VERTICAL INTEGRATION AND COLLUSION THE ROLE OF WHOLESALERS IN FACILITATING CARTEL STABILITY

Margaret Levenstein
University of Michigan

This paper examines, both theoretically and historically, the role of wholesale distributors in facilitating collusion in several nineteenth century industries. Earlier work on collusion in the nineteenth century bromine cartel suggests that wholesale distributors played a critical role in monitoring compliance with output restriction agreements (Levenstein 1989). These firms also invested in reputations as maintainers of collusive prices (or as price cutters in the case of "outsiders" to the cartel). Attempts by manufacturing firms to integrate forward into the distribution of their own products were seen as a direct threat to the cartel. More recent studies of the structure of collusion in the chloroform and bleach markets during the same period suggest that wholesalers played similar, though less formal, roles in facilitation collusion in those markets (Levenstein 1991, Chapter 3).

There are three aspects to the questions raised by these examples, focusing on the structure of the collusive organizations themselves, the nature of competition in these industries, and the organization of the participating firms. First, it is necessary to improve our knowledge of the vertical structure of turn-of-the-century collusive organizations in order to understand the functioning and incentive properties of many of these arrangements. Descriptions of nineteenth century pools frequently refer to common selling agencies begin established, or relied upon, but this aspect of pool structure is usually not explored in detail. The research on the cartels in which the Dow Company participated, which provides a more intimate picture of the operating of these cartels than many histories which are forced to rely on anti-trust prosecutions for evidence of cartel activity, suggests that these vertical structures deserve more attention than they have received in the past. The paper reviews the existing literature on nineteenth century pools, focusing on those in the chemical industry (see, for example, Ripley 1916, Eskew 1948, and Dewing 1924). It then analyzes the incentive properties of these organization and compares the more well-known characteristics of joint selling agencies with those of cartels which relied on independent wholesale distribution firms. (Bernheim and Whinston 1985 provides a rare formal treatment of this problem.)

Second, this period was one characterized by frequent charges of "cut throat competition." The most convincing explanation of this rampant price competition is that it was the result of the increased integration of markets (resulting from declines in transportation costs) combined with a high fixed cost (and relatively highly leveraged) structure (Lamoreaux 1985.) This created very impatient firms with no history or pattern of cooperation, and therefore, less ability to refrain from competition. (See Fudenberg 1991 for a review of the literature relating patience, reputation, and the ability to sustain cooperation).

This paper does not challenge that explanation of the increased price competition observed. However, it does raise another explanation of why some industries were less successful during this period than previously at sustaining collusive pricing strategies. This explanation, like the previous one, focuses on the growth of "large" firms, but what is important to this argument is not the increases in horizontal size or fixed costs, but rather the increasing vertical integration of these firms. In many instances, it appears, wholesale distributors were crucial to maintaining earlier collusive arrangements. As a result of vertical integration, these independent monitors of collusive arrangements were displaced, and collusion could no longer by successfully enforced. (Of course, if the integrating firm achieved the high market share that it usually aspired to, new methods of achieving monopolistic pricing were available to replace the old, ineffective ones.)

Second, once we recognize the importance of the vertical, as well as the horizontal, nature of these collusive organizations, new questions are raised about our understanding of changes in the organizations of the firms themselves (as opposed to the organization of markets). In contrast to the argument put forth by Chandler (1977) that horizontal and vertical integration were alternative strategies facing firms, we need to understand the intimate connection between the horizontal and vertical changes observed. In the cases of both DuPont and General Motors, we observe simultaneous changes in the horizontal and vertical structure of the industry. These firms were not willing to make the risky investment in vertical integration without the assurance of the modicum of control over market price afforded by the horizontal mergers (and the resulting increased market share) pursued.

Similar concerns dominate the thinking of the Dow Chemical Company in its consideration of vertical integration during this period. However, for the Dow case we have more detailed descriptions of the collusive organizations that existed in the various markets in which Dow participated. We also have explicit discussions of the possible disruptiveness of vertical integration to those collusive arrangements, and the importance of achieving monopoly power concomitant with vertical integration if this was to be a successful strategy.