A clear configuration of the design parameters has held up consistently in the research: highly specialized, routine operating tasks; very formalized procedures in the operating core; a proliferation of rules, regulations, and formalized communication throughout the organization; large-sized units at the operating level; reliance on the functional basis for grouping tasks; relatively centralized power for decision making; and an elaborate admin- istrative structure with a sharp distinction between line and staff.
1. The operating core
The obvious starting point is the operating core, with its highly ra- tionalized work flow. As a result of this, the operating tasks are simple and repetitive, generally requiring a minimum of skill and little training—often taking only hours, seldom more than a few weeks, and usually in-house. This leads to a sharp division of labor in the operating core—to narrowly defined jobs, specialized both vertically and horizontally—and to an em- phasis on the standardization of work processes for coordination. Thus, formalization of behavior emerges as the key design parameter. Because the workers are left with little discretion in their work, there is little pos- sibility for mutual adjustment in the operating core. The use of direct supervision by first-line managers is limited by the fact that standardiza- tion handles most of the coordination. Thus, very large units can be de- signed in the operating core. (There is, however, as we shall see below, need for another kind of direct supervision.)
2. The administrative component
The tight regulation of the operating work—in effect, the sealing off of the operating core from disruptive environmental influence—requires that the administrative structure be highly elaborated. First is the middle line, which is fully developed, especially well above the operating core, and is sharply differentiated into functional units. The managers of this middle line have three prime tasks. One is to handle the disturbances that arise among the highly specialized workers of the operating core. Although standardization takes care of most of the operating interdependences, am- biguities inevitably remain, and these give rise to conflicts. These cannot easily be handled by mutual adjustment among the operators, since infor- mal communication is inhibited by the extensive standardization. So they tend to be handled by direct supervision, the orders of first-line managers. And because many of these conflicts arise between operators adjacent to each other in the work flow, the natural tendency is to bring adjacent operators under common supervision—in other words, to group the oper- ators into units that deal with distinct parts of the work flow, which results in the functional basis for grouping operating units. For the same reason, this functional grouping gets mirrored all the way up the hierarchy, from the production and maintenance departments, which look to the plant manager to resolve many of their conflicts, to the manufacturing and mar- keting vice-presidents, who often expect the same of the company president.
A second task of the middle-line managers, which also explains why they are grouped on functional bases, is to work in a liaison role with the analysts of the technostructure to incorporate their standards down into the operating units. Their third task is to support the vertical flows in the structure—the aggregation of the feedback information up the hierarchy and the elaboration of the action plans that come back down. All these tasks of the middle-line managers require personal contacts—with their subordinates, the analysts, and their own superiors—which limit the num- ber of people they can supervise. Hence, units above the operating core tend to be rather small in size and the overall administrative hierarchy rather tall in shape.
The technostructure must also be highly elaborated. In fact, Stinch- combe identified the birth of this structure in early nineteenth-century industries such as textiles and banking with the growth of technocratic personnel. Because the Machine Bureaucracy depends primarily on the standardization of its operating work processes for coordination, the tech- nostructure—which houses the analysts who do the standardizing— emerges as the key part of the structure. This is so despite the fact that the Machine Bureaucracy sharply distinguishes between line and staff. To the line managers is delegated the formal authority for the operating units; the technocratic staff—officially, at least—merely advises. But without the standardizers—the cadre of work-study analysts, job-description design- ers, schedulers, quality control engineers, planners, budgeters, MIS peo- ple, accountants, operations researchers, and many, many more—the structure simply could not function. Hence, despite their lack of formal authority, considerable informal power rests with the analysts of the tech- nostructure—those who standardize everyone else’s work.
The informal power of the technostructure is gained largely at the expense of the operators, whose work the analysts formalize to a high degree, and of the first-line managers, who would otherwise supervise the operators directly. Such formalization institutionalizes the work of these managers, removing much of their power to coordinate and putting it into the systems designed by the analysts. The first-line manager’s job can, in fact, become so circumscribed that he can hardly be said to function as a manager at all (that is, as someone who is in charge of an organizational unit). The classic case is the foreman on the assembly line, although earlier we had the example of the branch managers of the large Canadian banks, and Jay (1970:66) describes the same phenomenon in his job as head of a program production department in the BBC television service.
The emphasis on standardization extends well beyond the operating core of the Machine Bureaucracy, and with it follows the analysts’ influ- ence. In other words, rules and regulations permeate the entire Machine Bureaucracy structure; formal communication is favored at all levels; deci- sion making tends to follow the formal chain of authority. In no other configuration does the flow of information and decision making more closely resemble the system of regulated flows presented in our second overlay of Chapter 1, with commands amplified down the vertical chain and feedback information aggregated up it. (This is not to suggest that the work of the senior managers is rigid and formalized, but rather that at every hierarchical level, behavior in the Machine Bureaucracy is relatively more formalized than that in the other configurations.)
A further reflection of this formalization is the sharp divisions of labor all over the Machine Bureaucracy. We have already discussed job special- ization in the operating core and the sharp division between line and staff. In addition, the administrative structure is sharply differentiated from the operating core. Unlike the case with the Simple Structure, here managers seldom work alongside operators. And the division of labor between the analysts who design the work and the operators who do it is equally sharp. In general, of the five configurations, it is the Machine Bureaucracy that most strongly emphasizes division of labor and unit differentiation, in all their forms—vertical, horizontal, line/staff, functional, hierarchical, and status.
In general, then, the Machine Bureaucracy functions most clearly in accord with the classical principles of management: formal authority filters down a clearly defined hierarchy, throughout which the principle of unity of command is carefully maintained, as is the rigid distinction between line and staff. Thus, the real error of the classical theorists was not in their principles per se, but in their claim that these were universal; in fact, they apply only to this and one other of the five configurations.1
3. The obsession with control
All this suggests that the Machine Bureaucracy is a structure with an obsession—namely, control. A control mentality pervades it from top to bottom. Three quotations illustrate this, each from a different hierarchical level. First, near the bottom, consider how a Ford Assembly Division gen- eral foreman describes his work:
I refer to my watch all the time. I check different items. About every hour I tour my line. About six thirty, I’ll tour labor relations to find out who is absent. At seven, I hit the end of the line. I’ll check paint, check my scratches and damage. Around ten I’ll start talking to all the foremen. I make sure they’re all awake, they’re in the area of their responsibility. So we can shut down the end of the line at two o’clock and everything’s clean. Friday night everybody’ll get paid and they’ll want to get out of here as quickly as they can. I gotta keep ’em on the line. I can’t afford lettin’ ’em get out early.
We can’t have no holes, no nothing, (quoted in Terkel, 1972:186)
At the middle level, the issues may be different, but the control mentality remains the same: “. . .a development engineer is not doing the job he is paid for unless he is at his drawing board, drawing, and so on. Higher management . . . cannot trust subo’rdinates when they are not demonstra- bly and physically ‘on the job'” (Burns, 1971:52-53). And at the strategic apex:
When I was president of this big corporation, we lived in a small Ohio town, where the main plant was located. The corporation specified who you could socialize with, and on what level. (His wife interjects: “Who were the wives you could play bridge with.”) The president’s wife could do what she wants, as long as it’s with dignity and grace. In a small town they didn’t have to keep check on you. Everybody knew. There are certain sets of rules, (quoted in Terkel, 1972:406)
The obsession with control reflects two central facts about these struc- tures: First, attempts are made to eliminate all possible uncertainty, so that the bureaucratic machine can run smoothly, without interruption. The operating core must be sealed off from external influence so that the standard outputs can be pumped off the assembly lines without disrup- tion—hence the need for rules from top to bottom. Second, by virtue of their design, Machine Bureaucracies are structures ridden with conflict; the control systems are required to contain it. The magnified divisions of labor, horizontal and vertical, the strong departmental differentiation, the rigid distinction between line and staff, the motivational problems arising from the routine work of the operating core, all these permeate the struc- ture with conflict. As Woodward noted, in these types of organizations, the ideal social and technical systems simply do not correspond:
Technical ends may best be served by conflict and pressure. Many of the conflicts that occurred in the firms studied seemed to be constructive by making a contribution to end results, and it was certainly not true to say that the most successful firms were those with the best relationships and closest identification between the staff and the company, (p. 45)
Hence, the development of the ubiquitous control mentality. The problem in the Machine Bureaucracy is not to develop an open atmosphere where people can talk the conflicts out, but to enforce a closed, tightly controlled one where the work can get done despite them.
The obsession with control also helps to explain the frequent pro- liferation of support staff in these structures. Many of the staff services could be purchased from outside suppliers. But that would expose the Machine Bureaucracy to the uncertainties of the open market, leading to disruptions in the systems of flows it so intently tries to regulate. So it “makes” rather than “buys.” That is, it envelops as many of these support services as it can within its own boundaries in order to control them, everything from the cafeteria in the factory to the law office at headquarters.
4. The strategic apex
The managers at the strategic apex of these organizations are concerned in large part with the fine-tuning of their bureaucratic machines. Hunt notes, as we saw earlier, that these are “performance organizations,” not “problem-solving” ones. Theirs is a perpetual search for more efficient ways to produce given outputs. Thus, the entrepreneur function takes on a very restricted form at the,strategic apex.
But all is not strictly improvement of performance. Just keeping the structure together in the face of its conflicts also consumes a good deal of the energy of top management. As noted earlier, conflict is not resolved in the Machine Bureaucracy; rather, it is bottled up so that the work can get done. And as in the case of the bottle, the seal is applied at the top; ultimately, it is the top managers who must keep the lid on the conflicts through their role of handling disturbances.
Direct supervision is another major concern of top management. For- malization can do only so much at the middle levels, where the work is more complex and unpredictable than in the operating core. The coordina-tion between the highly differentiated middle-level units—for example, between engineering, marketing, and manufacturing in the mass-produc- tion firm—often requires a flexible mechanism. The obvious choice would seem to be mutual adjustment. But its use is limited by the various blocks to informal communication—status differences between line and staff and between managers at different levels of the hierarchy, sharp differentiation between units at the same level of the hierarchy, and the general emphasis on formal communication and vertical reporting relationships. (In terms of our continuum of Figure 4-5, only the mildest liaison devices tend to be used in these structures—liaison positions and perhaps standing commit- tees, but not matrix structure and the like. The latter would destroy the chain of authority and the principle of unity of command, elements of central importance to the basic configuration.) So there remains the need for a good deal of direct supervision at the top. Specifically, the managers of the strategic apex must intervene frequently in the activities of the mid- dle line to effect coordination there. The top managers are the only general- ists in the structure, the only managers with a perspective broad enough to see all the functions—the means—in terms of the overall ends. Everyone else in the structure is a specialist, concerned with a single link in the chain of activities that produces the outputs.
All this leads us to the conclusion that considerable power in the Machine Bureaucracy rests with the managers of the strategic apex. That is, these are rather centralized structures; in fact, they are second in this characteristic only to the Simple Structure. The formal power clearly rests at the top; hierarchy and chain of authority are paramount concepts. But so also does much of the informal power since that resides in knowledge, and only at the top of the hierarchy does the segmented knowledge come together. The managers of the middle line are relatively weak, and the
workers of the operating core have hardly any power at all (except, as we shall see later, to disrupt the operations). The only ones to share any real informal power with the top managers are the analysts of the techno- structure, by virtue of their role in standardizing everyone else’s work. Hence, we can conclude that the Machine Bureaucracy is centralized in the vertical dimension and decentralized only to a limited extent in the hori- zontal one.
5. Strategy making
Strategy in these structures clearly emanates from the strategic apex, where the perspective is broad and the power is focused. The process of strategy making is clearly a top-down affair, with heavy emphasis on action planning. In top-down strategy making, all the relevant information is ostensibly sent up to the strategic apex, where it is formulated into an integrated strategy. This is then sent down the chain of authority for imple- mentation, elaborated first into programs and then into action plans.
Two main characteristics of this strategy-making system should be noted. First, it is intended to be a fully rationalized one, as described in our second overlay of Chapter 1. All the decisions of the organization are meant to be tied into one tightly integrated system. Exceptions flow up the chain of authority, to be handled at the level at which their effect is con- tained in a single unit, ultimately at the strategic apex if they cut across major functions. In turn, the resulting decisions flow down the chain for implementation in specific contexts. The structure that emerges is not so much one of work constellations, where groups at different levels make different kinds of decisions, as one of a hierarchy of ends and means, where managers at successively lower levels make the same kinds of deci- sions but with different degrees of specificity. For example, production decisions made at the vice-presidential level may concern what sum of money should be spent on new machinery; at the plant level, which ma- chines to buy; and at the foreman level, how these machines are to be installed. Second, unique to this structure is a sharp dichotomy between formulation and implementation in strategy making. The strategic apex formulates and the middle line and operating core implement. At least, in theory. We shall come to practice momentarily.
Figure 9-1 shows the Machine Bureaucracy symbolically, in terms of our logo, with a fully elaborated administrative and support structure— both staff parts of the organization being focused on the operating core— and large operating units but narrower ones in the middle line to reflect the tall hierarchy of authority.
Figure 9-1. The Machine Bureaucracy
Source: Mintzberg Henry (1992), Structure in Fives: Designing Effective Organizations, Pearson; 1st edition.