Attention is shifted here from a discussion of the disabilities of vertical integration to consider the general size limits to which the firm is subject. The issue can be put as follows: Holding the degree of vertical integration and organization form constant, are there simple size impediments that radial expansion of the firm eventually incurs? Bounded rationality, bureaucratic insularity, and atmospheric consequences are considered.
1. Bounded Rationality
Bounded rationality gives rise to finite spans of control78 together with the specialization of communication and decision-making functions. It is sufficient for our purposes here, however, merely to emphasize the finite span of control consequence. Faced with this condition, radial expansion of the enterprise requires that additional hierarchical layers be added. The consequence of this for internal communications has been examined elsewhere in conjunction with what I have referred to as the “control loss phenomenon” ( 1970, Chap. 2).
Boulding observes in this connection that “almost all organizational structures tend to produce false images in the decision-maker, and that the larger and more authoritarian the organization, the better the chance that its top decision-makers will be operating in purely imaginary worlds” and concludes that image distortion is ultimately responsible for diminishing returns to scale (1966, p. 8). Although some of this image distortion may be done strategically (and thus is attributable to the pairing of opportunism with information impactedness), the main points that I would emphasize here are (1) the extended hierarchy is made necessary by bounded rationality, and (2) even efforts to communicate in a purely instrumental way are subject to serial reproduction losses (Bartlett, 1932). Consequently, unless offset by other factors (of which the acquisition of monopoly power is one possibility), radial expansion of the firm eventually exhibits diminishing returns.
2. Bureaucratic Insularity
The issue here is related to what Michels has referred to as the iron law of oligarchy” — which is a manifestation of bureaucratic opportunism. Although Michels was principally concerned with voluntary organizations, especially labor unions and political parties, his remarks have relevance in other hierarchical contexts as well. Their special significance for present purposes is that bureaucratic insularity varies directly with organizational size (Michels, 1966, p. 71).
Nominally, and according to the letter of the rules, all acts of the leaders are subject to the ever vigilant criticism of the rank and file. In theory the leader is merely an employee bound by the instruction he receives. . . . But in actual fact, as the organization increases in size, this control becomes purely fictitious.
Given finite spans of control, increasing firm size leads to taller hierarchies in which leaders are less subject to control by lower-level participants. The resulting bureaucratic insularity of the leadership permits it, if it is so inclined, to both entrench and engross itself.
As compared, however, with a voluntary organization, the matter of legitimacy in the business firm is less in relation to lower-level participants than it is to the stockholders. The control problems in each case nevertheless turn on identical considerations, namely, the information impacted- ness issue. Since problems of stockholder control in this sense typically become more severe as the firm grows in size and complexity (see Section 3 of Chapter 8), larger size is associated with greater opportunities for discretion. Where the leadership exercises these opportunities by permitting slack and indulging in corporate personal consumption, size limitations necessarily follow — especially if lower- level performance varies directly with higher-level example, which normally is to be expected.
It is noteworthy, however, that business firms differ from other types of bureaucracies in that voting for the board of directors can be concentrated through share ownership (through direct purchase of shares, tender offers, and the like). This is not possible in most other types of organization where one-man, one-vote rules tend to predominate. This is a basic distinction between the business firm and bureaucracies more generally. Incumbent managements can be displaced more easily as a result — though I would concede that effecting displacement in the large firm is not always easy (see Chapter 9).
3. Cooperation Limits among Lower-Level Participants
Internal organization affords atmospheric advantages of two sorts. First, it offers associational relations that may be valued. Second, it supports involvements of a continuing sort in which members are more sensitive to part- whole relations. Increasing size, however, easily upsets this latter type of systems concern.
As Dahl and Lindblom observe (1963, p. 225), large size and hierarchical structure favor impersonality among the parties, which is more characteristic of a calculative orientation. This is partly attributable to the specialization of information gathering and the more limited disclosure of information (on a need-to-know basis) as firm size and hierarchical structure are extended ( organization form held constant). The corresponding assignment of decision- making to what are perceived by lower-level participants to be remote parts of the enterprise also contributes to this result. Although efficiency purposes are commonly served in this way, these must be weighed against the loss of moral involvement. To the extent that nonknowledgeability and nonparticipation impair moral involvement and larger-size results in role incompatibility, which it apparently does (March and Simon, 1958, p. 98), a more calculative orientation is to be expected. The zone of acceptance of the employment contract is narrowed, which serves to limit the attractiveness of an employment contract in relation to a sales contract. Put differently, attitudes of voluntary cooperation are supplanted by a quid pro quo orientation. Since, moreover, each individual in the large organization is small in relation to the whole, so that the percentage effects of individual behavior are perceived to be in- substantial, the large organization may be thought to be better able to tolerate perfunctory performance or even deviant conduct.
A reduction in group disciplinary pressures which, in a smaller firm, operate in the service of enterprise viability (by enforcing norms which extend the effective influence of supervisors) thus obtains. As Olson puts it, where “each member … is so small in relation to the total that his actions would not matter much one way or another…, it would seem pointless for one [member]… to snub or abuse another for a selfish [antifirm] action, because the recalcitrant’s action would not be decisive in any event” (Olson, 1968, p. 62). Indeed, should alienation from the enterprise develop among individual components of the firm, small-group powers may even be turned against the enterprise in subtle but significant respects. The disaffected group may allocate rewards and sanctions in a perverse fashion. Industrial sabotage is an extreme manifestation of this condition.
Source: Williamson Oliver E. (1975), Markets and hierarchies: Analysis and antitrust implications, A Study in the Economics of Internal Organization, The Free Press.