In order to avoid imputing benefits to hierarchy that can be had, in some degree, by simple nonhierarchical associations of workers, it will be useful to begin with an examination of worker peer groups. These groups involve collective and usually cooperative activity, provide for some type of other-than-marginal productivity and income-sharing arrangement, but do not entail subordination. Such collective organizations offer prospective advantages in indivisibility, risk-bearing, and associational respects.
Indivisibilities of two types can be distinguished. The first type involves scale economies associated with physical assets and is reasonably familiar. Larger scale units, provided they are utilized at design capacity, permit lower average costs to be realized. The group may thus be formed in order to assure that utilization demands will be sufficient.
The second type, somewhat less familiar, are the indivisibilities asso- ciated with information. Radner observes that “the acquisition of information often involves a ‘set up cost’; i.e., the resources needed to obtain the information may be independent of the scale of the production process in which the information is used” (1970, p. 457). Consequently, groups may also be formed in order to economize on information costs.
The realization of economies of either of these types, however, does not clearly imply, as a technological imperative, collective organization. Thus, technologically speaking, there is nothing that prevents one individual from procuring the physical asset in requisite size to realize the economies in question and contracting to supply the services of this asset to all of the members of the group. Similarly, there is no technological bar that prevents one individual from assuming the information gathering and dissemination function. All parties, suppliers and users of the specialized services alike, could be independent, yet scale economies of both types could also be fully realized. If, therefore, such specialization fails to materialize, it is not because there are technological impediments to monopoly ownership of the physical assets and information services in question. Rather, the problems are to be traced to transactional difficulties that attend market exchange in these circumstances. Accordingly, the incentive to collectivize activities for which indivisibilities are large in relation to the market are transactional in origin.
The arguments here are familiar. They are illustrated in my discussion of Stigler’s treatment of vertical integration in Chapter 1 and are developed more extensively in Chapter 5. Briefly, the argument is that market contracting of the complex contingent claims variety is infeasible, on account of bounded rationality, while sequential spot contracting is hazardous where suppliers acquire nontrivial learning-by-doing advantages and an absence of opportunism cannot safely be presumed. The upshot is that, where indivisibilities of either physical asset or informational types are large and learning-by-doing predictably (albeit uncertainly) obtains, collective organization may arise instead.
The advantage of collective organization, with respect to the exchange of the specialized services in question, is that the incentives of the parties to joint profit maximize is greater. Indivisible physical assets are owned and utilized by the group, rather than privately; associated learning-bydoing advantages thus accrue not to the strategic benefit of the monopolistic supplier but are shared by the membership instead. Informationgathering and dissemination likewise proceed with less concern for strategic nondisclosure or distortion of the data, since individual parties are unable, given the ownership arrangements, to appropriate the private gains to which such opportunistic behavior would otherwise give rise.
2. Risk-Bearing Factors
Group affiliation may be sought for insurance purposes if such member- ship can provide income guarantees to buffer the effects of unanticipated contingencies on terms superior to that which market insurance can provide. Recall in this connection the disabilities of insurance markets discussed in Chapter 1. A group will have an advantage over the market to the extent that it is better able to (1) limit membership in a discriminating way, thus mitigating problems of adverse selection attributable to ex ante information impactedness and opportunism on the part of insurance purchasers, and (2) check malingering and other ex post manifestations of moral hazard. Advantages of the first kind require that members be able to ascertain the characteristics of applicants in a discerning fashion, while those of the second require that performance be accurately assessed.
Groups of three types can be distinguished: groups engaged in unrelated tasks, groups engaged in a common task, and groups performing an integrated set of tasks. Although the peer group may have little to offer for purposes of membership screening or as a check on malingering if the group is of the unrelated task variety, the peer group may have an advantage, in relation to the market, in both of these respects where members are engaged in common or integrated tasks. Members of common or integrated task groups not only know the requisite attributes to look for in admitting a new member but are able, as a by-product of their working relationships, to mutually monitor one another — virtually automatically, at little incremental monitoring expense. Peer group affiliation may thus be sought in part because the group, rather than the individual, is better able to bear risks and because, in relation to the market, the group has superior ex ante screening and ex post monitoring capabilities.
3. Associational Gains
It will be convenient, for the purpose of discussing associational gains, to assume that markets experience no transactional disabilities except as these are explained by the attitudes that individuals have toward completing transactions by one mode of organization rather than another. Again, unrelated, common, and integrated task groups can be distinguished. Since work in an unrelated task group neither requires recurring contacts nor provides a strong basis for communicating common experience, associational gains in such groups are apt to be small.22 Attention is accordingly focused on common and integrated task groups.
Associational benefits can accrue to peer groups through increased productivity among members of the group who feel a sense of responsibility to do their fair share as members of a group but, left to their own devices, would slack off. Also, and more important, a transformation of “involvement” relations, from a calculative to a more nearly quasimoral mode, obtains. Especially if the metering of transactions among the members of a group is consciously suppressed and kept on an informal basis, which in a peer group is to be expected, involvement relations between the members of a group are less calculative than if the same transactions were to be conducted across a market. A written record of transactions is less fully maintained, and obligations are commonly settled in other than cash or pecuniary equivalent terms. At least some individuals experience a heightened sense of well-being in this way.
I hasten to add, however, that informal involvement relations are not identically valued by all individuals. Other things being equal, peer groups (or other internal modes) will presumably be organized by those individuals who find market transactions less satisfying than a nonmarket relationship, while markets will be favored by those who prefer a more exacting, transaction- specific correspondence between rewards and deeds. Thus, given diversity of tastes for involvement relations, neither mode need fully displace the other, but both modes will coexist and each will appeal selectively to that part of the population to whose involvement tastes it most nearly corresponds, ceteris panbus.
Source: Williamson Oliver E. (1975), Markets and hierarchies: Analysis and antitrust implications, A Study in the Economics of Internal Organization, The Free Press.