In Circle Y What is Mtu
During the past 30 years, managers have been bombarded with two competing approaches to the problems of human administration and organization. The first, usually called the classical school of organization, emphasizes the need for well-established lines of authority, clearly defined jobs, and authority equal to responsibility. The second, often called the participative approach, focuses on the desirability of involving organization members in decision making so that they will be more highly motivated.
Douglas McGregor, through his well-known “Theory X and Theory Y,” drew a distinction between the assumptions about human motivation which underlie these two approaches, to this effect:
- Theory X assumes that people dislike work and must be coerced, controlled, and directed toward organizational goals. Furthermore, most people prefer to be treated this way, so they can avoid responsibility.
- Theory Y—the integration of goals—emphasizes the average person’s intrinsic interest in his work, his desire to be self-directing and to seek responsibility, and his capacity to be creative in solving business problems.
It is McGregor’s conclusion, of course, that the latter approach to organization is the more desirable one for managers to follow.1
McGregor’s position causes confusion for the managers who try to choose between these two conflicting approaches. The classical organizational approach that McGregor associated with Theory X does work well in some situations, although, as McGregor himself pointed out, there are also some situations where it does not work effectively. At the same time, the approach based on Theory Y, while it has produced good results in some situations, does not always do so. That is, each approach is effective in some cases but not in others. Why is this? How can managers resolve the confusion?
A New Approach
Recent work by a number of students of management and organization may help to answer such questions.2
These studies indicate that there is not one best organizational approach; rather, the best approach depends on the nature of the work to be done. Enterprises with highly predictable tasks perform better with organizations characterized by the highly formalized procedures and management hierarchies of the classical approach. With highly uncertain tasks that require more extensive problem solving, on the other hand, organizations that are less formalized and emphasize self-control and member participation in decision making are more effective. In essence, according to these newer studies, managers must design and develop organizations so that the organizational characteristics
the nature of the task to be done.
While the conclusions of this newer approach will make sense to most experienced managers and can alleviate much of the confusion about which approach to choose, there are still two important questions unanswered:
1. How does the more formalized and controlling organization affect the motivation of organization members? (McGregor’s most telling criticism of the classical approach was that it did not unleash the potential in an enterprise’s human resources.)
2. Equally important, does a less formalized organization always provide a high level of motivation for its members? (This is the implication many managers have drawn from McGregor’s work.)
We have recently been involved in a study that provides surprising answers to these questions and, when taken together with other recent work, suggests a new set of basic assumptions which move beyond Theory Y into what we call “Contingency Theory: the fit between task, organization, and people.” These theoretical assumptions emphasize that the appropriate pattern of organization is
on the nature of the work to be done and on the particular needs of the people involved. We should emphasize that we have labeled these assumptions as a step beyond Theory Y because of McGregor’s own recognition that the Theory Y assumptions would probably be supplanted by new knowledge within a short time.3
The Study Design
Our study was conducted in four organizational units. Two of these performed the relatively certain task of manufacturing standardized containers on high-speed, automated production lines. The other two performed the relatively uncertain work of research and development in communications technology. Each pair of units performing the same kind of task were in the same large company, and each pair had previously been evaluated by that company’s management as containing one highly effective unit and a less effective one. The study design is summarized in Exhibit I.
Exhibit I. Study Design in “Fit” of Organizational Characteristics
The objective was to explore more fully how the fit between organization and task was related to successful performance. That is, does a good fit between organizational characteristics and task requirements increase the motivation of individuals and hence produce more effective individual and organizational performance?
An especially useful approach to answering this question is to recognize that an individual has a strong need to master the world around him, including the task that he faces as a member of a work organization.4
The accumulated feelings of satisfaction that come from successfully mastering one’s environment can be called a “sense of competence.” We saw this sense of competence in performing a particular task as helpful in understanding how a fit between task and organizational characteristics could motivate people toward successful performance.
Because the four study sites had already been evaluated by the respective corporate managers as high and low performers of tasks, we expected that such differences in performance would be a preliminary clue to differences in the “fit” of the organizational characteristics to the job to be done. But, first, we had to define what kinds of organizational characteristics would determine how appropriate the organization was to the particular task.
We grouped these organizational characteristics into two sets of factors:
1. Formal characteristics, which could be used to judge the fit between the kind of task being worked on and the formal practices of the organization.
2. Climate characteristics, or the subjective perceptions and orientations that had developed among the individuals about their organizational setting. (These too must fit the task to be performed if the organization is to be effective.)
We measured these attributes through questionnaires and interviews with about 40 managers in each unit to determine the appropriateness of the organization to the kind of task being performed. We also measured the feelings of competence of the people in the organizations so that we could link the appropriateness of the organizational attributes with a sense of competence.
The principal findings of the survey are best highlighted by contrasting the highly successful Akron plant and the high-performing Stockton laboratory. Because each performed very different tasks (the former a relatively certain manufacturing task and the latter a relatively uncertain research task), we expected, as brought out earlier, that there would have to be major differences between them in organizational characteristics if they were to perform effectively. And this is what we did find. But we also found that each of these effective units had a better fit with its particular task than did its less effective counterpart.
While our major purpose in this article is to explore how the fit between task and organizational characteristics is related to motivation, we first want to explore more fully the organizational characteristics of these units, so the reader will better understand what we mean by a fit between task and organization and how it can lead to more effective behavior. To do this, we shall place the major emphasis on the contrast between the high-performing units (the Akron plant and Stockton laboratory), but we shall also compare each of these with its less effective mate (the Hartford plant and Carmel laboratory respectively).
Beginning with differences in formal characteristics, we found that both the Akron and Stockton organizations fit their respective tasks much better than did their less successful counterparts. In the predictable manufacturing task environment, Akron had a pattern of formal relationships and duties that was highly structured and precisely defined. Stockton, with its unpredictable research task, had a low degree of structure and much less precision of definition (see Exhibit II).
Exhibit II. Differences in Formal Characteristics in High-performing Organizations
Akron’s pattern of formal rules, procedures, and control systems was so specific and comprehensive that it prompted one manager to remark:
“We’ve got rules here for everything from how much powder to use in cleaning the toilet bowls to how to cart a dead body out of the plant.”
In contrast, Stockton’s formal rules were so minimal, loose, and flexible that one scientist, when asked whether he felt the rules ought to be tightened, said:
“If a man puts a nut on a screw all day long, you may need more rules and a job definition for him. But we’re not novices here. We’re professionals and not the kind who need close supervision. People around here
produce, and produce under relaxed conditions. Why tamper with success?”
These differences in formal organizational characteristics were well suited to the differences in tasks of the two organizations. Thus:
- Akron’s highly structured formal practices fit its predictable task because behavior had to be rigidly defined and controlled around the automated, high-speed production line. There was really only one way to accomplish the plant’s very routine and programmable job; managers defined it precisely and insisted (through the plant’s formal practices) that each man do what was expected of him.
On the other hand, Stockton’s highly unstructured formal practices made just as much sense because the required activities in the laboratory simply could not be rigidly defined in advance. With such an unpredictable, fast-changing task as communications technology research, there were numerous approaches to getting the job done well. As a consequence, Stockton managers used a less structured pattern of formal practices that left the scientists in the lab free to respond to the changing task situation.
- Akron’s formal practices were very much geared to
concerns as its task demanded. For example, formal production reports and operating review sessions were daily occurrences, consistent with the fact that the through-put time for their products was typically only a few hours.
By contrast, Stockton’s formal practices were geared to
concerns, as its task demanded. Formal reports and reviews were made only quarterly, reflecting the fact that research often does not come to fruition for three to five years.
At the two less effective sites (i.e., the Hartford plant and the Carmel laboratory), the formal organizational characteristics did not fit their respective tasks nearly as well. For example, Hartford’s formal practices were much less structured and controlling than were Akron’s, while Carmel’s were more restraining and restricting than were Stockton’s. A scientist in Carmel commented:
“There’s something here that keeps you from being scientific. It’s hard to put your finger on, but I guess I’d call it ‘Mickey Mouse.’ There are rules and things here that get in your way regarding doing your job as a researcher.”
As with formal practices, the climate in both high-performing Akron and Stockton suited the respective tasks much better than did the climates at the less successful Hartford and Carmel sites.
Perception of structure:
The people in the Akron plant perceived a great deal of structure, with their behavior tightly controlled and defined. One manager in the plant said:
“We can’t let the lines run unattended. We lose money whenever they do. So we make sure each man knows his job, knows when he can take a break, knows how to handle a change in shifts, etc. It’s all spelled out clearly for him the day he comes to work here.”
In contrast, the scientists in the Stockton laboratory perceived very little structure, with their behavior only minimally controlled. Such perceptions encouraged the individualistic and creative behavior that the uncertain, rapidly changing research task needed. Scientists in the less successful Carmel laboratory perceived much more structure in their organization and voiced the feeling that this was “getting in their way” and making it difficult to do effective research.
Distribution of influence:
The Akron plant and the Stockton laboratory also differed substantially in how influence was distributed and on the character of superior-subordinate and colleague relations. Akron personnel felt that they had much less influence over decisions in their plant than Stockton’s scientists did in their laboratory. The task at Akron had already been clearly defined and that definition had, in a sense, been incorporated into the automated production flow itself. Therefore, there was less need for individuals to have a say in decisions concerning the work process.
Moreover, in Akron, influence was perceived to be concentrated in the upper levels of the formal structure (a hierarchical or “top-heavy” distribution), while in Stockton influence was perceived to be more evenly spread out among more levels of the formal structure (an egalitarian distribution).
Akron’s members perceived themselves to have a low degree of freedom vis-à-vis superiors both in choosing the jobs they work on and in handling these jobs on their own. They also described the type of supervision in the plant as being relatively directive. Stockton’s scientists, on the other hand, felt that they had a great deal of freedom vis-à-vis their superiors both in choosing the tasks and projects, and in handling them in the way that they wanted to. They described supervision in the laboratory as being very participatory.
It is interesting to note that the less successful Carmel laboratory had more of its decisions made at the top. Because of this, there was a definite feeling by the scientists that their particular expertise was not being effectively used in choosing projects.
Relations with others:
The people at Akron perceived a great deal of similarity among themselves in background, prior work experiences, and approaches for tackling job-related problems. They also perceived the degree of coordination of effort among colleagues to be very high. Because Akron’s task was so precisely defined and the behavior of its members so rigidly controlled around the automated lines, it is easy to see that this pattern also made sense.
By contrast, Stockton’s scientists perceived not only a great many differences among themselves, especially in education and background, but also that the coordination of effort among colleagues was relatively low. This was appropriate for a laboratory in which a great variety of disciplines and skills were present and individual projects were important to solve technological problems.
As we would expect, Akron’s individuals were highly oriented toward a relatively short time span and manufacturing goals. They responded to quick feedback concerning the quality and service that the plant was providing. This was essential, given the nature of their task.
Stockton’s researchers were highly oriented toward a longer time span and scientific goals. These orientations meant that they were willing to wait for long-term feedback from a research project that might take years to complete. A scientist in Stockton said:
“We’re not the kind of people here who need a pat on the back every day. We can wait for months if necessary before we get feedback from colleagues and the profession. I’ve been working on one project now for three months and I’m still not sure where it’s going to take me. I can live with that, though.”
This is precisely the kind of behavior and attitude that spells success on this kind of task.
Finally, the individuals in both Akron and Stockton perceived their chief executive to have a “managerial style” that expressed more of a concern for the task than for people or relationships, but this seemed to fit both tasks.
In Akron, the technology of the task was so dominant that top managerial behavior which was not focused primarily on the task might have reduced the effectiveness of performance. On the other hand, although Stockton’s research task called for more individualistic problem-solving behavior, that sort of behavior could have become segmented and uncoordinated, unless the top executive in the lab focused the group’s attention on the overall research task. Given the individualistic bent of the scientists, this was an important force in achieving unity of effort.
All these differences in climate characteristics in the two high performers are summarized in Exhibit III.
Exhibit III. Differences in “Climate” Characteristics in High-performing Organizations
As with formal attributes, the less effective Hartford and Carmel sites had organization climates that showed a perceptibly lower degree of fit with their respective tasks. For example, the Hartford plant had an egalitarian distribution of influence, perceptions of a low degree of structure, and a more participatory type of supervision. The Carmel laboratory had a somewhat top-heavy distribution of influence, perceptions of high structure, and a more directive type of supervision.
Because of the difference in organizational characteristics at Akron and Stockton, the two sites were strikingly different places in which to work. But these organizations had two very important things in common. First, each organization fit very well the requirements of its task. Second, although the behavior in the two organizations was different, the result in both cases was effective task performance.
Since, as we indicated earlier, our primary concern in this study was to link the fit between organization and task with individual motivation to perform effectively, we devised a two-part test to measure the sense of competence motivation of the individuals at both sites. Thus:
part asked a participant to write creative and imaginative stories in response to six ambiguous pictures.
asked him to write a creative and imaginative story about what he would be doing, thinking, and feeling “tomorrow” on his job. This is called a “projective” test because it is assumed that the respondent projects into his stories his own attitudes, thoughts, feelings, needs, and wants, all of which can be measured from the stories.5
The results indicated that the individuals in Akron and Stockton showed significantly more feelings of competence than did their counterparts in the lower-fit Hartford and Carmel organizations.6
We found that the organization-task fit is simultaneously linked to and interdependent with both individual motivation and effective unit performance. (This interdependency is illustrated in Exhibit IV.)
Exhibit IV. Basic Contingent Relationships
Putting the conclusions in this form raises the question of cause and effect. Does effective unit performance result from the task-organization fit or from higher motivation, or perhaps from both? Does higher sense of competence motivation result from effective unit performance or from fit?
Our answer to these questions is that we do not think there are any single cause-and-effect relationships, but that these factors are mutually interrelated. This has important implications for management theory and practice.
Returning to McGregor’s Theory X and Theory Y assumptions, we can now question the validity of some of his conclusions. While Theory Y might help to explain the findings in the two laboratories, we clearly need something other than Theory X or Y assumptions to explain the findings in the plants.
For example, the managers at Akron worked in a formalized organization setting with relatively little participation in decision making, and yet they were highly motivated. According to Theory X, people would work hard in such a setting only because they were coerced to do so. According to Theory Y, they should have been involved in decision making and been self-directed to feel so motivated. Nothing in our data indicates that either set of assumptions was valid at Akron.
Conversely, the managers at Hartford, the low-performing plant, were in a less formalized organization with more participation in decision making, and yet they were not as highly motivated like the Akron managers. The Theory Y assumptions would suggest that they should have been more motivated.
A way out of such paradoxes is to state a new set of assumptions, the Contingency Theory, that seems to explain the findings at all four sites:
1. Human beings bring varying patterns of needs and motives into the work organization, but one central need is to achieve a sense of competence.
2. The sense of competence motive, while it exists in all human beings, may be fulfilled in different ways by different people depending on how this need interacts with the strengths of the individuals’ other needs—such as those for power, independence, structure, achievement, and affiliation.
3. Competence motivation is most likely to be fulfilled when there is a fit between task and organization.
4. Sense of competence continues to motivate even when a competence goal is achieved; once one goal is reached, a new, higher one is set.
While the central thrust of these points is clear from the preceding discussion of the study, some elaboration can be made. First, the idea that different people have different needs is well understood by psychologists. However, all too often, managers assume that all people have similar needs. Lest we be accused of the same error, we are saying only that all people have a need to feel competent; in this
way they are similar. But in many other dimensions of personality, individuals differ, and these differences will determine how a particular person achieves a sense of competence.
Thus, for example, the people in the Akron plant seemed to be very different from those in the Stockton laboratory in their underlying attitudes toward uncertainty, authority, and relationships with their peers. And because they had different need patterns along these dimensions, both groups were highly motivated by achieving competence from quite different activities and settings.
While there is a need to further investigate how people who work in different settings differ in their psychological makeup, one important implication of the Contingency Theory is that we must not only seek a fit between organization and task, but also between task and people and between people and organization.
A further point which requires elaboration is that one’s sense of competence never really comes to rest. Rather, the real satisfaction of this need is in the successful performance itself, with no diminishing of the motivation as one goal is reached. Since feelings of competence are thus reinforced by successful performance, they can be a more consistent and reliable motivator than salary and benefits.
Implications for managers
The major managerial implication of the Contingency Theory seems to rest in the task-organization-people fit. Although this interrelationship is complex, the best possibility for managerial action probably is in tailoring the organization to fit the task and the people. If such a fit is achieved, both effective unit performance and a higher sense of competence motivation seem to result.
Managers can start this process by considering how certain the task is, how frequently feedback about task performance is available, and what goals are implicit in the task. The answers to these questions will guide their decisions about the design of the management hierarchy, the specificity of job assignments, and the utilization of rewards and control procedures. Selective use of training programs and a general emphasis on appropriate management styles will move them toward a task-organization fit.
The problem of achieving a fit among task, organization, and people is something we know less about. As we have already suggested, we need further investigation of what personality characteristics fit various tasks and organizations. Even with our limited knowledge, however, there are indications that people will gradually gravitate into organizations that fit their particular personalities. Managers can help this process by becoming more aware of what psychological needs seem to best fit the tasks available and the organizational setting, and by trying to shape personnel selection criteria to take account of these needs.
In arguing for an approach which emphasizes the fit among task, organization, and people, we are putting to rest the question of which organizational approach—the classical or the participative—is best. In its place we are raising a new question: What organizational approach is most appropriate given the task and the people involved?
For many enterprises, given the new needs of younger employees for more autonomy, and the rapid rates of social and technological change, it may well be that the more participative approach is the most appropriate. But there will still be many situations in which the more controlled and formalized organization is desirable. Such an organization need not be coercive or punitive. If it makes sense to the individuals involved, given their needs and their jobs, they will find it rewarding and motivating.
The reader will recognize that the complexity we have described is not of our own making. The basic deficiency with earlier approaches is that they did not recognize the variability in tasks and people which produces this complexity. The strength of the contingency approach we have outlined is that it begins to provide a way of thinking about this complexity, rather than ignoring it. While our knowledge in this area is still growing, we are certain that any adequate theory of motivation and organization will have to take account of the contingent relationship between task, organization, and people.
1. Douglas McGregor,
The Human Side of Enterprise
(New York, McGraw-Hill Book Company, Inc., 1960), pp. 34–35 and pp. 47–48.
2. See for example Paul R. Lawrence and Jay W. Lorsch,
Organization and Environment
(Boston, Harvard Business School, Division of Research, 1967); Joan Woodward,
Industrial Organization: Theory & Practice
(New York, Oxford University Press, Inc., 1965); Tom Burns and G.M. Stalker,
The Management of Innovation
(London, Tavistock Publications, 1961); Harold J. Leavitt, “Unhuman Organizations,” HBR July–August 1962, p. 90.
3. McGregor, op. cit., p. 245.
4. See Robert W. White, “Ego and Reality in Psychoanalytic Theory,”
Vol. III, No. 3 (New York, International Universities Press, 1963).
5. For a more detailed description of this survey, see John J. Morse,
Internal Organizational Patterning and Sense of Competence Motivation
(Boston, Harvard Business School, unpublished doctoral dissertation, 1969).
6. Differences between the two container plants are significant at .001 and between the research laboratories at .01 (one-tailed probability).
A version of this article appeared in the May 1970 issue of
Harvard Business Review.
In Circle Y What is Mtu