DOJ Antitrust Division, Business Review Request Letter 00-5, Containers America LLC (1999)
May 14, 1999
Garret G. Rasmussen
202-457-6343
grasmussen@pattonboggs.com
BY HAND
Hon. Joel I. Klein
Assistant Attorney General
Antitrust Division
Department of Justice
Washington, D.C.
Dear Joel:
This is a request for a business review letter concerning a joint venture. I represent Containers America LLC ("Containers America"), which is a newly-formed joint venture composed of five regional steel drum companies, each located in a different geographic market. The purpose of the joint venture is to enable its members, all of whom are regional companies, to compete against national companies for national contracts and to engage in group purchasing in order to qualify for volume discounts.
I have advised Containers America that the proposed group purchasing and joint bidding for national contracts are lawful, procompetitive, joint venture activities. These activities increase competition for national contracts, increase output, and reduce prices. Indeed, absent the joint venture, the individual members would remain excluded from a growing segment of the market which is composed of national buyers that are demanding a single, national supplier. Based on my advice, Containers America already has started its group purchasing activities, and it intends to bid on upcoming national contracts.
This request involves many of the same issues that the Division addressed in its May 20, 1997 business review letter concerning Russell Stanley's joint selling venture in the steel drum industry. In that case, Russell Stanley proposed acting as the general contractor while other steel drum manufacturers acted as subcontractors on bids for national supply contracts that could not be served efficiently by Russell Stanley alone. The Division wrote that it did not intend to take any action regarding that venture.
Containers America was formed in March 1999 by the following five regional steel drum manufacturers:
The members of Containers America primarily sell 55 gallon steel drums. Approximately 35 million steel drums are sold annually in the United States. Collectively, the Containers America members sell approximately 5 million steel drums.
Each of the five members of Containers America competes for sales of steel drums in a different geographic market, defined by a circle with a radius of between 100-200 miles from the member's plant. Within each of these regional markets, the members of Containers America face significant competition from other manufacturers of steel drums that are not members of Containers America.¹
¹ CP Louisiana has six competitors (none of whom are members of Containers America) located within 300 miles of its New Orleans plant, and its market share in that area is approximately 15%. North Coast has five competitors (none of whom are members of Containers America) within 200 miles of its plant in Cleveland, and its share of that market is approximately 30%. General Steel Drum has two competitors (none of whom are members of Containers America) that are also based in North Carolina. General Steel Drum's share of the North Carolina market is approximately 30%. Nesco is the only manufacturer of steel drums in St. Louis, and its share of the St. Louis market is approximately 60%. Trilla has five competitors (none of whom are members of Containers America) located in Chicago. Its share of the Chicago market is approximately 35%.
Because steel drums are bulky and because it is difficult to arrange backhauls, transportation costs are high, and, as a percentage of cost, they increase in direct proportion to the distance of shipment. Thus, most sales by the members of Containers America are to customers located near the plant. Between 50%-80% of each members sales is within 150 miles of each member's plant. Moreover, relatively few sales are made more than 200 miles from a plant. This is significant because each member's plant is located more than 300 miles from the next closest member's plant.²
Members occasionally make sales outside their geographic market, but only under special circumstances. One circumstance is the absence of another manufacturer located closer to the customer. Accordingly, Containers America members occasionally compete with each other. For example, General Steel Drums, located in Charlotte, North Carolina has made some sales into Florida, as has CP Louisiana, located in New Orleans. However, there is now a new entrant in Florida, which will make it more difficult for General Steel Drums and CP Louisiana to compete there. Members also occasionally make shipments outside their markets as an accommodation to a large customers located primarily within the market, but with some locations outside the market.
There is an emerging national market for steel drums. Customers with operations throughout the United States are increasingly demanding one national supplier,³ and there are a number of suppliers with multiple plants that are capable of serving national customers.⁴ The national suppliers include the following:
² The members with plants located closest to each other are Trilla and Nesco, whose plants are 308 miles apart. North Coast and Trilla are approximately 350 miles apart. All the others are even further apart.
³ Among the companies that have solicited national contracts are Dow Chemical, Mobil, Amoco, Witco, BASF, and Reichhold. Although some of these companies may at the same time request bids for regional supply contracts, they generally do so only to gain negotiating leverage with national suppliers.
⁴ National customers for steel drums generally have operations along the Mississippi River from the Gulf Coast to Chicago, as well as in the mid-west industrial belt, the mid-Atlantic states, and the Houston area.
⁵ Van Leer will become even more dominate, if, as announced, it merges with Huhtamaki, a Finnish firm that also manufacturers of steel drums.
While the five manufacturers who have formed Containers America are frequently the low bidders in their respective regions, none has ever won a national supply contract. That lack of success is of great concern because multi-regional purchasers are increasingly moving toward national contracts. In fact, each of the five single-plant manufacturers has either lost, or is faced with the prospect of losing, significant business to national suppliers:
Although none of the members of Containers American, acting alone, can effectively compete for multi-regional or national supply contracts, collectively they can serve the same regions as the multi-regional manufacturers who successfully bid on national supply contracts. Thus, as discussed below, the joint venture will allow these five manufacturers to effectively compete for the first time in the market for national supply contracts.
Every manufacturer of steel drums must buy steel, fittings (plug/flange sets, rings, gaskets, and nuts/bolts), lining, and paint. Suppliers of these materials typically offer volume discounts to large purchases. Because of the volume discounts, large manufacturers with multiple plants have been able to purchase materials at significantly lower prices than single-plant manufacturers, such as each of the five principals of Containers America. Given that steel constitutes approximately 50% of the cost of a steel drum and that the other materials account for another 15% of the cost, the large multi-regional manufacturers have a significant cost advantage over unaffiliated single-plant manufacturers. In fact, the individual purchases of both
Van Leer and Grief are equal to, or greater than, the combined purchases of the five principals of Containers America.
The joint venture intends to purchase sufficient quantities of steel, fittings, lining, and paint in order to qualify for volume discounts. A member will not be required to purchase through the joint venture, but may be asked to specify in advance how much of a particular input it will buy if a particular price can be obtained. The purchasing director, who initially will be the purchasing officer of one of the members, will approach manufacturers and ask for the lowest price at which that manufacturer will sell at the specified volume. He may also ask other members how low a price they have been able to obtain from particular manufacturers and use that information to negotiate even lower prices for the group. In order to qualify for the largest volume discounts, members might also agree to all use common specifications for a particular input.
The pro-competitive benefits of such group purchasing are substantial. The potential cost savings realized by the five single-plant manufacturers who have formed Containers America could exceed 10%. The cost savings will allow the members of Containers America to offer lower prices to consumers and eliminate the competitive advantage currently enjoyed by the multi-plant manufacturers who already receive volume discounts from suppliers. Collaboration among competitors to achieve "lower costs of production . . . [is] consistent with the purposes of the antitrust laws." VII Areeda, Antitrust Law ¶1504. Indeed, that justification has been recognized as sufficient to offset group purchasing agreements among companies who compete much more directly than is the case here. See, e.g., Instant Delivery Corp. v. City Stores Co., 284 F. Supp. 541 (E.D. Pa 1968).
In contrast, there is little, if any, potential for anti-competitive harm from group purchasing by the members of Containers America. One reason group purchasing will not lead to any significant anti-competitive effects in the steel drum market is that there is only de minimis competition between the five principals of Containers America. In fact, as described above, each competes primarily in a separate regional market, and none of the five principals has ever competed successfully in the market for national supply contracts.
Even if the five members did compete against one another to a greater degree, and even if each had market power, there still would be little potential for anti-competitive effects as a result of the group purchasing. Although steel and the other group-purchased materials may account for as much as 60%-70% of the cost of steel drums, the other pricing variables -- such as labor, overhead, profit margins, and the materials not purchased through Containers America -- would make it very difficult for one company to predict with a sufficient degree of precision the bid that its fellow member will submit for a particular job.⁶ Thus, it is unlikely that the group purchasing will result in anticompetitive price coordination.
Further, even if the joint purchasing were sufficient to allow one company to predict with reasonable precision the bid of another member, anticompetitive price coordination would still be impossible because each of the companies faces significant competition in its regional market from steel drum manufacturers that are not part of the joint venture. "A competitive market structure . . . eliminates the fear that information exchanges can bring about price coordination." See VII Areeda, Antitrust Law ¶1503
Finally, the joint venture will not have any monopsony power because its purchases will constitute only a small fraction of those made by all steel drum manufacturers. Indeed, the steel drum industry as a whole purchases only a small fraction of the rolled steel and paint sold in the United States. As such, any agreements that the principals of Containers America reach regarding purchasing guidelines -- such as specifications for certain materials, freight equalization requirements, and the frequency of delivery by the supplier -- will not result in any anti-competitive harm.
When, as here, joint selling arrangements are necessary to bring new competition into a market, they are tested under the rule of reason and upheld even though they literally may involve "price fixing" among members who occasionally compete with each other. VII Areeda, Antitrust Law ¶1500 ("reasonable collaboration among competitors does not violate Sherman Act §1").
⁶ To be sure, steel accounts for 50% of the cost of steel drums -- higher than the 20% safe harbor set forth in the Antitrust Division & FTC Statement of Policy in Health Care. Nevertheless, the potential for anticompetitive harm remains low because, among other reasons, the four companies have freedom to set their own prices for steel drums and because they are not required to buy all their steel through the group.
The rule of reason applies to the proposed joint bidding for two reasons. First, it is part of a legitimate joint venture. NCAA v. Board of Regents, 468 U.S. 85 (1984). Second, the joint bidding increases output because it results in new competition for national contracts. See Broadcast Music Inc. v. CBS, 441 U.S. 1 (1979)
In Broadcast Music, the Supreme Court held that the antitrust laws should not be applied in a literal or mechanical way that has the perverse effect of outlawing conduct that actually benefits consumers. Indeed, the reasoning of Broadcast Music is directly relevant here. In that case, composers and publishers who held copyrights to music and who competed in the licensing of those rights combined in two associations, ASCAP and BMI, which sold various packages of copyrights. The Supreme Court, reversing the lower court, ruled that, even though there was "price-fixing in the literal sense," the defendants had not acted unlawfully because they had expanded output. The Supreme Court held that "[n]ot all arrangements among actual or potential competitors that have an impact on price are per se violations of the Sherman Act or even unreasonable restraints." 441 U.S. at 23. In fact, reasoning that "easy [per se] labels do not always supply ready answers," the Supreme Court held that the "literal approach does not alone establish that this particular practice is . . . 'plainly anti-competitive' and very likely without any 'redeeming virtue.'" 441 U.S. at 8, 9. Looking to the economic effect of the blanket licenses, the Broadcast Music Court ruled that the per se doctrine should not apply because creation of the blanket licenses was a pro-competitive response to consumers' demand to avoid the high transaction costs of dealing with multiple suppliers. Id. at 20.
Another relevant case is Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210 (D.C. Cir. 1989). There, Judge Bork, writing for the D.C. Circuit, refused to apply the per se doctrine to an agreement among competing and potentially competing van line companies who combined through a national corporation to offer common prices for nationwide moving services. The court reasoned the collaboration was "designed to make the van line more efficient rather than to decrease the output of its service and raise its rates." 792 F.2d at 211. Because the restraints were ancillary to a pro-competitive contractual integration among competitors and potential competitors, the court applied the rule of reason and found no antitrust violation.
Similarly, the reasoning of the leading antitrust treatise provides for rule of reason analysis here:
Price fixing . . . or similar legal labels do not have absolute meaning. Before applying a rule about such matters we need to classify the conduct before the antitrust tribunal . . . . There are two routes to classification: linguistic and functional. The linguist consults a dictionary to determine whether he can reasonably describe the conduct before the court as, say, "price fixing. . . ." But when applying a judge-made rule, the court obviously knows the rule's rationale and should, as many do, define its scope in terms of the policies that give it life. Any other approach is inconsistent with those policies . . . . Virtually all horizontal agreements among competitors ultimately affect price. . . . Obviously some of these impacts will be wholly pro-competitive. Prohibiting them categorically as per se illegal price fixing would deprive society of beneficial arrangements promoting antitrust goals. . . . Suppose, for example, that two small producers, A and B, in a highly competitive market form a joint venture partnership, C, to realize economies of scale in purchasing a component or in selling a finished product. Assume that C negotiates a buying or selling price with suppliers or customers. In every ordinary sense, A and B are jointly fixing their buying or selling price; competition between them on the buying or selling side of the market is eliminated. Yet no one believes that they violate the per se rule against 'price fixing.' The obvious explanation is that the rationale of the per se rule does not apply. Joint selling (or buying) ventures are not always or almost always pernicious. Indeed, they are not pernicious at all in the absence of market power.
VII Areeda, Antitrust Law ¶ 1510.
The proposed joint bidding for national contracts is pro-competitive because it creates a new competitor for those contracts without reducing any existing competition. By increasing the number of bidders for national supply contracts, the joint bidding expands output and reduces prices. As discussed above, the members of Containers America, standing alone, cannot effectively compete for national supply contracts against multi-regional manufacturers of steel drums who have plants in different regions. At most, they can compete for a portion of the national account's business.
In contrast, there is little potential for anti-competitive effect because the joint bidding through Containers America will be limited to national and multi-regional contracts, and the members of the joint venture will remain free to offer competing facility-specific bids. Moreover, discussions among members regarding prices for steel drums will be limited to national and multi-regional bids.
In conclusion, we believe that the proposed group purchasing and group selling are pro-competitive. We would, however, appreciate your review of the proposed joint purchasing and joint selling.
If you need any additional information, please let me know.
Sincerely,
Garret G. Rasmussen