive, but the goals which drive the ethics discussion. Importantly, the power motivation in individuals can be divided into two types: A.

Personalized power—the degree to which a person expresses his or her power needs for personal gain and self aggrandizement; B. Socialized power—the degree to which a person expresses his or her power needs to get something done for the greater good. This distinction between types of power is significant, as will be seen later, because the former power drive can often invoke resistance in others and therefore be detrimental to an organization’s performance, while the latter is often essential.

A Hypothetical Scenario In order to demonstrate the three motives in action, let’s look at a scenario to see how three different people in the same company might handle it. These examples are all caricatures and are purposely exaggerated to make a point. Let’s imagine three people are driving to work who are involved in the same project. This is a project in trouble. It’s over budget, behind schedule, in disarray. Later, they will meet with the fourth person, their boss, a vice-president. No one slept well last night. Notice in the description below the different goal states each seeks, their differing thinking patterns, and the way that they direct and select different behavioral strategies to satisfy their dominant motive. Mr. nAch is driving and traffic is a nightmare; cars are moving at a snail’s pace. The radio says there’s a wreck ahead. Within 5 minutes of being stuck, Mr. nAch had identified 2 alternate routes to work and he’s off the freeway in a flash, trying alternate 1. He can’t figure out why in the world all these other people would sit in traffic when there is obviously a better way, after all, they’ve been on this freeway a thousand times, just like him. Now, he’s way ahead of the game, feeling a little better for his efforts. “Sheep,” he mutters to himself. “Not only that,” he thinks, “but I’ve got the solution to the work situation. We’ve got to get the decision process to work better. Expertise is the key. It’s no damn good that we take forever to decide everything. A geriatric unit makes a decision faster than us. Meetings about meetings. If I had the job, I’d get it done in no time. First, I’d reduce the paperwork 75%. No point to most of it. I already have the fastest turnaround of all the managers. I don’t know what they do with all their time. Maybe if they understood my system they could speed the process.” Mr. nAch felt very proud of the system he developed to process requests. It was unique to the best of his knowledge, and gave him a real advantage. Mr. nAch saw the light turn yellow and accelerated. “Made it,” he thought, “now if I keep it at 35, I’ll time ‘em all the way to Broad Street.”

Meanwhile, on another road, Mr. nAff is concerned that his group was demoralized by Mr. nAch’s and Ms. nPow’s tirades. He thinks, “Steve Smith’s wife just had a baby, and Mr. nAch did not take even a moment to congratulate him. He is so preoccupied that he doesn’t even seem to care about people. I just don’t understand people who don’t care about others.” As traffic slows down to a crawl, he thinks about each person’s needs and preferences. “Smith is at a major crossroads in his career. His workload is increasing and yet he is being pulled to be at home more with his family. How will he create the balance he needs in life. Mr. nAch and Ms. nPow are almost unbearable. They are at each others’ throats. You’d think we weren’t on the same team. Yet they each have a legitimate concern, stemming from their highly aggressive goals, which conflict with each other. Unfortunately, this conflict has not been dealt with. “We’ve worked together for years; we don’t want to become enemies,” he’d say to them later. “That’s the whole problem; we’ve lost each other’s support; who can work with all the back stabbing and anger.” Mr. nAff thinks about how he can help them resolve their differences. “I know people, how to get them back together, and once that happens, the rest will be easy. People can think again. No one wants to work when everyone’s angry and resentful of one another; what’s the point?” Relationships are the key. On a different freeway, Ms. nPow was focusing on the company as a whole. “We’ve reached a major crossroads in our company,” she thinks, “and we’ve got to change direction or we will surely perish as a company. She turns her attention to Mr. nAch, and thinks about her frustration with how driven he is to succeed. She appreciates his drive, and sees him as one of the most productive people she has ever met, but also feels like if he succeeds in his sales goals, the company will likely fail. “I can’t seem to get this through to him,” she thinks, “and I wonder how to help him see the bigger picture. If he does, and we can harness his enthusiasm to help affect change in the company, then we have a good chance of success.” Ms. nPow then begins to think about other people in the company and how she can influence them to help the company shift gears toward a new product line and new strategy. These three vignettes are not just about what people think in their cars that day. The kinds of thoughts, the themes of the thought, what each person sees as important and meaningful, will play themselves out repeatedly, throughout these three people’s lives. Mr. nAch has developed what are technically referred to as neuro-networks that unconsciously pattern his thoughts around how to do things better. Mr. nAff’s neuro-networks unconsciously pattern his first concerns around relationships. Ms. nPow’s neuro-networks revolve around influence and having an impact on her world. It is both reasonable to conclude and more importantly, we have empirical evidence to categorically prove, that they will adopt different patterns of behavior.

The Motive “Signature” We describe each of these three social motives in order to accentuate their differences. In fact, everyone has each of these needs in varying degrees, and indeed, one can be high in all three. In other words, the amount of need one has is not a zero sum situation. and the relative

amounts of each of these we have represents our motive profile. Because we each have distinct and different profiles, these differences make up what we might refer to as a motive “signature.” It is this signature that, in part, makes us each unique in terms of our personality. Figure 5. Three Different Motive “Signatures”

90 80 70

Person A

Person B

60 50 Person C 0 30 20 10

n Achievement

n Affiliation

n Power

As will be demonstrated below, these social motives have significant implications for ones leadership style. For example, contrary to popular belief, high nAch people do not always make good leaders. In fact, an individual’s need to achieve can result in behaviors that unintentionally undermine the drive in others to achieve by overly competing, or by overly controlling a given situation. In addition, contrary to popular belief, high nAff leaders also do not always make good leaders. Sometimes, in their strong preference for harmony, for example, they avoid conflict, and sometimes these conflicts are a necessary and important part of organizational life. At the same time, in the right setting, high nAch leaders and high nAff leaders can be very effective. Where immediate results are needed, and where positive models of the achievement drive is needed, high nAch people can often thrive. And where the job requires the leader to integrate different functions and promote teamwork, high nAff leaders can be particularly effective. Ultimately, however, to understand leadership fully, we must understand the nature of power, for it is power and influence that moves people, organizations, and society as a whole when it cannot be done simply by example.

Does a Strong Power Motive Mean You’re a Good Leader? One might be tempted to jump to the conclusion that a strong power (nPow) motive – the desire to influence others -- makes for a good leader. As will be shown, a natural power motive helps,

but isn’t required so long as you develop a conscious ability to influence others. This can evolve over time. To understand power in action as it relates to leadership, an early study by D. G. Winter,1[1] an associate of McClelland’s, is instructive. To test the impact of power on others, Winter examined the effect of Kennedy’s inspiring inaugural address on others. Messages that inspire others are, by definition, designed to influence and are therefore power driven. He found in his study that students exposed to Kennedy’s address were uplifted by his message. After hearing the address, their frequency of power thoughts as measured by a test called the Thematic Apperception Test statistically increased. This is significant from the point of view of leadership because people who are positively influenced by power often feel stronger, more capable, and take more actions to reach organizational aims. To explain his findings, let’s first draw upon McClelland’s explanation of the difference between a high nPow manager versus a high nAch manager. This explanation is taken directly from one of his most famous Harvard Business Review Articles: Power is the Great Motivator.2[2] Consider the case of Ken Briggs, a sales manager in a large U.S. corporation who joined one of our managerial workshops. About six years ago, Ken Briggs was promoted to a managerial position at headquarters, where he was responsible for salespeople who service his company’s largest accounts. In filling out his questionnaire at the workshop, Ken showed that he correctly perceived what his job required of him, namely, that he should influence others’ success more than achieve new goals himself or socialize with his subordinates. However, when asked with other members of the workshop to write a story depicting a managerial situation, Ken unwittingly revealed through his fiction that he did not share those concerns. Indeed, he discovered that his need for achievement was very high - in fact, over the ninetieth percentile - and his need for power was very low, in about the fifteenth percentile. Ken’s high need to achieve was no surprise - after all, he had been a very successful salesman - but obviously his motivation to influence others was much less than his job required. Ken was a little disturbed but thought that perhaps the measuring instruments were not accurate and that the gap between the ideal and his score was not as great as it seemed. Then came the real shocker. Ken’s subordinates confirmed what his stories revealed: he was a poor manager, having little positive impact on those who worked for him. They felt they had little responsibility delegated to them, he never rewarded but only criticized them, and the office was not well organized but was confused and chaotic. On all those scales, his office rated in the tenth to fifteenth percentile relative to national norms. As Ken talked the results of the survey over privately with a workshop leader, he became more and more upset. He finally agreed, however, that the results confirmed feelings he had been afraid to admit to himself or others. For years, he 1[1]

Winter, D. G., Power Motivation in Thought and Action. Unpublished Ph.D. thesis, Harvard University, 1967. McClelland, D. C. and Burnham, D. H., Power is the Great Motivator, Harvard Business Review, 1976, 54, 100111. 2[2]

had been miserable in his managerial role. He now knew the reason: he simply did not want, and he had not been able, to influence or manage others. As he thought back, he realized he had failed every time he had tried to influence his staff, and he felt worse than ever. Ken had responded to failure by setting very high standards - his office scored in the ninety-eighth percentile on this scale - and by trying to do most things himself, which was close to impossible; his own activity and lack of delegation consequently left his staff demoralized. Ken’s experience is typical of those leaders whose strong need for achievement combined with low need for power causes them to be frustrated with others and move toward a more coercive approach to leadership. Although Ken demonstrates power, it was not focused on the needs of others, but on his own needs. As a result, he uses power as a blunt sword. Like many high achievement driven people, Ken was very successful in his individual contributor role and, as a consequence, was promoted into a managerial position for which he, ironically, was probably not well suited. This explains a long misunderstood phenomenon as to why individuals who are highly productive in one job often demonstrate the “Peter Principle” when promoted into a management position.

Power Motive: Good or Bad? A clear eyed view of the concept of motives must be taken to realize that none of the three motives, nAch, nAff, or nPow, are intrinsically “good” or “bad”. A power motive isn’t bad per se, unless it is used to accomplish bad things. Closer to the point is the notion that different motives and motive patterns tend to “fit” different situations better. The many traits of a powermotivated leader tend to fit the requirements of many (but not all) management positions. When it comes naturally, it works best, but the traits of a power motive can also be developed. Unfortunately, because power is such a loaded term for many and in our society often seen as undesirable, often people deny or don’t develop and refine their ability to influence others toward productive organizational outcomes.

What Kind of Power? If achievement motivation does not, by itself make a good leader, asserted McClelland, socialized power likely will. To demonstrate this, 16 sales districts from a major U.S. company were tested and then rated for morale, as measured by the degree to which people were clear about roles, responsibilities, and organizational direction (referred to below as “organizational clarity”) and the degree to which there was strong and effective teamwork (referred to below as “team spirit”). The percentage gains and losses in sales for each district were then compared to their prior year. The difference in sales figures by district ranged from a gain of nearly 30% to a loss of 8% with a median gain of about 14%. The relationship between morale and sales was then ascertained showing a clear positive correlation.3[3] 3[3]

Ibid., p. 114.

Figure 6: Relationship Between Morale and Sales

The motive scores of more than 50 managers of both high and low morale units were then examined, revealing an inescapable pattern. More than 70% of the managers in high morale units scored high in need for power. Moreover, the best of these managers, as judged by the morale of those working for them, tended to score the highest on power motivation. Still, the whole story is not revealed. 80% of the better managers had scores of power motive stronger than their need for affiliation as compared to only 10% of the lower performing managers. This pattern held true throughout the whole company. In the research, product development, and operations divisions, for example, 73% of the better managers had a stronger need for power than a need to be liked, as compared with only 22% of the lower performing managers.

Towards a Leadership Motive Pattern McClelland’s most definitive study demonstrating the impact of the motive patterns on leadership was carried out at AT&T.4[4] In their study, the careers of 311 entry level managers were tracked over a 16 year period. In this study, they hypothesized that managers whose predominant needs were high in socialized power and where those needs were higher than their need for affiliation would likely be more successful than others. We call this the leadership motive pattern – a “signature” of motives best suited for leadership. Their findings indeed reinforced the hypothesis. Those managers who were characterized by the leadership motive pattern at entrance to the company were more likely to be promoted to higher levels of management over time. Moreover, nearly half of the high level management people followed the leadership motive pattern. By way of contrast, those with high nAch profiles, while successful in lower management levels where technical skills were more important, tended to peak in their careers at lower levels and not move up the corporate ladder as readily. In a similar study of successful Navy executive officers, the same leadership motive pattern distinguished the more effective Navy officers from their less effective counterparts. Interestingly, the characteristics of successful petty officers were similar, with two outstanding differences. Successful petty officers saw their work as oriented around getting things done (achieving) and relating effectively to others (affiliation), with less emphasis on the importance of getting things done through others. In other words, they did not see the game of “influence” as important as those that moved up the Navy’s organizational ranks.5[5] To further test the leadership motive pattern, McClelland and his associates distinguished between three kinds of managers—institutional managers, affiliative managers, and personal power managers: *Institutional managers are high in power motivation, low in affiliation motivation, and high in inhibition, that is, able to consciously inhibit their personal ego gratification in lieu of seeking goals for the greater good. 4[4]

McClelland, D. C. and Boyatzis, R. E., Leadership motive pattern and long term success in management. Journal of Applied Psychology, 1982, 67, 737-743. 5[5] Winter, D. G. Navy leadership and management competencies: Convergence among tests, interviews, and performance ratings. Boston: McBer, 1978.

*Affiliative managers have affiliation needs stronger than their need for power. *Personal power managers have high needs for power, higher than their need for affiliation, while their inhibition score is low. In one of McClelland’s most illustrative studies, he examined a Sales Division of a large company whose managers closely matched the three types above. By correlating the type of manager with three different features of an organization’s effectiveness, a clear picture of the efficiency of the institutional manager emerged.6[6] Figure 7: Leadership Behavior: Develop Sense of Responsibility in Subordinates Figure 8: Leadership Behavior: Develop Organizational Clarity

Figure 9: Leadership Behavior: Develop Team Spirit

By further studying the behaviors that institutional managers tend to exhibit, McClelland found five characteristics that seem to distinguish them from the other types. *They are more organization minded; they tend to join more organizations and feel responsible for building them. They also tend to believe in the importance of centralized authority. *They report that they like to work—they like the discipline of work, and they like getting things done in an orderly way. *They are willing to sacrifice their own self-interest to meet the needs of the organization. This is in contrast to the personal power manager whose own personal interests tend to be paramount. *They have a keen sense of justice. They believe in fairness and that people who work hard and sacrifice for the good of the organization should get rewarded for their efforts.

6[6]

Ibid., p. 105.

*They tend to be more mature—less egotistic, less defensive, and more willing to seek advice from experts. In summary, the above research seems to strongly suggest that, to be effective, a leader may need to possess a high need for power directed toward the benefit of the organization as a whole. Moreover, the need for power needs to be greater than his or her need for affiliation. A leader with a high and dominant need for affiliation will have a tendency to want to stay on good terms with all people and will likely make exceptions to corporate policies. In addition, it is often easy to recognize the affiliation motive in their desire to avoid interpersonal conflict. The desire to please will often result in an inability to provide focus and direction, and thereby cause confusion and loss of confidence. In addition, in their effort to please employees, they will likely apply policies differentially, causing mistrust on the part of others.

“to be effective, a leader may need to possess a high need for power directed toward the benefit of the organization as a whole”

To the undiscerning eye, these findings might be interpreted to conclude that to be an effective leader, one must have a high need for power, and without it, you will likely be ineffective. This would be an overstatement of the case. Rather, the findings suggest that power is (unconsciously) a primary factor in most leaders’ effectiveness, and the effective and mature expression of power is likely to yield positive results in many, if not most, leadership situations.

“…power is (unconsciously) a primary factor in most leaders’ effectiveness, and the effective and mature expression of power is likely to yield positive results in many, if not most, leadership situations.”

What If You Have a Low Power Motive? If the need for power is not present in your motive pattern, and it is needed for you to be an effective leader, the game is not lost. The challenge for you may be to consciously think more about how to influence others, a thinking pattern that others high in nPow do more subconsciously. This will enable you to cultivate your ability to express power more effectively and thereby level the playing field in terms of producing more influence behaviors.

While McClelland’s findings clearly suggest that people with high nPow tend to be more effective as leaders, the behaviors that are exhibited by high nPow individuals are paradoxically often seen as undesirable in our society. As a society, we tend to have contradictory feelings about power. We both want it, and yet don’t want to be affected by others who want it. We respect people who get things accomplished, yet often resent their approach. The distinction between personalized power and socialized power is particularly important here because it sheds light on the paradox. People who are high in personalized power tend to see the world as a win-lose game. “Survival of the fittest” is their credo, and they will do whatever it takes to get ahead. They tend to like prestige and, in the extreme, tend to be impulsive in their application of their power needs. This invites resistance to them and their goals. In contrast, people who are high in their socialized power needs often exercise their power for the benefit of others. They seek win-win solutions to conflict and tend to effectively suppress their inner need for domination. Hence, they tend to be much more effective in their organizational leadership than their personal power counterparts and exhibit behaviors that are, more often than not, seen as socially desirable. This invites support to both them and their goals. This distinction became particularly compelling to McClelland later in his research career when he became interested in the question of whether and how power needs differ as people mature. In exploring the question, he distinguished between four different stages of maturity as they relate to power needs.

Figure 10: Relation of Power Motivation and Inhibition to Maturity and Leadership Potential7[7] Leadership Leadership Power Needs Maturity Tendencies Level

Often not enough Stage I assertive Low power motivation

7[7]

McClelland, D.C. “Power Motivation and Organizational Leadership.” In Power: The Inner Experience, by D. C. McClelland, New York: Irvington Publishers, 1975.

Often overlyMotivation dominant High Power Stage II (particularly with low nAff) Low Inhibition High Power Motivation Often overly efficient Stage III High Inhibition (particularly with low nAff) High Power Motivation Stage IV High Inhibition Often selfless and efficient High Stage IV scores

McClelland hypothesized that these stages, in effect, form a natural hierarchy of power, where each stage represents a “more mature” approach to power and influence and will therefore result in greater leadership capability. He formed his hypothesis by observing and comparing people who used fundamentally different leadership styles and their relation to power. Later, research on how people progress through different stages of ego maturity and their relationship to power and influence strongly corroborated McClelland’s assertion. In each of the studies described below, ego maturity is measured by the degree of complexity in ones thinking, the ability to differentiate among multiple feelings, the ability to recognize and value differences between ones view and that of others, and the ability to focus on the long term, rather than strive solely to meet immediate needs. In a study by S. Smith,8[8] she found that people with more mature egos tended to be less driven by their needs for personal power and control and more by their need to focus on the needs of others and the organization as a whole. In contrast, people with less developed egos tend to exercise coercive power—that is the need to control others and their actions. As managers they tend to prefer to enforce decisions rather than make them. They tend to think in terms of right or wrong, and are often extremely rule bound. In a similar study, H. Lasker9[9] found a clear relationship between ego maturity and motive patterns. He found that managers at lower stages of ego maturity tend to have stronger personal power drives while managers with greater ego maturity tend to be more driven by their needs for achievement. Inspired by these findings, W. Torbert found a strong correlation between ego maturity and progression up the corporate ladder of success.10[10] In another study, W. Torbert and D. Fisher found that leaders with higher ego development were far more capable of exercising socialized power and leading organizational change efforts than their less mature counterparts.11[11]

8[8]

Smith, S., Development and the Problems of Power and Agreement in Organizations, Unpublished doctoral dissertation, George Washington University, 1980. 9[9] Lasker, H., Ego Development and Motivation: A Cross-Cultural Cognitive-Development Analysis of Achievement, Unpublished doctoral dissertation, University of Chicago, 1978. 10[10] Torbert, W., “Leading Organizational Transformation,” Research in Organizational Change and Development, Vol. 3, pages 83-116, 1989, JAI Press Inc. 11[11] Fisher, D. and Torbert, W., “Transforming Managerial Practice: Beyond the Achiever Stage,” Research in Organizational Change and Development, Volume 5, pages 143-173, 1991, JAI Press Inc.

Daniel Goleman and his colleagues at the Consortium for Research on Emotional Intelligence in Organizations at Rutgers University, have re-labeled nPower as “Social Skills,” in his Harvard Business Review article. He defines this as “friendliness with a purpose, moving people in the direction you desire.” This is as close to McClelland’s definition of Power as one can get.12[12] The above research, while strong in their implications of which profile will likely result in more effective leadership, does not suggest that only one motive pattern will yield leadership success in all cases. Other motive profiles may also be more likely to be effective in different situations. For example, the role of integrator is a critical organizational function, especially important in organizations that require cross functional communication. In this case the integrative leader is one who facilitates communication across organizational lines and helps groups work through conflict. In a study by P. Lawrence and J. Lorsch, they found that effective integrators differed from their less effective counterparts by their higher need for affiliation.13[13] They pay more attention to others and their feelings, they try harder to establish friendly relationships in meetings, and they take on more assignments that offer opportunities for interaction. In addition, they tend to have average, but not high needs for achievement as well as average, but not high needs for power. In other words, effective integrators have balanced motive profiles and their affiliation needs are higher than those that are ineffective integrators. McClelland drew a similar conclusion in a study of 96 Navy Human Relations officers, where the most successful ones displayed moderate and balanced needs for Achievement, Affiliation, and Power. He called this the integrator motive profile.14[14]

Power Is Great For Leaders, But What About Affiliation? As organizations have become less hierarchical, it appears that the implications of McClelland’s early research for today’s leadership may need to be revisited. While the ability to influence others continues to be a critical source of leadership effectiveness, the ability to establish collaborative relationships has become more and more important in the past decade. Hence, some combination of socialized power with medium affiliation drives may be the profile needed to succeed in many of today’s flatter, more collaborative environments. As leaders need to become more facilitative rather than authoritative in their orientation, so too must their profile change. This flies in the face of McClelland’s earlier assertion that in order to be effective, a leader has to have need for power stronger than their need for affiliation. This may still be true, yet a potent dose of nAff is extremely important to be a good facilitative manager. So which is it? Does a good leader need to be low in affiliation needs relative to power, or does he or she need a medium or even higher amount of nAff? And where do standards come from if achievement is not high? To shed light on this issue, R. Boyatzis, former President of McBer, the consulting firm founded by David McClelland, made a distinction between two types of affiliation needs—affiliative 12[12]

Goleman, D., “What Makes a Leader,” Harvard Business Review, Nov.-Dec., 1998, pages 93-102. Lawrence, P. R. and Lorsch, J. W., “The New Management Job: The Integrator,” Harvard Business Review, 1967, 45, pp. 142-151. 14[14] Winter, D. G. Navy leadership and management competencies: Convergence among tests, interviews, and performance ratings. Boston: McBer, 1978. 13[13]

assurance and affiliative interest. The need for affiliative assurance is based on the need to be liked and approved by others and often results in behaviors designed to avoid rejection. The need for affiliative interest, in contrast, is based on a genuine concern for others and their well being and often results in “joining” behaviors such as being involved in teams, social clubs, etc. One motive is self-focused (affiliative assurance), the second is other-focused (affiliative interest). In a working paper on the subject, Boyatzis hypothesized that the affiliative assurance manager will likely behave in ways that are counterproductive to organizational needs.15[15] He would make decisions seeking acceptance and approval from others rather than make tough decisions. He would equate harmony with success rather than performance with success, and would feel uncomfortable confronting others for fear that they would lose others’ approval. In contrast, a manager driven by affiliative interest will likely be compassionate, yet direct in their decisionmaking and less concerned about the consequences of making tough decisions. While there is no direct research to date on the subject, C. Lafferty and his associates have for many years been studying the relationship between 12 different personality traits and leadership effectiveness. Their research has strongly corroborated Boyatzis’ hypothesis. Two of Lafferty’s 12 personality traits make the same distinction as Boyatzis, and in multiple studies, Lafferty found that strong leaders exhibited low needs for approval, and high needs for affiliation—that is, a high interest in and care for the needs of others. Moreover, they tended to express that concern as a leader through coaching of others toward increased performance. 16[16] This is distinct from coaching that is driven by the desire to maintain a relationship.

Conclusion – Motives and Leadership When taken as a whole, the research on motives as it relates to leadership forms a compelling guide. The research is telling us that people do have different needs, perceptions, and drives. At a deep unconscious or pre-conscious level these form the basis for their leadership style. We have come to refer to this as being on “automatic pilot.” While no style is perfect, and each style can be effective in different situations, the task of a leader is quite clear: to get the most out of ones people, and to sustain their performance over time. To do this, a leader must have influence, and must think about the motives of his or her people, and how they will respond to his or her leadership differently based on these motives. This influence must be directed toward the greater good of the whole or people will resist it. Moreover, leaders’ attention to influence must outweigh their need to be liked. If their need for approval is too strong they will not make difficult decisions, ones that are unpopular. And these decisions sometimes have to be made for the greater good of the whole. Finally, a leader has to have some achievement drive, which produces the thinking patterns from which high standards emanate.

We haveMotivation: come toArefer asPerspective.” being onIn Human Motivation: A Book of Boyatsis, R. E., “Affiliation Reviewto andthis a New Readings, Edited by D. C. McClelland“automatic and R. S. Steele.pilot.” General Learning Press, Morristown, NJ 1973. 16[16] For further information on the subject, contact Clayton Lafferty at Human Synergistics, Plymouth, Michigan. 15[15]

These leadership principles do not stand alone. When applied in a mature fashion, a leader can directly and significantly impact the climate of the organization as a whole. Certain leadership styles impact the climate differently, which in turn affects and directs the attention of people in the organization. In the following section, we will look at this relationship, and begin to see a clear picture of the kind of leadership style that most powerfully affects and creates an organizational climate for success.

Part II

THE LEADERSHIP CHAIN CONTINUED: THE RELATIONSHIP BETWEEN STYLE AND CULTURE

Organizations and their cultures do not just emerge overnight. They build and evolve over time. They evolve as a result of the personality of the leaders, the environment in which the organization exists— its geographical location, its industry, etc.-- and the key decisions it makes that determine its history. We define “culture” as “the way we do things in this company in order to succeed.” As previously stated, we are focusing on climate variables which we define as “the subset of measurable variables in the culture that affect motivation.” These are the factors that influence an organization’s effectiveness and its results over long periods of time. Of all the factors that determine how and why a particular organizational culture emerges, perhaps the most powerful impact is the style of the leader. The leader of an organization, by virtue of his or her personality and behavior, makes an indelible mark on the organization’s culture, either consciously or unconsciously. People in organizations are constantly adjusting to the expectations of the leader. They are looking to the leader for signs of what is okay to do or not to do, they pay attention to what the leader ignores and what the leader focuses on, and they are following the example she provides. In other words, a leader cannot help but directly impact the culture/climate of her organization. Witness the case of Southwest Airlines. As a company, its people are known for their enthusiasm and playfulness, and unique way of operating. When you fly on one of their airplanes, its spirit is palpable. Etched on the walls are its motto: “nourished by our people’s indomitable spirit, boundless energy, immense goodwill and burning desire to excel.” Anyone who has ridden on a Southwest Airlines plane knows what we mean here as they are treated to playful, humorous, and often counter-industry comments of the flight attendant manning the microphone during takeoff and landings. Not surprisingly, the organizational climate of Southwest Airlines mirrors the personality of its leader—Herb Kelleher. Kelleher is known far and wide as a fun loving, aggressive, hard drinking president, who is adulated by the people in the company for his good heart, and playful leadership. Hence, in this example, and countless others throughout corporate America, the quality of leadership and the character of an organization’s culture appear to be inexorably linked.17[17] In this, the second part of the article, we explore the second link of the leadership chain: namely the relationship between a leader’s style and the climate of the organization. It is this link that we are most familiar intuitively, yet at the same time, we fear. When we were at the individual contributor level in the organization, we blamed upper management for what ails the corporation. Now in the role of leader, we have no where to turn to blame the dysfunctional features of the organization except ourselves.

17[17]

While linked, it does not follow that an organization’s culture is only driven by the leader’s style of leadership. Factors such as the nature of the industry, the organization’s long term strategic philosophy, and its past history of success, may be more critical to forming a successful culture.

While we use well known publicized case studies such as Southwest Airlines (Herb Kellerher), WalMart (Joe Walton), IBM (Thomas Watson) and Microsoft (Bill Gates) to demonstrate the forceful impact a personality has on the organization, it is often less clear in your system. That is not because it is less impactful, but rather that it is hard to distinguish day to day behaviors we become used to as seminal in the organic behavior of the entire system.

What the Research Shows One of the earliest attempts to understand the linkage between leadership and organizational climate was the research of two colleagues of McClelland’s: G. Litwin and R. Stringer. In their work, Litwin and Stringer began with the assumption that the success of an organization is dependent on the behaviors of its people, and that the most powerful determinant of those behaviors is the degree to which people’s motives are aroused. An aroused motive is like an itch that needs to be scratched. When a motive is aroused, people behave to meet the needs of that motive. For example, if affiliation is aroused, then people will want to be with others, to interact socially. If a power need is aroused, they will want to control or influence others. If achievement is aroused, people will want to be more productive, innovative, and goal directed—to achieve something of value. Of the three motives, the achievement motive is most likely to be correlated with individual performance. In other words, the more the achievement motive is aroused in individual contributors, the greater the impact on organizational productivity. Based on this aroused motive premise, Litwin and Stringer then hypothesized that certain climates tend to stimulate or suppress certain motives and that different leadership styles will create different climates. Their causal chain of logic – quite similar to that of the Leadership Chain -- is explained in the following figure (figure 11). Figure 11: The relationship between leadership style and results

To test their hypotheses, Litwin and Stringer began to experiment with a number of different variables to measure organizational climate. They drew the conclusion there are a discreet set of characteristics that are most critical to the productive functioning of an organization.18[18] 1. Conformity* (originally referred to as structure) - the feeling that employees have about the constraints in the group, how many rules, regulations, procedures there are; is there an emphasis on “red tape” and going through channels, or is there a loose and informal atmosphere. 2. Responsibility - the feeling of being your own boss; not having to double-check all your decisions; when you have a job to do, knowing that it is your job. 3. Reward - the feeling of being rewarded for a job well done; emphasizing positive rewards rather than punishments; the perceived fairness of the pay and promotion policies. 18[18]

Litwin, G.H., and Stringer, R.A., Jr. Motivation and Organizational Climate. Boston: Harvard University, Graduate School of Business Administration, Division Research, 1968. Page 81-82.

4. 5. 6. 7. 8. 9.

Risk - the sense of riskiness and challenge in the job and in the organization; is there an emphasis on taking calculated risks, or is playing it safe the best way to operate. Warmth - the feeling of general good fellowship that prevails in the work group atmosphere; the emphasis on being well-liked; the prevalence of friendly and informal social groups. Support - the perceived helpfulness of the managers and other employees in the group; emphasis on mutual support from above and below. Standards - the perceived importance of implicit and explicit goals and performance standards; the emphasis on doing a good job; the challenge represented in personal and group goals. Conflict - the feeling that managers and other workers want to hear different opinions; the emphasis placed on getting problems out in the open, rather than smoothing them over or ignoring them. Identity - the feeling that you belong to a company and you are a valuable member of a working team; the importance placed on this kind of spirit.

They then determined that the best way to test the validity of these variables was to see whether and to what extent different organizations differ along these variables. They found, for example, that in the 13 research organizations they examined tended to be high in perceived conformity (20.2) and low in risk taking (9.9). In contrast, the 12 production organizations were lower in perceived conformity (17.6) and higher in risk tendencies (11.1). The following table depicts their early results.19[19]

19[19]

Ibid., pg. 70.

Figure 12: Climate Differences in Four Industries

Climate Scale

Military N=8

Researcha N=13

Sales N=15

Production N=12

Conformity Responsibility Risk Reward Warmth and Support Conflict Expect Approval

23.4 13.1 8.9 21.0 17.0 8.3 25.2

20.2

19.1 16.4 10.5 27.5 21.4 6.9 24.0

17.6

15.4

9.9 26.3 20.3 7.9 21.9

16.4

11.1 26.1 19.0 8.9 24.2

To study the impact of organizational climate on the motive drives of people, they then studied two sales organizations thought to be quite different in organizational climate. The first sales consisted of 19 sales people from a computer sales company; the second organization consisted of 35 sales representatives from an electronics components firm. While both organizations had much in common in terms of the types of people who staffed them, the markets they served and the salaries sales people earned, they differed considerably in terms of the climates that characterized them. The computer time salesmen, for example, were managed in a way that emphasized personal relations while the electronics components sales people were managed in a way that encouraged aggressive selling techniques. Here then, are the differences in climate as measured by the above climate variables:20[20] Figure 13: A Comparison of Climates in Two Companies

Computer Time Sales (N=19)

Climate Scale Conformity 18.7 Responsibility 16.7 Risk 11.4 Reward 28.2 Warmth and Support 21.8 Conflict 7.4 * Differences were statistically significant

Electronics Components Sales (N=35)

Difference

21.0

2.3*

16.5 11.9 27.3

.2 .5 .9

20.4

1.4*

7.7

.3

While many climate characteristics were similar, the two climate scales demonstrably different were Conformity and Warmth and Support. As you might predict, greater room to make choices (lower conformity) combined with greater expression of support and care yielded a higher achievement climate. Figure 14: Mean Motivation Scores for Salesmen in Two Different Sales Organizations21[21] 20[20] 21

Ibid., Page 77.

Computer Time Sales (N=19) n Achievement 10.9 n Power 7.7 n Affiliation 5.3 * Difference was statistically significant

Electronics Components Sales (N=35)

Difference

9.2

1.7

8.3 5.1

.6 .2

Moreover, when asked to rate the success of the salesmen on a four-point scale, it became clear that greater achievement motivation scores yielded higher success. This makes good sense, for the quality and characteristics that describe nAch drives are also the same that are associated with effective goal driven behavior. Figure 15: Correlation of Motivation with Manager Ratings of Success in Two Different Sales Organizations22[22]

n Achievement and Success

Computer Time Sales (N=19)

Electronics Components Sales (N=35)

Difference

+.25

--.23

.48*

+.09

.22

.00

.08

n Power and Success -.13 n Affiliation and Success +.08 * Differences are statistically significant

“the quality and characteristics that describe nAch drives are also the same that are associated with effective goal driven behavior” To test the influence of leadership style on organizational climate, Litwin and Stringer then created three simulated business operations (described as organizations A, B, & C) made up of 15 members, each headed by a president. The presidents were instructed regarding the leadership style they were to maintain so that leadership style became the major variable input. All other factors were controlled as carefully as possible. The physical locations were identical, the technology and essential tasks were the same, and the members of the organizations were matched with respect to age, sex, background,

22[22]

Ibid., Page 79

motive patterns, and personality characteristics. The following description is drawn directly from Litwin and Stringer:23[23] Each simulated business operated in a 100-seat classroom. The work involved the production of miniature construction models of radar towers and radar-controlled guns of various kinds from “Erector Set” parts. A typical product was comprised of from 30 to 50 parts. Innovations were called for, and none of the businesses could afford to stand still. 15 subjects were assigned to each business including 13 men and 2 women. The subjects ranged in age from 18 to 29 years. All were hired (at an average hourly wage of $1.40) to participate in a study of “competitive business organizations.” 45 subjects were selected from an initial subject pool of 78 because they composed the “best” matched groups. The dimensions along which the groups were matched were: age, college major, business or other work experience, n Achievement, n Affiliation, n Power scores, and California Psychological Inventory personality profiles (particularly overall elevation, and elevation in the major areas). Attention was given to careful matching with respect to initial motive score, since aroused motivation was a major output measure. The experiment was conducted over a two-week period, comprising eight actual days of organizational life. The workday averaged about six hours. During the course of the experiment, daily observations were made, and periodic readings were taken using questionnaires and psychological tests. These data were used to provide feedback to the presidents indicating to what extent they were achieving the intended leadership styles. The leader of Organization A was instructed to place strong emphasis on the maintenance of a formal structure. The leader of Organization B was instructed to create a loose, informal structure stressing friendly, cooperative teamwork. The leader of Organization C was instructed to place high emphasis on productivity, encouraging people to set their own goals and take personal responsibility for results. Litwin and Stringer hypothesized that the leadership style of leader A would create a climate that stimulated the need for power while the need for achievement and for affiliation would be reduced. Leader B would create a climate that would arouse the need for affiliation, while power would be suppressed. And finally Leader C would create a climate that would arouse the need for achievement while the need for affiliation and for power would be unaffected. Not surprisingly, over the two week period, each organization developed distinct organizational climates as a result of the different leadership styles. While not all of the initially hypotheses were confirmed, the most salient result of the study was that the leadership style and its resulting climate of Organization C far more than the other two aroused the achievement drives of the people in the organization. At the same time power drives that tend to be counterproductive were 23[23]

Ibid. , pp. 94-96

stimulated by the leadership style of Leaders A and B. Affiliation drives were relatively unaffected. As a result, profitability in organization C was far greater than A and B (explained more fully in the third section of this article). Figure 16: The Effect of Climate on Situationally Aroused Motivation24[24]

n Achievement

n Power

n Affiliation

high

high

high

low

low

low

week 1

week 2

week 1

week 2

week 1

week 2

Organization A Organization B Organization C Tests of social motives administered several weeks later corroborated these findings, strongly suggesting that not only does the leadership style affect the climate and therefore the aroused needs of people, this affect can be long-lasting. After several more studies on the relationship between corporate climate and success, Litwin and Stringer modified their original set of climate variables, and eventually landed on the following six as representing those variables that most impact long term organizational performance. These six will be referred to for the remainder of this article.

The Six Leadership Behavior Styles The following six distinct leadership behavior styles have the greatest impact on climate and thus long term organization performance: Domineering:

Demands immediate compliance, controls tightly, gives corrective negative feedback

Governing:

Gives long-term direction and vision, accepts input, gives balanced feedback

24[24]

Ibid., Page 122.

Placating:

Promotes harmony, considers people’s feelings as strongly as the task, gives inconsistent or exclusively positive feedback unrelated to performance

Involving:

Works for collaborative commitment, empowers others to act, gives feedback for adequate performance rather than clearly differentiating levels of performance

Soloing:

Works to own high standard, ignores or micro-controls depending on assessed quality of employee, gives no feedback, has trouble delegating

Coaching:

Develops employee for long term, empowers to learn and develop, gives feedback on performance for improvement

These styles have as their origins the thought processes of the motives. The “raw style” from nAch is Soloing, from nAff is Placating and from nPow is Domineering. The “learned style” from nAch is Coaching, from nAff is Involving and from nPow is Governing. With few exceptions the “learned” (adapted) style is more effective and appropriate than the raw style. In a recent study, researchers at McBer found a strong relationship between specific leadership styles and the six important organizational climate dimensions, further corroborating and building on Litwin and Stringer’s findings.25[25] For this study they used a measure of managerial style that distinguishes between six differing approaches. An analysis of 3871 managers and executives of major international companies in 9 different industries who rated their immediate supervisor between 1986-1992 found that domineering and soloing styles of management tended to correlate with low achievement climates, while the use of governing, placating, involving and coaching styles tended to correlate with higher achievement climates. In particular, the data showed the following effects on climate variables.

Figure 17: The Effect of Leadership Style on Organizational Climate 25[25]

Kelner, Stephen, P., Rivers, C. A., O’Connell, K. H., “Managerial Style as a Behavioral Predictor of Organizational Climate,” 1994, McBer & Company.

Leadership Behavior Styles

Organizati on Climate Dimensio ns Confor mity Responsibi lity Standards Rewards Clarity Team Spirit

Domineer ing

Governi ng

Pl acating

+

-

+

-

+

+

0 -

+ + + +

+ + + +

In volvin g

-

Soloi ng

Coachi ng

-

-

+

0

0

+ + + +

-

+ + + +

Equally as significant is the fact that differences in managerial style account for between 54% and 72% of variance in climate. In other words, far more than any other factor, one’s management style affects the climate of your immediate work group. This is particularly interesting when one thinks about how often managers say they are powerless to affect their organization and its results. They complain that their manager ties their hands, that the company’s policy inhibits their performance, etc. This data shows, to the contrary, that their ability to influence their organization lies squarely on their shoulders.

Coercive

Low

Authoritative

High

Affiliative

Medium

Democratic

High

Pace-setting

Low

Coaching

High

Figure 8: General Preferred Leadership Style To understand this relationship more fully, let’s look at a couple of actual leaders, each demonstrating different leadership approaches. Bill Toomey is the head of the Real Estate Department of a Fortune 500 electronics company. For a long time he was highly successful as an individual contributor in the company, and was eventually promoted to his position as manager of the department due to his individual skills and enthusiasm. However, because he was so achievement driven, he tended to use a soloing style of management. He would often show his direct reports how to best make real estate deals, and interceded when he thought he could do it better. Moreover, he often criticized them when they made mistakes, even minute ones. Because of his micro-management style, many of his direct reports felt belittled by him. While Bill thought he was coaching, they felt judged and undermined by his over-controlling presence. Interestingly, by his own admission, he “took over” too much, but he could not help himself. He genuinely thought he could do it better, and that by showing them the way, he was helping his direct reports become more effective. They, on the other hand, felt like they had little room to maneuver, and little room to influence the direction of the department or the key decisions. Moreover, when things got tight, Bill would revert to a domineering style of management, thereby exacerbating the feeling of “no freedom to move” and too much micro-managing. When tested with a normed survey, the results showed that they felt they had to conform to Bill’s rigid expectations, that they had little responsibility for the department and its results, they were unclear about their roles and responsibilities, and team spirit was at an all time low.

Figure 18: Early Climate in the Real Estate Department 80 70 60 % of norm

50 40 30 20 10 0 conformity

standards responsibility rewards

clarity team spirit

In a workshop designed to learn about the leadership chain and the impact Bill had on his organization and its results, Bill looked closely at how he was creating such an achievement depressed climate and sought ways to change his leadership style to more positively impact the climate. He quickly decided that, in spite of his own needs to drive toward personal excellence by demanding compliance, in the long run results would improve if he increased: 1.) the degree to which he gave others authority and autonomy to make decisions, 2.) the degree to which he included them in the overall direction of the department and 3.) coached them towards increased performance, rather than do it for them. As a result, in 6 months, the climate dramatically improved. Conformity went down (due to less domineering leadership), standards, rewards, clarity, and team spirit increased significantly. Figure 19: Later Climate in the Real Estate Department 80 70 60 % of norm

50 40 30 20 10 0 conformity standards responsibility rewards

clarity team spirit

While the overall performance results are not yet in, the improved climate of the organization as a direct result of Bill’s shift in leadership has already paid off. One of his employees, for example, just closed a major, multi-million dollar deal; the biggest he had ever done and the most successful. Said the employee, “I felt I was over my head in the deal. I had never done such a deal, and Bill gave me room to do it. I really rose to the occasion.” This would likely not have occurred under Bill’s previous leadership style. In our experience working with thousands of managers and executives, we have seen dramatic results time and time again. The above early studies on leadership and organizational climate keep showing us that not only does ones leadership make a difference, changes in leadership style produce rapid and often significant improvement in its results. The key is knowing specifically where and how to make the changes.

“far more than any other factor, ones management style affects the climate of the organization”

In related studies by other researchers, the relationship between leadership style and organizational climate continues to show the same patterns. For example, in studies by Clayton Lafferty (a student and colleague of McClelland’s), he and his associates have found a direct correlation between leaders who care about their people, who establish challenging, yet achievable goals, and who rarely use coercive means to achieve these results, are consistently seen as more effective leaders than others. Moreover, they create climates that tend to perform more consistently and effectively over long periods of time. Similarly, John Kotter has drawn the same conclusion from his studies of the differences between leaders and managers. Their long term focus, ability to invite and sustain organization change, and their powerful vision distinguishes effective leaders from their less effective counterparts.

Conclusion The second link on the leadership chain is now secured. It has shown us that while different leadership styles work in different situations, there is a particular style that is likely to produce greater goal attainment in most business situations. This style is visionary (governing), involving, and encourages others to learn, grow, and achieve great results through coaching. While variations do exist, few leaders will go wrong in adopting this approach. Perhaps, even more significant is the compelling findings of the more recent McBer study that shows that the leader’s style accounts for over 50% of the variance between climates. In other

words, as a manager, no longer can we blame the organization above us for the quality of the organization below us. We can and do directly impact our organizations, our departments and our teams by the kind of leader we are. If we want more results, and we want these results to sustain, the most powerful lever for change is our leadership behavior.

Part III THE

RELATIONSHIP

ORGANIZATIONAL

BETWEEN

CLIMATE

AND

PERFORMANCE

So far we have explored the relationships between the first three links on the leadership chain. We have seen how the motive profile of a leader impacts his or her leadership style and that his or her style influences the organization’s culture. In this third section, we will explore the relationship between the third and fourth links on the leadership chain: namely the relationship between an organization’s culture and its success. As we have previously indicated, for the purposes of this article we will limit the discussions of Dr. Schneider’s research on culture to the variables that impact motivation. This will focus us on the subset of variables originally labeled “climate” by Litwin and Stringer. An extremely important conclusion of Dr. Schneider is that each of the cultures that is “in balance” (therefore effective and productive) has the same climate which we will refer to as “achievement aroused.” While this makes sense in retrospect, it is a relatively new finding with major implications, particularly for today’s knowledge based companies. While claims abound about the relationship between an organization’s climate and its performance there is surprisingly little published research evidence to support this claim. Fortunately, the evidence that does exist is extremely compelling.

The Impact of Climate on Performance In an early attempt to measure the impact of climate on performance, Litwin and Stringer hypothesized that an achievement aroused climate will likely produce greater profits than an affiliation or power aroused climate. Dr. Schneider would define these as major factors contributing to a culture being “out of balance.” By “achievement aroused climate,” they meant a climate that stimulates the thoughts and behaviors leading to the desire to perform at higher levels of capability. To test this hypothesis they designed and studied a simulated organization populated by University students described in detail in the second of this series of articles. In

their study, the achievement aroused company (organization C) produced far greater profits than A and B, the power and affiliation aroused organizations. Figure 20: The Effects of Climate on Performance in a Simulated Setting26[26]

Contracts Completed Contracts in Process No. of Contracts Worked On No. of New Products Total Revenue Total Costs Charged to Completed Contracts Profit; ( ) = Loss % Profit; ( ) = Loss Materials-Saving Innovation Units Rejected (requiring work)

Company A Power 6 4 10 4 $948.20

Company B Affiliation 6 3 9 6 $661.50

Company C Achievement 8 3 11 8 $619.80

$940.50 $7.70 .81% $0.00 0

$666.80 $(5.30) (.80%) $25.10 1

$547.50 $72.30 11.7% $43.80 4

This finding was profound in its implications for it pointed to a measurable variable that predicted differentiated results. Moreover, at the time, this finding flew in the face of the popular belief that performance often is at the cost of job satisfaction. In this study, job satisfaction was measured as extremely high in C, as well as, the degree of innovation. Interestingly, job satisfaction was also high in B (people felt good), but as an organization it failed miserably in terms of performance.

“this finding flew in the face of the popular belief that performance often is at the cost of job satisfaction”

Climate Patterns Litwin and Stringer went on to further test their hypotheses in a number of field studies, each of which shed light on the powerful influence of climate on the motive patterns of employees and the performance of the organization. In fact, just as with motives, climate variables align to form climate patterns which are useful to predict outcomes. In one study of a service department of a public utility, for example, they found the climate pattern looked as follows:

26[26]

Ibid., Page 139.

Figure 21: Climate Pattern in a Service Department27[27]

90 80 70 % of norm

60 50 40 30 20 10 conformity

standards responsibility rewards

clarity team spirit

As can be seen, these office workers perceived a very high degree of conformity and a relatively high degree of standards (in their words—standards were “nothing short of perfection”) while at the same time little responsibility, reward, and team spirit. While high expectations and challenge can support achievement, without the concomitant support, encouragement and rewards, such an environment can easily suppress achievement drives. This was indeed the case. Employee motive patterns derived by the Thematic Apperception Test (an exercise that measures motive patterns) revealed the following averages: Figure 22: Motive patterns of Service Employees28[28] 90 80 70 60 50 40 30 20 10

n Achievement

27[27] 28[28]

Ibid., p. 149 Ibid., p. 151

n Affiliation

n Power

Not surprisingly, it was reported that service was generally average or below average and sales performance considerably below average. Subsequent attempts to further tighten performance standards were greeted with derision by the employees, and were therefore not successful. In another study of two chemical plants, Litwin and Stringer found the climates of both organizations to be similar in many respects, including conformity, standards, two climate variables no longer measured-- support and conflict, scoring average on each. One of the organizations, however, scored higher in responsibility, reward and morale. People in that organization perceived the climate to be far more favorable than people in the other organization. A summary of subsequent research on the relationship between each of the Litwin and Stringer climate dimensions and organization performance shows the following correlations: Figure 23: The Impact of Climate on Performance Growth in Sales Conformity Responsibility Standards Rewards Clarity Team Spirit Morale (Clarity + Team Spirit) Total Climate Index

Return on Sales

Growth Earnings

-.50 +.34 +.35

+.40

+.40

in

Waste/ Efficiency -.50 +.40 +.57 +.42 +.48 +.59

+.52 +.49

+.56

+.52

The Total Climate Index, in this case, is total of the climate variables when taken as a whole. As can be seen, it correlates strongly with growth on sales, return on sales, and higher efficiency.

“the Total Climate Index …correlates strongly with growth on sales, return on sales, and higher efficiency”

The Hockey Stick From the above research, a model of an ideal climate for high performance comes to sharp focus. While no ideal can be applied to all situations, such a model points the direction for the kind of climate that, in principle, more fully than any other, stimulates achievement motives in employees. Time and again we have tested this model by asking people, both managers and employees, to describe their ideal climate. Time and again, this same profile emerges. We call this “the high performance” climate (see figure 24). Note the resemblance in shape to a hockey stick; which is in fact used as a mnemonic for this achievement-arousing climate pattern. Figure 24: The High Performance Climate 90 80 70 % of norm

60 50 40 30 20 10 conformity

standards responsibility rewards

clarity team spirit

Time and again we have tested this model by asking people to describe their ideal climate. Time and again, this same “hockey stick” profile emerges.

To demonstrate in real life the relationship between climate and performance, let’s look at one organization, whose dramatic change in climate produced a resultant change in performance.

Leadership Style, Climate, and Performance – A Case Example Ardent Insurance, Inc. is a large, well established insurance company, based primarily in the Western part of the United States. For many years it was among the leading insurance companies in the industry. About 8 or 9 years ago, however, it fell onto hard times and was losing money and market share right and left, to the point of being on the brink of extinction. As a result, a new CEO named John Swift, took over the company with a clear mandate to turn the situation around (name of company and CEO changed to protect anonymity). For the first couple of years of John’s tenure, he adopted a style of leadership that was hard nosed and bottom line in nature. His was a “take no prisoners” style that, he believed, would produce immediate results in a situation that required urgency and focus. As it relates to the variables described in article #2 of this series, John’s leadership style, at the time, could be characterized as highly domineering in nature, with a strong dose of soloing. He showed people what he wanted, how he wanted it done, and if it did not get done the way he wanted it, he would get extremely upset, so much so that people feared him. Here was his leadership style profile as perceived by his direct reports (figure 25). Figure 25: John Swift’s Early Leadership Style

100 90 80 70 Percentile compared to average manager

60 50 40

30 20 10 0 Domineering Placating Soloing Governing Involving

Coaching

This leadership style was a natural outgrowth of John’s motive profile, which matched what we call “the entrepreneurial V,” a style that defines and describes many entrepreneurs (figure 26).

Figure 26: John Swift’s Motive Pattern 90 80 70 60 50 40 30 20 10

n Achievement

n Affiliation

n Power

In what is fairly typical of leaders with this profile, his strong need for achievement in a situation requiring immediate results led him to do a lot of soloing, while his strong need to influence combined with a quick trigger finger led to a highly domineering tendency. As a result of his style, as well as other factors including the bureaucratic nature of Ardent Insurance as it existed at the time, the climate of the organization was in significant disarray. To many people, it was considered a company wherein you had to watch your back at all times, and only say what was politically expedient. Not surprisingly, while John thought he was providing strong standards for success, people felt his standards were impossible to achieve. This caused much anger and frustration, feelings that people were not able to express for fear they would be fired. In addition, people had little hope that these conditions would change. Interestingly, contrary to what you would expect, turnover was rather low. This was due, to a large extent to the fact that they were in a recession where there were few jobs available. So while a few of the best people left, the rest stayed, many of whom resented being there. Although many stayed, John felt he needed to take drastic action. He identified people who he wanted to keep, and laid off many of the rest. The company went from 6000 employees to 3200 employees in a two-year period. While the financial bleeding was stopped, the company remained highly unstable in that it was at risk of losing its best employees. In short, while the leadership decisions may have made sense given John’s motive profile and characteristics and the financial crisis, it did little to inspire the workforce to give their best.

Here are the results of a climate survey done for the organization as a whole at the time (figure 27). Figure 27: The Early Climate at Ardent Insurance 90 80 70 % of norm

60 50 40 30 20 10 conformity

standards responsibility rewards

clarity team spirit

When John looked at the company climate and saw how his style contributed to it, he was surprised and troubled by the results and did a number of concrete things to change the situation. He changed his style from domineering to governing in nature. As part of this, he set in motion a series of steps to focus not so much on protecting the company from failure, but on creating a vision for the future, one that was exciting and compelling. He also began delegating much more responsibility to a few trusted direct reports and reduced his soloing tendencies. As a result, the climate in the organization began to shift. While not as comprehensive as the first (we only have data for part of the company), most employees now report the company as clearer in its direction, less conformity driven, and where people have a stronger sense of responsibility. Thus, they experience it as a more rewarding a place to work (figure 28).

Figure 28: Recent Climate at Ardent Insurance

90 80 70 % of norm

60 50 40 30 20 10 conformity

standards responsibility rewards

clarity team spirit

Ardent Insurance is now the most successful insurance company in its industry segment. It is highly profitable, the employee size has grown to 17,000 (with acquisitions), the stock has outperformed its industry by a factor of 16 and the company no longer compares itself to similar companies in the industry. Rather, they are aiming to be among the best companies in the world. This is a far cry from a few years ago when it was on the brink of extinction.

Recent Research Since the Litwin and Stringer study, the relationship between climate and performance has drawn considerable attention. What has emerged by way of research is exciting and highly reinforcing of Litwin and Stringer’s original studies. In a recent study of the relationship between climate and performance, for example, Daniel Denison found a number of climate factors that relate directly to organizational performance. Drawing from a sample of 43,747 people who assessed the climate in 6,671 organizations, groups, and companies, he found that companies which have well organized work systems (related to clarity in Litwin and Stringer’s terms) perform on average far better than those that don’t (see figure 29).29[29]

Figure 29: Organization of Work and Return on Investment30[30] 29[29] 30[30]

Denison, Daniel R., Corporate Culture and Organizational Effectiveness. New York: John Wiley & Sons, 1990. Ibid., p. 62.

20% 18% Return on Investment

high Clarity Organizations

14% 12% 10% 8% 6%

Low Clarity Organizations

4% 2% 0%

1

2

3

4

5

Years

Denison also found the same relationship between organization of work and return on investment. Probing the relationship between climate and performance further, Denison found clear relationships between high involvement decision making practices and return on investment and return on sales (see figure 30). By high involvement, he is referring to decision making practices which are more participatory and collaborative. Interestingly, there appears to be some lag time before the introduction and creation of high involvement decision making climate yields significant results. This is to be expected for it often takes a while for a high involvement decision making practice to take root. Once it does, experience and research shows it often produces more long term results.

Figure 30: High Involvement Workplaces and Return on Investment31[31]

20% High Involvement Organizations

18% Return on Investment

14% 12% 10% 8% 6% 4% Low Involvement Organizations 2% 0%

1

2

3 Years

4

5

Again, there is a similar relationship between high involvement workplaces and return on sales. Some of the other climate variables Denison found to correlate with strong financial results were: ϒΕ ϒΕ ϒΕ ϒΕ

Interunit coordination Strong supervisory leadership Job rewards Job clarity

These latter two are particularly interesting in that they corroborate and reinforce Litwin and Stringer’s original research (rewards and clarity). Perhaps one of the most exciting pieces of research to date linking climate with organizational performance is a study which began in 1993 by the Hay Group who teamed up with the Life Office Management Association to investigate what outstanding life insurance CEOs actually do to succeed.32[32] In their research, they studied 28 CEOs and senior executives--overseeing more than $194 billion in assets--from top performing life insurance companies in North America. Of the 28 CEOs and executives, 19 underwent systematic individual and organizational assessment. In their study, the pool of subjects were split into two groups of performers (outstanding and good) based on their company’s financial performance. The outstanding performers were, on average, in the 75th percentile of their industry in Total Capital and Surplus Growth, while the good performers were in the 25th percentile. The goal of the study was to determine what differentiated outstanding leaders from others. 31[31] 32[32]

Ibid., p. 63. Newsbreak Bulletin, August, 1995, published by Hay McBer Innovation and Resource Center.

Consistent with earlier research and our leadership chain theory, the leadership style of outstanding CEOs had organizational climates that matched our high performance profile. Their total organizational climate scores averaged at the 75th percentile, in contrast with good CEOs whose total climate scores averaged just below the 50th percentile (figure 31). Figure 31: Differences in climate between outstanding and good performers 90

Outstanding CEOs

80 70 % of norm

60 50

Good CEOs

40 30 20 10 conformity

standards responsibility rewards

clarity

total climate team spirit

Note the “hockey stick” pattern evident in the “Outstanding CEO” climate. Interestingly, in addition to creating a high performance climate, the outstanding CEOs also created climates wherein the gap between the actual climate of the organization and the climate that people desired was rather small (22% average gap) while the gap in the less productive organizations was 45%. In short, the relationship between climate and performance is clear and unequivocal: high performance climates clearly yield high financial results (figure 32).

Figure 13: Organization Climate and Financial Success in the Life Insurance Industry

Outstanding CEOs 75%

50% Good CEOs 25%

25%

50%

75%

Financial Success (total capital surplus growth over four years)

What caused such a clear difference in climate and performance? While all leaders drew upon authoritative, democratic, and pace setting styles of leadership, the very best CEOs used the coaching and involving style significantly more than their counterparts. Interestingly, none of the CEOs in the study exhibited a strong domineering style of leadership. Underlying the strong coaching and involving style are a set of competencies that seemed to distinguish outstanding CEOs from good ones. The leadership study found that over 80% of the outstanding CEOs (and only 9% of the rest) were strong in at least 4 of the following 6 competencies: 1. 2. 3. 4. 5. 6.

Socialized Power: finding personal satisfaction from seeing others learn, grow, and improve their performance. Need to Achieve: continuously improving and taking entrepreneurial risks. Change leadership: delivering a message of change; empowering others to change. Curiosity and Information Seeking: probing for information; searching widely for useful ideas and new opportunities. Developing and Empowering: providing long-term encouragement and coaching. Listening and Responding: hearing people’s personal concerns and being aware of their unspoken thoughts and feelings.

“the relationship between climate and performance is clear and unequivocal: high performance climates clearly yield high financial results”

This study adds to the model by including the variable of competencies and reinforces the research that is overwhelming and quite clear. The links on the chain are strong connections, binding leadership competencies to coaching driven leadership style to high performance climate to superior financial results.

THE LEADERSHIP CHAIN SECURED We have now come full circle. We have explored how motive patterns in leaders directly impact their leadership style. We have seen how different styles affect the climate of an organization and which leadership style seems to have the most positive impact in most situations. We have explored how climate impacts performance, and the kind of climate that will stimulate the greatest achievement drives in employees and therefore the greatest results. The leadership chain is now complete. From motive drives to leadership to climate to performance. If any link on the chain is broken, the results are not predictable. Most importantly, since the beginning link starts off the chain reaction, then if that link is weak, performance is inevitably impaired. Most mediocre leaders have difficulty accepting this final proposition. They would rather blame the lack of results of the organization on factors outside their control. The problem is the business environment, or product timing, or poor employees, or other companies and their unfair competitive practices. Strong leaders, however, know this is not the case. All companies face similar challenges. The ones that rise to the top see that the one variable they affect more than any other is the climate in their organization. And they see that for the climate to produce significant and enduring results, it is up to them to learn how to lead the organization to success. Any other focus is doomed to failure.

Addendum to The Leadership Chain

Part IV AN ORGANIZATIONAL MODEL

ALIGNMENT

As described earlier in this article, organizational culture is a composite of climate and many other variables. Climate refers to” a set of measurable variables in the culture that directly affect motivation.” What we have not emphasized is the business strategy. Strategy is the combination of specific forces designed to provide (sustainable) competitive advantage. When individual motives are aligned to the organizational culture which is in turn aligned to business strategy, the building blocks of success are in place. Success becomes most likely when leadership style successfully arouses the motives necessary to meet organizational needs. Analyzing the “connect” – or the disconnect – between climate and business strategy in today’s businesses can provide revealing insights into organizational successes or failures.

While these articles are focused on motivation, no amount of behavioral sciences research or other “magic” can make up for a poor strategy. You must start with strategy in this model and work backwards. Simply put, find out why customers will beat your doors down demanding to buy product from you and then build on that. A great strategy poorly executed often yields better results than a poor strategy brilliantly carried out. Fortunately, one must not be forced between these Hobson’s choices.

The basis for building the organizational alignment model comes from connecting several pieces of recent organizational research and applying them to contemporary organizations. Dr. William Schneider’s33[33] construct of organization effectiveness actually starts with David McClelland’s work on motives. His research suggests that effective organizations are in alignment; i.e.: the motive lines up with the appropriate culture (by definition therefore the climate variables are also in alignment) and the cultures are lined up with the appropriate strategy. It all becomes clearer in the following diagram: What does this all mean? At the risk of oversimplifying, what this model indicates is that companies like Dell, Amazon, etc. are successful because their organizational strategies are right and their organizational strategies, their core cultures, and internal motive patterns are aligned. Dell succeeds because its core philosophy is “operational excellence.” That strategy is best achieved through a culture and leadership style of “control,” which is in turn best achieved through a governing style. Control costs, gain market share, control the market – that’s what Dell is all about, and they have consciously and unconsciously created a culture to achieve those ends. On the other hand, companies like Cisco get and keep customers through the strategy (value proposition) of “product leadership.” This is best supported by the culture labeled “competency,” which is most successful when the achievement motive is aroused. Hence, that becomes the leadership style. And finally – though the affiliative, collaborative style seems less effective in most situations, it can be very effective when the customer relationship -- customer intimacy -- is the strategy that drives competitive advantage. “Alignment” occurs when the chosen strategy is backed up by the appropriate culture and the executives have the appropriate motives to unconsciously drive the style to support the culture. Thus if Wal-Mart chooses a cost leadership (operational excellence) strategy, in order to be in alignment the best culture for them is “control” and the best style “governing” which is automatic pilot from socialized power. This is rowing with the current. If the strategy were operational excellence but the culture were collaborative and the styles of the executives “Soloing,” it would be like rowing against a strong current and soon the organization would wear down with relatively less progress. Not only is this a product of research, but it is common sense -- which is an added bonus! Note that no amount of “organizational alignment” will compensate for a poorly conceived strategy. If the strategy is wrong in the first place, it doesn’t help to lead in that direction. But the right strategy with inappropriate leadership approaches will invariably lead to underperforming systems.

33[33]

Schneider, W.E., The Reengineering Alternative, A Plan for Making Your Current Culture Work, McGraw Hill, 1994

Part V REVIEW AND CONCLUSIONS

SUMMARY

REVIEW AND SUMMARY CONCLUSIONS ϒΕ

A motive is a recurrent concern for a goal state (measured in fantasy) which drives, directs, or selects a person’s behavior.

ϒΕ

The three “Social Motives” are achievement (nAch), affiliation (nAff) and power (nPow).

ϒΕ

People can (and do) possess more than one motive, but usually one is dominant.

ϒΕ

The leadership chain connects motives, leadership style, organizational climate and performance.

ϒΕ

While other factors certainly impact an organization’s climate, the most potent impact is the behavioral style of leadership of the immediate manager.

ϒΕ

Your motive pattern combined with your ability to express those motives maturely and effectively has a direct impact on your leadership style.

ϒΕ

Your leadership style directly impacts the organization climate you create, which in turn affects the degree to which people desire to achieve and perform in your organization.

ϒΕ

This desire translates into “discretionary effort” and collaboration.

ϒΕ

Discretionary effort and collaboration improve results.

APPENDIX: A Brief History of McClelland (1917—1998) And His Research McClelland began his research just after World War II when he discovered the work of Henry Murray. At the time, Murray was one of the foremost experts in the field of psychology and the study of human motives. The study of motives began when scientists sought to understand why people behave the way they do. If we can understand the “why” behind behavior, then we can predict and even impact people’s behavior. Dr. Henry Murray understood the relationship between motives and patterned behavior well and sought to measure these underlying drives or motives. Murray believed the underlying pattern of thinking or motives would be best discovered by analyzing people’s responses to particular situations and events. To help discover patterns in people’s responses, he developed what is known as the Thematic Apperception Test, or TAT for short. The TAT is a series of illustrations designed to measure motives as revealed by a person’s spontaneous responses to the picture. The person is shown these pictures and asked to tell stories about them. The stories, when analyzed by experts using the exact same metrics, reveal the underlying concerns or thought patterns of the story teller. Based on this premise and subsequent studies, Murray discovered and defined numerous measurable motive patterns. They included such motives as the need for dominance, the need to avoid harm, the need for achievement, as well as many others. (See table below)

Examples of Murray’s Original Motives Abasement Autonomy Dominance Humiliation Play

Achievement Counteraction Exhibition Avoidance Rejection

Affiliation Defendance Harm Avoidance Nurturance Sentience

Aggression Deference Order Sex

As a result of these and subsequent efforts, there is an extremely large data base of information gathered and thoroughly analyzed regarding motives both clinical and social. Early in his career in the 1950s, McClelland was fascinated by the ability of the TAT to ascertain people’s underlying motives and chose to focus on the three motives that research has shown to be prevalent in social activity (i.e.—those motives that would help us understand why people behave the way they do while working with others or engaged in social interaction).

They were: the need for achievement, for affiliation, and for power. In shorthand, these are referred to as nAch, nAff, and nPow, the “n” referring to a “need.” This particular focus occurred just before his appointment as the Chairman of the School of Social Psychology at Harvard University. This prestigious platform afforded him the opportunity to continue his research interests with ample financing and manpower in the form of doctoral students seeking to add to his growing knowledge base. During his time at Harvard, McClelland modified the TAT to best bring out the three primary social motives. One TAT picture, for example, depicts a man sitting at a desk with a picture of his family in front of him. He appears to be peering at the picture in a thoughtful manner. Another depicts two people engaged in work in what appears to be a chemical laboratory or a medical laboratory. The set of six TAT pictures tend to invoke responses that reveal a person’s social motives and the relative strength of these motives compared with one another. In one of the first applications of his focus of study, McClelland demonstrated that a high achievement motivation in students (high n Ach), as measured by the TAT, corresponded with a higher tendency to perform certain tasks and puzzles quickly and successfully. This as well as similar studies culminated in his first book—The Achievement Motive.34[34] Tackling an even more ambitious subject, by studying the primary messages in children’s stories of different countries, he and his associates were able to demonstrate that the amount of achievement-related messages in these stories directly correlated with the average annual per capita consumption of coal usage, which is a direct measure of economic performance. Published in his seminal book—The Achieving Society35[35]— he elaborated on this research and successfully accounted for the rise and fall of entire economies by identifying fluctuations in achievement motives. In effect, economies grew when their achievement levels were high, and they declined when achievement levels were low. In the 1960s, challenging some fundamental beliefs of his contemporary psychologists that motives are fixed and firm in people, and therefore cannot be altered, McClelland became interested in finding ways to arouse the achievement motives. To accomplish this, he and his associates developed and offered a training program to 16 executives of AT&T. The program lasted a week and was designed to teach the executives about the achievement motive, the research underlying it, and how to think, act and perceive the world as someone with high achievement needs. A follow up study two years later showed conclusively that those who attended the course had been promoted faster than the average executive in the company, thereby reinforcing the value of such a training.

The Distinction between Needs and Values 34[34]

McClelland, D. C., Atkinson, J. W., Clark, R. A., and Lowell, E. L. The Achievement Motive. Appleton-CenturyCrofts, New York, 1953. 35[35] McClelland, D. C., The Achieving Society, Van Nostrand, New York, 1961.

Later research by McClelland and others has distinguished between “needs” and “values.” Needs are underlying patterns that are not conscious in nature. We all have them, and our behavior is driven to meet them. We also have values. Values are principles of thought and behavior we deem important. Values are more often than not conscious and learned through social reinforcement. Like needs, they also affect our behavior, but they are usually conscious. Since they are often learned, they can be unlearned, and therefore are much more changeable than motives or needs.

A remarkable finding. Based on this initial success, McClelland and his associates then offered the training course in a number of small cities in India where the cultural habits of people seemed to suppress achievement needs. In one particular controlled study, McClelland compared two economically and socially similar cities in India: Kinkinada and Rajamundi using the latter as a control group and the former as a test group. In the test group, a 10 day achievement need stimulation program was able to raise the achievement needs in 52 participants of the training. The training helped participants feel strong, competent, and effective. As a result, 2/3 of the participants of the training became unusually active in business, much more than they had before. Examples of high business activity were: starting a new business, expanding an old business, greatly increasing profits, or taking active steps to generate new products. Prior to the training, only 1/3 of the participants were active in business. Overall, the results of the training stimulated significant economic activity in Kinkinada. This was in contrast to the control city which had no appreciable increase in economic activity. Prior to these findings, the economic growth in both cities had been the same for 3200 years. In another test program in Bombay, as a result of the same 10-day training, participants demonstrated an unusually high amount of entrepreneurial activity, resulting in new business starts and efforts that earned significant salary increases. These and other studies have strongly confirmed two of McClelland’s hypothesis—that the achievement need can indeed be stimulated and that raises in achievement drives clearly results in measurable increases in economic growth for individuals, groups, and organizations. McClelland passed away in March of 1998, but his numerous students continue their research into motives and their impact on behavior. Together, McClelland, his associates and students began a legacy of learning that continues to this day. More importantly, as his research evolved, his interests started to move toward the subject of leadership and organizational performance. Going against society’s often unstated discomfort with power, he learned that power and leadership are inextricably linked and that the need for power is a far greater predictor of leadership effectiveness than any other singular

motive. The results of his more recent interests in leadership and power have proven one more contribution of this giant and stimulating thought leader.

Related Documents

The Chain
October 2019 34
The Leadership
October 2019 19
The New Food Chain
May 2020 12

More Documents from "Lina Krishnan"