The argument that successive modes of organization represent efficiency ad- vances on earlier modes proses for the radical economics literature a dilemma that was apparent even in Karl Marx. I review some of the consequent tensions here.
1. Origins of the Division of Labor
Early in his chapter on the division of labor and manufacture, Marx describes an organization where a capitalist employs a number of artificers. Initially each artificer, with the help of one or two apprentices, “makes the entire commodity, and he consequently performs in succession all the operations necessary in his old handicraft way” (Marx, 1967, p. 337). Except for workshop ownership, that appears to correspond with the Communal-emh mode of organization. This continues until external circumstances change. For example, an “increased quantity of the article has perhaps to be delivered within a given time” (p. 377). As a consequence of the changes, work is temporarily reorganized. “Instead of each man being allowed to perform all the various operations in succession, these operations are changed into disconnected isolated ones, carried on side by side; each is assigned to a different artificer. . . . This accidental repartition gets repeated, develops advantages of its own, and gradually ossifies into a systematic division of labor” (p. 337). The resulting division of labor thus appears to arise as an efficiency response to changing circumstances rather than as a capitalist scheme to divide and conquer.
2. The Demise of Putting-Out
Similarly, Harry Braverman reports that the early phases of industrial capitalism “were marked by a sustained effort on the part of the capitalist to buy labor in the same way he bought his raw materials … This attempt took the form of a great variety of subcontracting and ‘Putting-Out’ systems” (f974, pp. 60-61). Braverman then goes on to observe that the “subcontracting and ‘Putting- Out’ systems were plagued by problems of irregularity of production, loss of materials in transit and through embezzlement, slowness of manufacture, lack of uniformity and uncertainty of the quality of the product. But most of all, they were limited by their inability to change the processes of production” (p. 63). Unsurprisingly, those early forms of organization were supplanted by others that had better product flow, task assignment, and incentive attributes. Again, however, the changes are drivenTby efficiency; a pernicious scheme to divide and conquer is not needed to reach those results.
Hudson’s account of the differences between the woolen and worsted branches of the textile industry in embezzlement and organizational respects is also instructive. She observes that frauds were responsible for “considerable inefficiencies and diseconomies” under the Putting-Out system in the late eighteenth century. They were more severe in the worsted branch than in woolens, mainly because there were fewer wage workers in woolens “and those there were, tended to be closely supervised in small workshops. In the worsted branch, however, woolcombers commonly embezzled their employers’ wool and the spinners reeled ‘false’ or shortvam. Combinations of operatives were often successful in ensuring that these appropriations continued with impunity” (Hudson, 1981, p. 50). The differential ease of embezzlement contributed to the more rapid transition to factory production in the worsted branch (p. 52).
3. Water Versus Handmills
Marglin reviews the handmill-watermill controversy in feudal England and observes that the centralization of milling under watermills had contract en- forcement advantages over the handmill: “It must have been extremely difficult to prevent the peasant from ‘embezzling’ the lord’s ‘rightful’ portion of grain if the milling operation took place within the peasant’s own home. Bloch mentions the ‘lawsuits which grimly pursued their endless and fruitless course, leaving the tenants always the losers’—but at great expense of time, effort, and money to the lord as well” (1974, p. 56; emphasis added). Despite the aforementioned costs, Margliti interprets the prohibition of handmills as an exercise of power and a manifestation of class conflict (pp. 55-58) since the handmill was, in his judgment, on a technological parity with the watermill.
There are two problems. First, if the watermills had only policing benefits and offered no technological advantages over handmills, then the obvious way to mill grain would be to concentrate all of the handmills at a central location and insist upon their use there. Inasmuch as handmills were sunk costs, investment in new equipment would thereby be avoided. But second, and more important, assessing the choice of milling technique in technology versus power terms is unacceptable if transaction cost differences are operative—as they plainly were.
Transaction cost disabilities of two types can be associated with the local handmilling of grain. First, actual compensation will differ from reported compensation in favor of those in the peasant population who are most prepared to lie, cheat, and steal. Such a compensation scheme is, among other things, shot through with adverse selection incentives. Second, and related, the embezzlement of grain will elicit protective responses by lords, policing that is costly and is appropriately included in the social calculus.
That is not to say that metering is an unmixed blessing and cannot be taken to excess. It can be and sometimes is. The issues here, however, are not the ones addressed by Marglin but arise in conjunction with the economics of atmosphere (Williamson, 1975, pp. 37—38) and in distinguishing between perfunctory and consummate cooperation (pp. 69-70). Those are important matters with which the organization of work is legitimately concerned. They are briefly considered in Chapter 10, but a much more complete treatment is needed.
4. Inside Contracting in Steel
The principal historical study to which Bowles and Gintis (1976) refer is the article by Stone, in which the transformation of the steel industry is examined. According to Stone, the organization of the steel industry in the late nineteenth century corresponded approximately to the Inside Contracting system described and discussed above. The Amalgamated Association of Iron, Steel, and Tin Workers, which was the union to which the skilled workers belonged and was reported to be the strongest union of its day, gave “the skilled workers authority over every aspect of steel production” (Stone, 1974, p. 64). The costly haggling and inflexibility to which Inside Contracting is subject predictably resulted. Operating inefficiency developed, and innovations were suppressed. Examples cited by Stone (1974, pp. 64-65) include the following:
- The consent and approval of the executive committee within each department was needed to fill a vacant position.
- The details of the work were subject to recurring dispute.
- Output per worker was restricted.
- Production procedures were proscribed: The “proportion of scrap that might be used in running a furnace was fixed; the quality of pig- iron was stated; the puddlers’ use of brick and fire clay was forbidden, with exceptions; the labor of assistants was defined. ”
- Presumably to perfect and maintain their monopoly over jobs, skilled workers were prohibited from teaching other workers.
- Changes in the physical plant could not be made without the approval of the executive committee of the union, which prevented tl^e company from realizing greater labor productivity by reorganizing or mechanizing labor tasks.
- Innovations of a labor-saving kind were discouraged: “The many innovations introduced between 1860 and 1890, of which the most notable was the Bessemer converter, increased the size and capacity of the furnaces and mills, but they generally did not replace men with machines. ”
The resulting inefficiencies were apparent to the companies. Andrew Carnegie and Henry Clay Frick resolved to challenge the union at Carnegie’s Homestead mill, which was reputed to be the strongest lodge of the Amalga- mated Association. A lockout was ordered in 1892, and Frick announced that the mill would thenceforth be operated nonunion. Violencg resulted, with members of the union pitted against scabs and Pinkerton agents. The support of state and federal governments helped Carnegie and Frick prevail. Whether emboldened by the success of Carnegie and Frick or out of realization that their competitive viability rested on their being likewise able to disaffiliate with the Amalgamated Association, other steel companies challenged and beat the union as well. Association membership, which peaked at twenty-five thousand in 1892, was down to ten thousand in 1898. By 1910 the entire steel industry was nonunion. The effects of breaking the power of the skilled workers are summarized by Stone as follows:
The decade that followed the Homestead defeat brought unprecedented develop-ments in every stage of steelmaking. The rate of innovation in steel has never been equaled. Electric trolleys, the pig casting machine, the Jones mixer, and mechanical ladle cars transformed the blast furnace. Electric traveling cranes in the Bessemer converter, and the Wellman charger in the open hearth did away with almost all the manual aspects of steel production proper. And electric cars and rising-and-falling tables made the rolling mills a continuous operation, [Stone, 1974, p. 661
Breaking the union’s grip on procedures did not, however, assure the steel industry that its labor force would thereafter be organized efficiently. Such efficiency required that new institutional structures be devised. The objectives of the steps taken seem mainly to have been designed to (1) supply affirmative incentives for productivity, (2) tie the interests of the workers to the firm over the ibng term, (3) develop the requisite work skills among inexperienced workers, and (4) organize the work to preclude subsequent loss of control by the company. Stone interprets the various steps taken to realize those objectives as pernicious and evidence of a continuing class struggle between workers and employers. But there is another possibility: The incentive to challenge the union in the first place and the efforts to organize labor subsequently were principally geared to achieving efficiency, the rewards for which, once the new methods were imitated by rivals and rates of return were driven down to competitive levels, were diffused throughout society.
Put differently, were it not that the Amalgamated Association had pro-hibited efficiency gains and impaired efficiency incentives, Carnegie’s challenge to the union is plausibly interpreted as a contest for raw power—its purpose being to redistribute income away from workers in favor of capital. Given, however, the large efficiency gains that Stone reports, the efficiency hypothesis (or a combined efficiency-power hypothesis) cannot be rejected. The efforts to organize labor in the post-Homestead era are also broadly consistent with the efficiency hypothesis.
Stone nevertheless asserts that the benefits of reorganization described above could have been realized without the adverse, oppressive effects of hierarchy. She contends that “a system of job rotation, one in which the workers themselves allocated work, would have been just as rational and effective a way of organizing production” (Stone, 1974, p. 66). While the details of such an organizational arrangement are not supplied, the rotation arrangement appears to correspond to the Communal-emh system91 described in section 3 and evaluated above. That the Communal-emh mode has the worst efficiency properties of all six modes examined in this chapter might be arguable. But that it is shot through With adverse incentives and maladaptive attributes is, I think, beyond dispute.
5. Forward Integration in Meatpacking
The power approach to vertical integration appears to assume that everything that feasibly can be integrated will be—or at least that is the result that William Ouchi and I reached in attempting to interpret Charles Perrow’s views on integration (Williamson and Ouchi, 1981, pp. 321-24). The belief, evidently, is that more integrated systems are more powerful than less. Branco Horvat’s position on this is explicit: “Corporations strive for vertical integration in order ta,control prices and other conditions of supply… Ever expanding corporations try to internalize all decisions concerning production, buying, selling, and financing” (1982, pp. 15-16; emphasis added).
The efficiency hypothesis, by contrast, is that vertical integration will occur selectively rather than comprehensively, that mistaken vertical integration can rarely be sustained, and that more efficient modes will eventually supplant less efficient modes—though entrenched power interests can sometimes delay the displacement. Evidence bearing on the selectivity hypothesis is developed in Chapter 5.
A particularly interesting confrontation between efficiency and power occurred in the meatpacking industry late in the nineteenth century. Here, moreover, power was aligned against integration.
Gustavus Swift believed that the practice of shipping Western cattle to Eastern markets alive rather than slaughtered and dressed was unnecessarily expensive. He proposed to realize economies by butchering the animals in the West and shipping the meat east in refrigerated cars, where it would be received and distributed from a network of refrigerated storage houses. Not only did this involve investments in specialized assets, the value of which would be limited should Swift’s strategy fail, but itTtiet determined opposition:
Railroads, startled by the prospect of losing their livestock business, which was an even greater producer of revenue than grain on the west to east routes, refused to build refrigerated cars. When Swift began to construct his own, the Eastern Trunk Line Association refused to carry them. Only by using the Grand Trunk, then outside of the Association, was Swift able to bring his cars east. At the same time he had to combat boycotts by local wholesalers, who in 1886 formed the National Butchers’ Protective Association to fight “the bust.” These butchers attempted to exploit a prejudice against eating fresh meat that had been killed days or even weeks before, more than a thousand miles away. [Chandler, 1977, p. 300]
Despite the opposition from the railroads and butchers, Swift’s “high quality and low prices” combined with “careful scheduling” prevailed (p. 300). Other packers soon thereafter realized that “if they were to compete with Swift in the national market they must follow his lead” (p. 301). Efficiency thus evidently swamped the resistance of entrenched power interests— though this is not to say that Swift won easily. Had the efficiency gains been much smaller or had the entrenched interests been better organized, power might well have defeated Swift. I submit, however, that large efficiency differences place entrenched interests under great strain. (The aphorism “if you can’t beat them, join them” is often the way by which such interests secure relief.)
Source: Williamson Oliver E. (1998), The Economic Institutions of Capitalism, Free Press; Illustrated edition.