William H. Sullivan II v. National Football League, & Members of the National Football League

The National Football League and twenty-one organizations owning NFL franchises (referred to collectively as the “NFL”) appeal the judgment entered against them after a jury found that the NFL violated the antitrust laws by restricting owners of member football clubs from selling shares in their teams to the public. Plaintiff-appellee, William H. Sullivan, former owner of the New England Patriots football team (the “Patriots”), was awarded a total of $51 million in damages for the losses Sullivan incurred when he had to sell the Patriots to a private buyer after the NFL prevented him from offering 49% of the team to the public in the form of publicly traded stock. Because several prejudicial errors were committed during the trial, we vacate the judgment and remand for a new trial.

I. BACKGROUND

Under Article 3.5 of the NFL’s constitution and by-laws, three-quarters of the NFL club owners must approve all transfers of ownership interests in an NFL team, other than transfers within a family. In conjunction with this rule is an uncodified policy against the sale of ownership interests in an NFL club to the public through offerings of publicly traded stock. The members, however, retain full authority to approve any given transfer by a three-quarters vote according to Article 3.5.

Sullivan owned the Patriots from the team’s inception in 1959 until October of 1988. When Sullivan formed the Patriots, he and his partner sold non-voting shares of the team to the public beginning in 1960. At that time, the Patriots were in the old American Football League (“AFL”), which was separate from the NFL, and which had no policy against public ownership of teams. In 1966, the AFL and the old NFL merged into a single league. Under the terms of the merger, the new NFL would adopt the old NFL’s policy against public ownership. The Patriots, however, were allowed to retain their level of public ownership as a special exception to the rule under a grandfather clause.

In 1976, Sullivan sought to acquire the publicly held shares of the Patriots through a merger of the club into a new Sullivan-owned company. Stockholders approved the transfer and the transaction was subsequently consummated, although some shareholders subsequently brought suit, challenging the sufficiency of the purchase price. After protracted litigation, the shareholders obtained a judgment requiring Sullivan to pay them a higher price for their shares. The Patriots then became a fully privately owned club.

Sullivan and his son, Chuck Sullivan, who owned the stadium where the Patriots played, began to experience financial difficulties and increasing debt burdens in the mid-1980s. The Sullivans decided that they needed to raise capital to alleviate their financial problems. After the Boston Celtics professional basketball franchise made a public offering of 40% of the team in December of 1986, the Sullivans decided to pursue a similar deal with the Patriots in order to raise cash to cover some of their debts.

On October 19, 1987, the Sullivans met with Stephens, Inc., a small investment banking firm in Little Rock, Arkansas. They discussed a debt financing deal whereby Stephens would loan the Sullivans $80 million dollars, with half going to the Patriots and the other half to Chuck Sullivan’s company which owned the Patriots’ stadium. The Patriots’ portion of the loan would be repaid out of the proceeds of the sale of 49% of the Patriots through the offering of public stock. Stephens agreed to look into the possibility of arranging the deal, but informed the Sulli-vans that they would first have to get NFL approval. Sullivan ultimately never obtained NFL approval and the deal with Stephens never progressed beyond some preliminary discussions.

At a meeting of the NFL owners on October 27, 1987, Sullivan raised his stock sale *1096 idea with the other owners and asked for a modification of the NFL’s policy against public ownership to allow for certain controlled sales of minority interests in NFL clubs. Alternatively, Sullivan requested a waiver from the public ownership policy for his contemplated public offering of the Patriots. Sullivan’s request was eventually tabled at this meeting. Discussions continued among the owners and, at one point, Sullivan counted 17 of the 21 owners needed for approval as being in favor of allowing him to make his public offering (seven owners were still undecided). Pete Rozelle, NFL Commissioner at the time, told Sullivans that he was not in favor of Sullivan’s proposals and that league approval was “very dubious.” Sullivan ultimately never asked for a vote on amending the ownership policy or on waiving the policy for the Patriots, and the NFL never held such a vote. Sullivan claims that he did not ask for a vote because it would have been futile.

In October of 1988, Sullivan sold the Patriots for approximately $83.7 million to KMS Patriots L.P. (“KMS”), a limited partnership owned by Victor Kiam and Francis Murray. Sullivan alleges that, absent the NFL’s public ownership policy, he would have been able to retain a majority share of a rapidly appreciating asset with a high potential for future profits. Instead, Sullivan asserts, he was forced to sell the Patriots at a depressed price to private buyers.

On May 16, 1991, Sullivan sued the NFL claiming that, among other things, the NFL had violated the Sherman Anti-trust Act, 15 U.S.C. §§ 1-2, by preventing him from selling 49% of the Patriots to the public in an equity offering. Sullivan alleged that, as a result, he was forced to sell the entire team to a private buyer at a fire sale price in order to pay off existing debts. Prior to trial, the district court dismissed Sullivan’s claim under § 2 of the Sherman Act along with various state law claims. After a trial on Sullivan’s claim under § 1 of the Sherman Act, the jury rendered a verdict for Sullivan in the amount of $38 million, which the judge later reduced through remittitur to $17 million. Pursuant to 15 U.S.C. § 15, which provides for treble damages for antitrust violations, the court entered a final judgment for Sullivan of $51 million.

II. ANALYSIS

The NFL has raised a number of issues on appeal concerning the application of § 1 of the Sherman Act to the facts of this case, which, according to the NFL, entitle it to judgment as a matter of law. We address these issues first to see if the present case should be dismissed, and we ultimately conclude that it should not. We next address the NFL’s allegations of trial error and we find that several of them require that we overturn the verdict in this case and order a new trial.

The first set of issues involves the district court’s denial of the NFL’s motions for judgment as a matter of law under Fed. R.Civ.P. 50. We review the court’s decision de novo, using the same stringent decisional standards that controlled the district court. Gallagher v. Wilton Enterprises, Inc., 962 F.2d 120, 125 (1st Cir.1992); Hendricks & Assocs., Inc. v. Daewoo Corp., 923 F.2d 209, 214 (1st Cir.1991). Under these standards, judgment for the NFL can only be ordered if the evidence, viewed in the fight most favorable to Sullivan, points so strongly and overwhelmingly in favor of the NFL, that a reasonable jury could not have arrived at a verdict for Sullivan. Gallagher, 962 F.2d at 124-25; Hendricks, 923 F.2d at 214.

III. ISSUES ALLEGEDLY REQUIRING JUDGMENT FOR THE NFL

A. Lack of Antitrust Injury

To establish an antitrust violation under § 1 of the Sherman Act, Sullivan must prove that the NFL’s public ownership policy is “in restraint of trade.” Monahan’s Marine, Inc. v. Boston Whaler, Inc., 866 F.2d 525, 526 (1st Cir.1989). Under antitrust law’s “rule of reason,” the NFL’s policy is in restraint of trade if the anticompetitive effects of the policy outweigh the policy’s legitimate business justifications. Id. at 526-27 (citing Business Electronics Corp. v. Sharp Electronics Corp., 485 U.S. 717, 723, 108 S.Ct. 1515, 1519, 99 L.Ed.2d 808 (1988)). Anticompetitive effects, more commonly re *1097 ferred to as “injury to competition” or “harm to the competitive process,” are usually measured by a reduction in output and an increase in prices in the relevant market. National Collegiate Athletic Ass’n v. Board of Regents of Univ. of Okla., 468 U.S. 85, 104-07, 104 S.Ct. 2948, 2962-63, 82 L.Ed.2d 70 (1984) (“Restrictions on price and output are the paradigmatic examples of restraints of trade”) (hereinafter “NCAA ”); Chicago Professional Sports Ltd. Partnership v. National Basketball Association, 961 F.2d 667, 670 (7th Cir.), cert. denied, — U.S. -, 113 S.Ct. 409, 121 L.Ed.2d 334 (1992). Injury to competition has also been described more generally in terms of decreased efficiency in the marketplace which negatively impacts consumers. Town of Concord v. Boston Edison Co., 915 F.2d 17, 21-22 (1st Cir.1990), cert. denied, 499 U.S. 931, 111 S.Ct. 1337, 113 L.Ed.2d 268 (1991); Interface Group, Inc. v. Massachusetts Port Auth., 816 F.2d 9, 10 (1st Cir.1987). Thus, an action harms the competitive process “when it obstructs the achievement of competition’s basic goals— lower prices, better products, and more efficient production methods.” Town of Concord, 915 F.2d at 22.

The jury determined in this case, via a special verdict form, that the relevant market is the “nationwide market for the sale and purchase of ownership interests in the National Football League member clubs, in general, and in the New England Patriots, in particular.” The jury went on to find that the NFL’s policy had an “actual harmful effect” on competition in this market.

The NFL argues on appeal that Sullivan has not established the existence of any injury to competition, and thus has not established a restraint of trade that can be attributed to the NFL’s ownership policy. The league’s attack is two-fold, asserting (1) that NFL clubs do not compete with each other for the sale of ownership interests in their teams so there exists no competition to be injured in the first place; and (2) Sullivan did not present sufficient evidence of injury to competition from which a reasonable jury could conclude that the NFL’s policy restrains trade. Although we agree with the NFL that conceptualizing the harm to competition in this case is rather difficult, precedent and deference to the jury verdict ultimately require us to reject the NFL’s challenge to the finding of injury to competition.

Critically, the NFL does not challenge on appeal the jury’s initial finding of the relevant market and no corresponding challenge was raised at trial. 1 As a result, the NFL faces an uphill battle in its attack on the presence of an injury to competition. Given the existence of a relevant market for ownership interests in NFL teams, it is reasonable to presume that a policy restricting the buying and selling of such ownership interests injures competition in that market. The NFL nevertheless maintains that NFL teams do not compete against each other for the sale of their ownership interests, even if we accept that a market exists for such ownership interests.

1. No Competition Subject to Injury as Matter of Law

The NFL correctly points out that member clubs must cooperate in a variety of ways, and may do so lawfully, in order to make the football league a success. See United States Football League v. National Football League, 842 F.2d 1335, 1372 (2d Cir.1988); Los Angeles Memorial Coliseum Comm’n v. National Football League, 726 F.2d 1381, 1391-92 (9th Cir.), cert. denied, 469 U.S. 990, 105 S.Ct. 397, 83 L.Ed.2d 331 (1984) (hereinafter “L.A Coliseum”), North American Soccer League v. National Football League, 670 F.2d 1249, 1251 (2d Cir.), cert. denied, 459 U.S. 1074, 103 S.Ct. 499, 74 *1098 L.Ed.2d 639 (1982) (hereinafter “NASL”). On the other hand, it is well established that NFL clubs also compete with each other, both on and off the field, for things like fan support, players, coaches, ticket sales, local broadcast revenues, and the sale of team paraphernalia. Mid-South Grizzlies v. National Football League, 720 F.2d 772, 786-87 (3d Cir.1983), cert. denied, 467 U.S. 1215, 104 S.Ct. 2657, 81 L.Ed.2d 364 (1984); L.A. Coliseum, 726 F.2d at 1390, 1393, 1395, 1397. The question of whether competition exists between NFL teams for sale of their ownership interests, such that the NFL’s ownership policy injures this competition, is ultimately a question of fact. The NFL would have us find, however, that, as a matter of law, NFL teams do not compete against each other for the sale of their ownership interests. We decline to make such a finding.

The NFL relies on a series of eases which allegedly stand for the “well established” rule that a professional sports league’s restrictions on who may join the league or acquire an interest in a member club do not give rise to a claim under the antitrust laws. Seattle Totems Hockey Club, Inc. v. National Hockey League, 783 F.2d 1347 (9th Cir.), cert. denied, 479 U.S. 932, 107 S.Ct. 405, 93 L.Ed.2d 357 (1986); Fishman v. Estate of Wirtz, 807 F.2d 520 (7th Cir.1986); Mid-South Grizzlies, 720 F.2d at 772; Levin v. National Basketball Ass’n, 385 F.Supp. 149 (S.D.N.Y.1974). These cases, all involving a professional sport’s league’s refusal to approve individual transfers of team ownership or the creation of new teams, do not stand for the broad proposition that no NFL ownership policy can injure competition. See, e.g., NASL, 670 F.2d at 1259-61 (finding that the NFL’s policy against cross-ownership of NFL teams and franchises in competing sports leagues, which also effectively barred certain owners who owned other sports franchises from purchasing NFL teams, injured competition between the NFL and competing sports leagues and thus violated § 1 of the Sherman Act).

None of the cases cited by the NFL considered the particular relevant market that was found by the jury in this case or a league policy against public ownership. Seattle Totems and Mid-South Grizzlies considered potential mier-league competition when a sports league rejected plaintiffs’ applications for new league franchises. Seattle Totems, 783 F.2d at 1349-50; Mid-South Grizzlies, 720 F.2d at 785-86. Those decisions found no injury to competition because the plaintiffs were not competing with the defendant sports leagues, but rather, were seeking to join those leagues. Seattle Totems, 783 F.2d at 1350; Mid-South Grizzlies, 720 F.2d at 785-86. Mid-South Grizzlies left open the possibility that potential ráíra-league competition between NFL football clubs could be harmed by the NFL’s action, but found that the plaintiff in that case had not presented sufficient evidence of harm to such competition. Mid-South Grizzlies, 720 F.2d at 786-87.

The Fishman and Levin cases concerned the National Basketball Association’s (“N.B.A.”) rejection of plaintiffs’ attempts to buy an existing team. Fishman, 807 F.2d at 525-31; Levin, 385 F.Supp. at 150-51. Those cases also based their finding that there was no injury to competition on the fact that the plaintiffs were seeking to join with, rather than compete against, the N.B.A. Fishman, 807 F.2d at 544; Levin, 385 F.Supp. at 152. Neither ease considered whether competition between teams for investment capital was injured. As pointed out in Piazza v. Major League Baseball, 831 F.Supp. 420 (E.D.Pa.1993), Fishman explicitly recognized the potential for competition in the market for ownership of teams, although the plaintiff had failed to raise the issue, and Levin simply presumed, incorrectly, that there could never be any competition among league members. Piazza, 831 F.Supp. at 430-31 & n. 16 (citing Fishman, 807 F.2d at 532 n. 9; and Levin, 385 F.Supp. at 152).

The important distinction to make between the cases cited by the NFL and the present case is that here Sullivan alleges that the NFL’s policy against public ownership generally restricts competition between clubs for the sale of their ownership interests, whereas in the aforementioned cases, a league’s refusal to approve a given sale transaction or a new team merely prevented particular outsiders from joining the league, but did not *1099 limit competition between the teams themselves. To put it another way, the NFL’s public ownership policy allegedly does not merely prevent the replacement of one club owner with another—an action having little evident effect on competition—it compromises the entire process by which competition for club ownership occurs. 2

We take a moment to briefly address a related argument raised by the NFL to the effect that NFL clubs are unable to conspire with each other under § 1 of the Sherman Act because they function as a single enterprise in relation to the league’s public ownership policy. The NFL asserts that the Supreme Court’s holding in Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 104 S.Ct. 2731, 81 L.Ed.2d 628 (1984), controls the facts of this case and overturns prior caselaw holding that NFL clubs do not constitute a single enterprise but rather, are separate entities which were capable of conspiring with each other under § 1. See L.A. Coliseum, 726 F.2d at 1387-90; NASL, 670 F.2d at 1256-58.

We do not agree that Copperweld, which found a corporation and its wholly owned subsidiary to be a single enterprise for purposes of § 1, Copperweld, 467 U.S. at 771, 104 S.Ct. at 2741, applies to the facts of this case or affects the prior precedent concerning the NFL. See McNeil v. National Football League, 790 F.Supp. 871, 879-80 (D.Minn.1992) (holding that Copperweld did not apply to the NFL and its member clubs and finding the clubs to be separate entities capable of conspiring together under § 1). Copperweld’s holding turned on the fact that the subsidiary of a corporation, although legally distinct from the corporation itself, “pursue[d] the common interests of the whole rather than interests separate from those of the corporation itself.” Copperweld, 467 U.S. at 770, 104 S.Ct. at 2741. As emphasized in City of Mt. Pleasant, Iowa v. Associated Elec. Co-op., Inc., 838 F.2d 268 (8th Cir.1988), upon which the NFL relies for the application of Copperweld to this case, the critical inquiry is whether the alleged antitrust conspirators have a “unity of interests” or whether, instead, “any of the defendants has pursued interests diverse from those of the cooperative itself.” Id. at 274-77 (defining “diverse” as “interests which tend to show that any two of the defendants are, or have been, actual or potential competitors”). As we have already noted, NFL member clubs compete in several ways off the field, which itself tends to show that the teams pursue diverse interests and thus are not a single enterprise under § 1.

Ultimately, the NFL’s Copperweld challenge is subsumed under the question of whether or not the evidence can support a finding that NFL teams compete against each other for the sale of their ownership interests. Proof of such competition defeats both the NFL’s challenge to the existence of an injury to competition and the NFL’s Cop-perweld argument as well. Insufficient proof of such competition would require a judgment in favor of the NFL anyway, regardless of the implications under Copperweld. As we discuss below, the jury’s finding that there exists competition between teams for the sale of ownership interests was based on sufficient evidence.

2. Insufficient Evidence of Harm, to Competition

The NFL contends that Sullivan did not present sufficient evidence concerning: (1) the existence of competition between NFL clubs for the sale of ownership interests, or (2) a decrease in output, an increase in prices, a detrimental effect on efficiency or other incidents of harm to competition in the relevant market, from which a reasonable jury could conclude that the NFL’s policy injured competition. Although we agree that *1100 the evidence of all these factors is rather thin, we disagree that the evidence is too thin to support a jury verdict in Sullivan’s favor.

With respect to evidence of the existence of competition for the sale of ownership interests, one of Sullivan’s experts, Professor Roger Noll, testified that “one of the ways in which the NFL exercises monopoly power in the market for the franchises and ownership is by excluding certain people from owning all or part — any type part of an NFL franchise.” Dr. Noll explained that this “enables a group of owners, in this case, you only need eight owners, to exclude from the League and from competing with them, people who might be more effective competitors than they are.” The record also contains statements from several NFL owners which could reasonably be interpreted as expressions of concern about their ability to compete with other teams in the market for investment capital in general, and for the sale of ownership interests in particular. For example, Arthur Rooney II of the Pittsburgh Steelers stated in a letter that he did not “believe that the individually or family owned teams will be able to compete with the consolidated groups.” Ralph Wilson of the Buffalo Bills stated that big corporations should not own teams because it gives them an “unfair competitive advantage” over other teams since corporations will funnel money into the team and make it “more competitive” than the other franchises. Former NFL Commissioner Pete Rozelle admitted that similar sentiments had been expressed by NFL members.

Although it is not precisely clear that the “competition” about which Noll, Rooney, and Wilson were discussing is the same competition at issue here — that is competition for the sale of ownership interests— a jury could reasonably interpret these statements as expressing a belief that the competition exists between teams for the sale of ownership interests. The statements of the two NFL owners imply that greater access to capital for all teams will put increased pressure on some teams to compete with others for that capital, and all the statements reveal that the ownership rules, particularly the rule against public ownership, is the main obstacle preventing such access. The fact that ownership by “consolidated groups” is not necessarily the same as public ownership does not affect the conclusion that teams face competitive pressure in selling their ownership interests generally to whoever might buy them. We also note that evidence of actual, present competition is not necessary as long as the evidence shows that the potential for competition exists. See L.A. Coliseum, 726 F.2d at 1394 (discussing significance of potential competition, especially where challenged policy limits such competition so that it is not evident in practice). It would be difficult indeed to provide direct evidence of competition when the NFL effectively prohibits it.

The NFL focuses on the fact that Professor Noll testified that many of the purchasers of Patriots’ stock would be New England sports fans and others in the New England area. The NFL points out that other NFL teams would not compete with the Patriots for the sale of stock to their own fans. This argument slightly distorts Professor Noll’s testimony. Professor Noll stated that local souvenir buyers would be one portion of the market for Patriots stock. Professor Noll also testified several times that other investors would buy Patriots stock as well, for investment purposes. Noll’s point was that the souvenir buyers would serve to bid up the price of the stock above what the price would normally be if the Patriots were a regular company. His testimony did not preclude a finding that NFL teams compete against each other for investment capital via the sale of ownership interests.

The record also contains sufficient evidence of the normal incidents of injury to competition from the NFL’s policy — reduced output, increased prices, and reduced efficiency — to support the jury’s verdict. As Dr. Noll pointed out in his testimony, the NFL’s policy “excludes individuals ... who might want to own a share of stock in a professional football team.” Several NFL officials themselves admitted that the policy restricts the market for investment capital among NFL teams. There is thus little dispute that the NFL’s ownership policy reduces the available output of ownership interests.

*1101 The NFL is correct that, in one sense, the overall pool of potential output is fixed because there are only 28 NFL teams and, although their value may fluctuate, the quantity of their ownership interests cannot. However, the NFL’s public ownership policy completely wipes out a certain type of ownership interest—public ownership of stock. By restricting output in one form of ownership, the NFL is thereby reducing the output of ownership interests overall. In other words, the NFL is literally restricting the output of a product—a share in an NFL team.

There was considerable testimony concerning the price effects of the NFL policy. Both of Sullivan’s experts testified that the policy depressed the price of ownership interests in NFL teams because NFL franchises would normally command a premium on the public market relative to their value in the private market, which is all that the league currently permits. Professor Noll testified that fan loyalty would push up the price of ownership interests if sales to the public were allowed. Even former Commissioner Pete Bozelle acknowledged that “it was pointed out, with justification, it has been over the years, that [the ownership policy] does restrict your market and, very likely, the price you could get for one of our franchises if you wanted to sell it, because you are eliminating a very broad market-And they have said that there is a depression on the price they could get for their franchise.”

The NFL points out that the alleged effect of its ownership policy is to reduce prices of NFL team ownership interests, rather than to raise prices which is normally the measure of an injury to competition. E.g., Town of Concord, 915 F.2d at 22. We acknowledge that it is not clear whether, absent some sort of dumping or predatory pricing, see, e.g., Monahan’s Marine, Inc. v. Boston Whaler, Inc., 866 F.2d 525, 527 (1st Cir.1989), a decrease in prices can indicate injury to competition in a relevant market. The Supreme Court has emphasized, however, that overall consumer preferences in setting output and prices is more important than higher prices and lower output, per se, in determining whether there has been an injury to competition. NCAA, 468 U.S. at 107, 104 S.Ct. at 2963. In this case, regardless of the exact price effects of the NFL’s policy, the overall market effects of the policy are plainly unresponsive to consumer demand for ownership interests in NFL teams. Dr. Noll testified that fans are interested in buying shares in NFL teams and that the NFL’s policy deprives fans of this product. Moreover, evidence was presented concerning the public offering of the Boston Celtics professional basketball team which demonstrated, according to some of the testimony, fan interest in buying ownership of professional sports teams. Thus, a jury could conclude that the NFL’s policy injured competition by making the relevant market “unresponsive to consumer preference.” Id. 3

As for overall efficiency of production in the relevant market, 4 Sullivan’s experts testified that the NFL’s policy hindered efficiency gains, and that allowing public ownership would make for better football teams. Professor Noll stated that the NFL’s public ownership policy prevented individuals who *1102 might be “more efficient and much better at running a professional football team” from owning teams. Dr. Noll also stated that publicly owned NFL teams would be better managed, and produce higher quality entertainment for the fans. Noll testified that the ownership rule excluded certain types of management structures which would likely be more efficient in running the teams, resulting in higher franchise values. One NFL owner, Lamar Hunt, acknowledged that increased access to capital can improve a team’s operations and performance. A memorandum prepared by an NFL staff member stated that changes to the NFL’s public ownership policy could contribute to each NFL team’s own financial strength and viability, which in turn would benefit the entire NFL because the league has a strong interest in having strong, viable teams.

The NFL presented a large amount of evidence to the contrary and now claims on appeal that Sullivan’s position was based on nothing more than sheer speculation. We have reviewed the record, however, and we cannot say that the evidence was so overwhelming that no reasonable jury could find against the NFL and in favor of Sullivan. We therefore refuse to enter judgment in favor of the NFL as a matter of law.

B. Ancillary Benefits

The NFL next argues that even if its public ownership policy injures competition in a relevant market, it should be upheld as ancillary to the legitimate joint activity that is “NFL football” and thus not violative of the Sherman Act. We take no issue with the proposition that certain joint ventures enable separate business entities to combine their skills and resources in pursuit of a common goal that cannot be effectively pursued by the venturers acting alone. See, e.g., Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 99 S.Ct. 1551, 60 L.Ed.2d 1 (1979). We also do not dispute that a “restraint” that is ancillary to the functioning of such a joint activity — i.e. one that is required to make the joint activity more efficient — does not necessarily violate the antitrust laws. Broadcast Music, 441 U.S. at 23-25, 99 S.Ct. at 1564-1565; Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210, 223-24 (D.C.Cir.1986), cert. denied, 479 U.S. 1033, 107 S.Ct. 880, 93 L.Ed.2d 834 (1987); see also Northwest Wholesale Stationers, Inc. v. Pacific Stationery & Printing Co., 4