OPINION OF THE COURT

BECKER, Circuit Judge.

This is an antitrust case under Section 1 of the Sherman Act. Plaintiff InterVest Financial Services, Inc. ("InterVest") created an electronic trading platform where its subscribers could trade bonds and other forms of fixed income securities, and entered into a contract with Bloomberg, L.P. ("Bloomberg") to place its system on Bloomberg's information network, which is widely used in the financial world. According to InterVest, its trading system sought to revolutionize the bond market by allowing investors access to real-time pricing information and lower transaction costs per trade. However, InterVest's relationship with Bloomberg was unsuccessful, and Bloomberg terminated its contract with InterVest only 14 months after InterVest went live on the Bloomberg network. Alleging that S.G. Cowen Securities Corp. ("Cowen") and certain other broker-dealers in the bond market pressured Bloomberg to dump InterVest from its system because the broker-dealers were threatened by the prospect of InterVest undercutting the profits they earned by exploiting their monopoly over bond pricing information, InterVest brought suit under the Sherman Act against the broker-dealers and Bloomberg in the District Court for the Eastern District of Pennsylvania. InterVest also alleged that the broker-dealers tortiously interfered with its contract with Bloomberg.

All of the defendants settled with InterVest, except for Cowen. After the completion of discovery, Cowen moved for summary judgment, which the District Court granted. In reviewing Cowen's motion for summary judgment, the District Court applied the special standard for Sherman Act cases articulated by the Supreme Court in Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984), and Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). In those cases, the Supreme Court explained that "[c]onduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy." Matsushita, 475 U.S. at 588, 106 S.Ct. 1348. Therefore in a conspiracy case, a nonmoving plaintiff "must present evidence that `tends to exclude the possibility' that the alleged conspirators acted independently" in order to survive a motion for summary judgment. Id. (quoting Monsanto, 465 U.S. at 764, 104 S.Ct. 1464).

Cowen maintains that the Monsanto/Matsushita standard was properly applied. InterVest argues that the Court should not have used this standard because Cowen's participation as a broker-dealer in the bond market — a market in which these firms controlled pricing information on bonds and could therefore charge high spreads (or markups) on transactions — was direct evidence of a conspiracy. We disagree. As we will explain, the lack of price transparency in the bond market benefits investors who wish to transact anonymously and thus reduce the market impact of their trades; furthermore, broker-dealers provide the needed liquidity for investors who deal with thinly traded bonds. And there is nothing in the structure of the bond market that prevents the entry of new broker-dealers. We do not believe that the entire bond market, which includes thousands of broker-dealers trading various types of securities, can fairly be described as a conspiracy.

Moreover, the cases require that direct evidence of an illegal agreement be established with much greater clarity. And as the District Court concluded, Cowen's desire to make money as a broker-dealer in the bond market "is, in and of itself, perfectly rational and legal," Intervest Financial Services, Inc. v. S.G. Cowen Securities Corp., 206 F.Supp.2d 702, 717 (E.D.Pa. 2002), not direct evidence of an antitrust violation. Because the evidence InterVest submits is at most ambiguous regarding Cowen's participation in a conspiracy to injure InterVest, we believe that the Court correctly applied the summary judgment standard articulated by the Supreme Court in Monsanto and Matsushita.

In order to survive Cowen's motion, InterVest must present evidence that tends to exclude the possibility that Cowen acted independently and leads to the reasonable inference that Cowen engaged in an illegal conspiracy to keep InterVest out of the bond market. InterVest cannot meet this standard. There is no evidence in the record that Cowen communicated with other broker-dealers regarding InterVest. Although InterVest produces evidence tending to show that Bloomberg might have severed its relationship with InterVest at least in part due to pressure from broker-dealers, InterVest does not present evidence indicating that Cowen threatened Bloomberg into doing so or that there was an agreement or conspiracy between Cowen and Bloomberg to harm InterVest. Moreover, there is ample evidence in the record suggesting that Cowen's decision not to deal with InterVest was made independently in light of (1) Cowen's desire to maintain its business, which depended on its ability to obtain the profits earned from the spreads, or markups, on the transactions it brokered in the bond market, and the confidentiality between the brokerage and dealer aspects of the business; (2) InterVest's unproven technology; and (3) investors' apparent lack of interest in InterVest's system as evidenced by the few transactions the company conducted. Given this backdrop and the lack of evidence tending to show that Cowen conspired with other broker-dealers or Bloomberg rather than acting individually, we will affirm the District Court's order granting Cowen's motion for summary judgment on the antitrust claim.

Finally, we also will affirm the District Court's order granting summary judgment in favor of Cowen on InterVest's tortious interference with contract claim because InterVest can only point to an alleged complaint by Cowen to Bloomberg to support its claim, an action which we do not believe amounts to an improper interference under § 767 of the Restatement (Second) of Torts.

I. Background A. The Bond Market and the Role of Broker-Dealers; Positive and Negative Aspects

We begin with a summary of the operation of the bond market as it functioned during the relevant period (1993-1998). Intervest sought to create an electronic exchange in the secondary bond market, where bonds are bought and sold among institutional investors and broker-dealers after they are issued. All U.S. government and municipal bonds, as well as the vast preponderance of corporate bonds, are traded over-the-counter ("OTC") through direct trades between a buyer and seller. See Robert Zipf, How the Bond Market Works 139 (1997). A very small percentage of bonds are traded on established exchanges, such as the New York Stock Exchange.

In the vernacular of the bond market, the "buy-side" consists of the customers, mostly institutional investors, such as mutual funds, insurance companies, and pension funds. The "sell-side" of the market is made up of thousands of broker-dealers, including firms like Cowen. Until very recently, broker-dealers transacted business primarily over the telephone; electronic exchanges were not widely introduced until the late 1990s. "Sell-side" firms are called "broker-dealers" because they serve two main functions: (1) as brokers, by acting as an agent for customers who wish to buy and sell bonds; and (2) as dealers, by serving as principal investors through the trading of bonds in their own inventories. Id. at 140. Broker-dealers may also act as "inter-dealer brokers" when they serve as an agent in a transaction between two other broker-dealers.

Broker-dealers receive different forms of compensation depending on which role they are undertaking in the transaction; however, a firm may only act in one role in any single transaction. When serving as a broker (or an inter-dealer broker), the firm will receive a commission for executing the customer's order. By contrast, when acting as a dealer, the firm will charge a markup on sales and a markdown on purchases. Id. This markup or markdown is the dealer's spread (or profit), the amount of which is rarely disclosed to the customer. Id. at 145.

The OTC bond market functions very differently from the stock or equities markets. The stock markets may be characterized as "auction" markets, where the price of the stock is set by the highest bid or the lowest offer. The bond market, however, is a "negotiated" market in which the transaction price is set individually by the broker-dealer and the investor. Id. at 140. Importantly, this characteristic means that the bond market lacks price transparency, i.e., the ability of the investor to know the market price of the bond for which he is bidding, particularly in the trading of corporate bonds. See Speech of SEC Chairman Arthur Levitt, Sep. 9, 1998, 1998 WL 614902 at *2. Whereas an investor in the equities markets has ready access to current price information on stocks, the same is not available in the bond market. As Chairman Levitt put it. "The sad truth is that investors in the corporate bond market do not enjoy the same access to information as a car buyer or a home buyer, or, dare I say, a fruit buyer." Id. at *1.

This difference in relative price transparency between the bond market and other markets is due to several unique characteristics of the bond market, three of which are particularly significant. First, the bond market is dominated by a relatively small number of broker-dealers who serve as market makers for particular types of bonds. These market makers maintain a large inventory of bonds, which they trade with customers as dealers or with other firms as inter-broker dealers. Zipf, at 140. Market makers take advantage of their control over information on bond prices, availability, and market activity, to charge a larger spread than would be possible in a market with greater price transparency.

A New York Times article illustrated how a professional investor might interact with a market maker in order to complete a transaction in the corporate bond market:

To get a market price, [a professional investor] must conduct an unwieldy and imprecise shopping expedition among a dozen or more dealers, or rely on what one dealer tells them — or, for less active issues, fall back on what may be little more than a wild guess.

Robert B. Hershey, Jr., Trading In Bonds On Line, At Last, The New York Times, June 27, 1999, at Cl. Yet, broker-dealers also need to be resourceful in order to make money in the bond market. As one book on the subject explained:

For all bond traders . . . the best source of information consists of their contacts in the business. Who among other firms' traders has a certain type of bond in inventory? Which of them owes a favor, and to whom? Who quoted a favorable price on a certain bond just this morning? Who is trying to sell a weak issue? And so on. Information like this is so specialized and often so fleeting that not even a highly automated quotation system like NASDAQ can capture it. This is the kind of information that is gathered during the course of countless phone calls and used at the precise time of its greatest effectiveness.

Zipf, at 143. Thus, broker-dealers provide unique and specialized analysis to their customers, a service whose cost is partly reflected in the spread.

Second, the number and complexity of bonds means that they are much more thinly traded than equities. Broker-dealers provide liquidity to the market by relieving investors of the expense and difficulty of trying to find another buyer or seller for a bond. They do this by serving as an intermediary; they may keep bonds in their inventory until a buyer arrives. Broker-dealers can also buy a large block of bonds from a single investor, disaggregate the package, and then sell smaller blocks to investors as demand arises. By creating a market for the investor, broker-dealers absorb the risk associated with the transaction, i.e., the prospect that they cannot sell the bond for more than its value when they bought it from the investor. In the stock market, by contrast, brokers merely execute the transaction on an exchange for a fee, and do not accept any market risk associated with this service.

Third, investors enjoy anonymity in their transactions with broker-dealers. By not advertising publicly a selling offer, an investor may gain from a dealer's ability to gain a higher price for the bond through its own contacts in the market. This is even more important when an investor needs to sell a large amount of a certain bond. If the market were aware of the investor's trading pattern, the value of the bond would sink, thus harming the investor. The anonymity provided by broker-dealers reduces the impact of transactions on the market.

To summarize, the lack of price transparency in the bond market has both positive and negative aspects. On the plus side, investors benefit from anonymity in their transactions and the specialized information and liquidity broker-dealers provide. Yet on the negative side, investors are unable to access a significant amount of information on the market and thus are forced to pay higher spreads than would exist if there was greater price transparency.

B. InterVest's Business Model

Beginning in 1991, InterVest attempted to start an online bond trading system that would allow investors and broker-dealers to trade bonds with each other electronically and anonymously. InterVest envisioned a system that would display all current bids and offers as well as the actual price at which trades had been executed, thus providing the possibility of greater price transparency for its users. This type of electronic trading system is known as a "cross matching system," because it automatically matches up subscribers, either investors or broker-dealers, who post buy and sell orders with other subscribers who can fill the requests with a "contra-order." See BMA, Review of Electronic Transaction Systems: Cross-Matching Systems, available at http://www.bondmarkets.com/research/ecommerce/crossmatch.shtml.

InterVest claimed that its system would increase market efficiency and price transparency as well as eliminate the multiple intermediary fees (or spreads) incurred by investors. Unlike most broker-dealers, InterVest charged a fixed commission for transactions on its system, which would be less than the amount investors would otherwise pay a broker-dealer in the form of the transaction's spread. InterVest's business model relied on the prospect of attracting a sufficient number of subscribers, with their inventory of bonds to trade, in order to maintain an active exchange. This is because in an exchange-type system, it is the participants themselves who provide liquidity, as opposed to a pure dealer system, in which the dealers provide liquidity by acting as market intermediaries.

Cowen submits that InterVest's software, the InterVest 2.0 trading platform, was incapable of supporting the trading volume that InterVest anticipated in its business plan. Cowen supplied strong evidence of this through the report of a computer programming expert who analyzed InterVest 2.0, and concluded that the trading platform was plagued by numerous design flaws, including a risk of data loss and inadequate system support. However, the expert did not observe this software or the next version in development, InterVest 3.0, in operation. In its brief, InterVest rejects this expert's conclusions, but fails to point to any specific evidence in the record for support; rather, the company relies on conclusory statements that InterVest 3.0 was "state of the art" and would remedy the problems in InterVest 2.0. The District Court did not address this issue, however, because it determined that InterVest could not prove that Cowen had participated in concerted action in violation of the Sherman Act, and thus the Court did not need to consider whether Cowen's actions caused InterVest any injury. Intervest Financial Services, Inc. v. S.G. Cowen Securities Corp., 206 F.Supp.2d 702, 720 (E.D.Pa. 2002).

C. InterVest's Early Contact with the Broker-Dealers

InterVest was founded by Larry Fondren, who beginning in 1991 had started a number of other companies that sought to provide electronic trading services to the bond market. One such company, the Inter-Dealer Consortium, offered an electronic trading system similar to that ultimately developed for InterVest, except that its subscribers were restricted to broker-dealers. When Fondren approached broker-dealers seeking their participation in the Inter-Dealer Consortium, he was universally rebuffed.

For example, in his deposition, Fondren stated that a vice president of Merrill Lynch informed him that he hoped the project would "crash and burn." Fondren also testified that an employee of Salomon Brothers ("Salomon"), which is now Salomon Smith Barney, was concerned about joining the Inter-Dealer Consortium because Salomon was aware of Fondren's idea for InterVest, a company that might threaten the broker-dealers' market power. Specifically, Fondren alleged that Salomon characterized InterVest as "not playing by the rules" because it would allow investors to see the prices at which bonds were trading. Fondren received a similar response from Bear Stearns, which "indicated that [InterVest] was contrary to the way the market worked" because broker-dealers "prefer to have the market not see prices at which transactions are occurring." Fondren stated that Bear Stearns intended to make its concerns known to other broker-dealers. [ Id.]

InterVest also points to an ACT Note, a record stored in an electronic database system, from a meeting between Fondren and William Matthews, the executive at Cowen in charge of the bond desk, on June 11, 1993. The Note states that Matthews "showed concern when it was explained that institutional investors would have equal and anonymous access to all pricing and order execution" because, quoting Matthews, "this is just not the way the street works." Matthews also opined that because Cowen acts both as a broker-dealer and as an inter-dealer broker, it would be an inappropriate breach of confidentiality for the firm to share the prices of its bond transactions with InterVest. [ Id.] InterVest challenges this contention.

D. InterVest's Relationship with Bloomberg and the Response of the Broker-Dealers

InterVest achieved a potential breakthrough in 1995 when it entered into a contract with Bloomberg to provide its system on Bloomberg's network of terminals. Bloomberg is a staple in financial firms, who rely on the myriad information services accessible through "Bloomberg terminals." Firms on both the buy-side and sell-side of the bond market, as well as those involved in other financial markets, subscribe to the Bloomberg service. Between July 1995 and July 1998, the number of Bloomberg terminals installed in financial firms doubled from 50,000 to 100,000. Importantly, Bloomberg is a "two-sided" business, meaning that it must attract subscribers from both the buy-side and the sell-side of the financial markets in order to remain successful. Its revenues come primarily from subscription fees for the general use of the terminals and its information outlets, not surcharges from the use of special services, such as InterVest. Although Bloomberg is known mostly for its financial information dissemination business, in the 1990s it began to offer subscribers the ability to execute transactions in various markets through proprietary services as well as those offered by third-parties. InterVest was one of these third-party providers.

The relationship between Bloomberg and InterVest quickly soured. After Bloomberg released an advertisement stating that InterVest would soon be available on its network, Bloomberg received several complaints from broker-dealers. For example, an employee at ABN Amro Securities, a broker-dealer not named in this suit, emailed Bloomberg, stating his discomfort with the "concept of InterVest" because it is "a service that competes with [Bloomberg's] subscribers," and that he was sure that he was "not the only one that feels [that] way."

In response to these expressions of concern, Charles Garcia, Bloomberg's "relationship manager" for InterVest, called Fondren on December 27, 1995, and notified him that Bloomberg was placing a moratorium on its efforts to integrate the InterVest system on the Bloomberg network. Fondren testified that Garcia informed him that the moratorium was Bloomberg's response to the complaints from several broker-dealers. [ Id.] In response, InterVest's counsel threatened Bloomberg with "legal recourse." Bloomberg then lifted the moratorium and stated that it would support InterVest for a minimum of one year after the system went live. In December of 1996, InterVest and Bloomberg announced the inauguration of the InterVest system on the Bloomberg network at a press conference at Bloomberg's headquarters. An article appeared in the New York Times the next day, previewing InterVest's capabilities and its ambition to change the operation of the bond market. Fondren also gave an interview to a Bloomberg reporter that was instantly published through the "Bloomberg Forum" on the Bloomberg network.

The day after InterVest went live on Bloomberg, Garcia spoke with William Matthews from Cowen, who mentioned that he had seen the New York Times article on InterVest. That same day, Garcia called Fondren and notified him that Bloomberg was placing a moratorium on the development of a municipal bond trading platform for InterVest on its network. According to Fondren, Garcia told him that Bloomberg was "getting tremendous heat from Inter-Dealer Brokers" and specifically named Cowen as well as a municipal bond broker-dealer in Tokyo. Either on that day or shortly thereafter, Bloomberg removed the interview with Fondren that it had placed on the "Bloomberg Forum."

Fondren also testified that Bloomberg limited the functionality of InterVest on its network by not fixing a service InterVest sought to provide that would allow subscribers to benchmark the price of bids and offers of corporate bonds against U.S. Treasury bonds. This feature, sometimes referred to as "Treasury benchmarking," is the standard method of pricing corporate bonds. Neither did InterVest appear on Bloomberg's menu of "Electronic Trading Systems" on the network.

In September 1997, Bloomberg began to wind down its relationship with InterVest. Lou Eccleston, Bloomberg's Director of Sales, barred Bloomberg employees from speaking to the press about InterVest. On December 11, 1997, a subordinate of Eccleston emailed him with the question, "when does our contract with InterVest let us throw them off the system . . . it would be nice to close [InterVest] down." On February 5, 1998, about 14 months after the launch date, Bloomberg officially terminated its agreement with InterVest. During the time InterVest was available on Bloomberg, only 10 bond trades were executed, all of which had been pre-arranged over the telephone.

E. Bloomberg and BondNet (Another Electronic Exchange)

In order to establish a pattern of Bloomberg's hostility to electronic exchanges like InterVest that would provide pricing information to investors on the buy-side, InterVest points to the experience of BondNet, another electronic trading system that had a relationship with Bloomberg. Originally, BondNet operated as a trading platform for inter-dealer brokers, and did not provide access to pricing information to investors on the buy-side of the bond market. In 1997, BondNet was acquired by the Bank of New York ("BONY"), and with its easy access to BONY's buy-side customers, BondNet contemplated opening its system to investors, thus essentially following InterVest's business model.

Bloomberg imposed similar restrictions on BondNet's functionality as it did on InterVest. This is reflected in an internal memorandum by an employee of BONY stating that in mid-December 1997, "Bloomberg advised us . . . that they wanted to stop adding any `Buy side' customers because of pressure from Wall Street."

F. Other Evidence of InterVest's Contact With Cowen

Aside from Garcia's statement that Bloomberg received criticism from Cowen for placing InterVest on its network, InterVest points to several other pieces of evidence that it asserts support its theory that Cowen participated in a conspiracy to keep InterVest out of the bond market. While InterVest was preparing to appear on the Bloomberg network, Fondren and other InterVest employees met with representatives of Cowen to urge them to use InterVest's system. On April 10, 1996, Fondren met with the two sons of Joseph Cohen, Cowen's chairman at the time. They informed Fondren that Cowen was opposed to showing prices to investors on the buy-side. They indicated, however, that if InterVest limited itself to broker-dealers, then Cowen would participate and perhaps even invest in the operation.

On December 2, 1996, just a week before InterVest went live on Bloomberg, Fondren met with Matthews, the head of Cowen's bond desk, and Joseph Cohen himself. Notes written that day by an unidentified Cowen employee stated, "Dealers want present structure not price disclosure," and asked, "What about Cowen brokers and spreads?" Fondren testified that Matthews informed him that InterVest's business plan was not "playing by the rules." Fondren also recorded in an ACT Note that Cowen indicated that InterVest could make more money if it did not attempt to "break the spreads."

According to his ACT Notes, Fondren met again with Matthews and Tom Evans, another Cowen executive, on May 29, 1997. Fondren recorded that Evans initiated the meeting to inquire whether InterVest might assist Cowen in developing its own bond trading exchange over the internet. Matthews, however, seemed less interested in dealing with InterVest and told Fondren that doing business with InterVest was "viewed by the street as `unhealthy.'" When Fondren responded by charging that locking InterVest out of the bond market was anti-competitive and could lead to a lawsuit, Matthews allegedly laughed and said that even if Cowen was forced to pay $10 or $20 million, "it would be just a cost of doing business."

Finally, InterVest cites deposition testimony from Joseph Cohen that it submits illustrates Cowen's motives. Cowen operated both as a broker-dealer and as an inter-dealer broker. Thus, its relationship with other broker-dealers was particularly critical. Cohen explained that Cowen could not share with InterVest the prices of the bonds Cowen was trading and breach the anonymity of its clients' transactions, because other broker-dealers would not use Cowen's inter-dealer services. Notably, Cohen also said that the same boycott would occur if Cowen let its dealers see the types of bonds and their prices that its interdealer brokers were trading.

G. The Present Suit; the District Court's Opinion

The underlying litigation originated when Bloomberg filed suit against InterVest on June 25, 1998, seeking the recovery of approximately $100,000 plus interest for unpaid advertising expenses (the "Bloomberg Action"). InterVest counterclaimed and alleged that Bloomberg had participated in an antitrust conspiracy involving several broker-dealers, including Cowen. InterVest then filed a second lawsuit on November 3, 1999, against Cowen and the following broker-dealers: Bear Stearns, Co., Inc.; Cantor Fitzgerald Securities and several other Cantor Fitzgerald entities; Deutsche Bank Securities Corp.; Liberty Brokerage Securities, Inc. and several other Liberty entities; Merrill Lynch Co., Inc.; J.P. Morgan Securities, Inc.; and Salomon Smith Barney, Inc. (the "Broker-dealer Action"). InterVest charged the broker-dealers with a violation of section 1 of the Sherman Act, 15 U.S.C. § 1, participation in a restraint of trade in violation of Pennsylvania law, and tortious interference with the contractual relationship between Bloomberg and InterVest.

The two cases were consolidated for discovery purposes only. In the meantime, InterVest settled with Bloomberg, and the District Court dismissed the Bloomberg Action with prejudice. InterVest also settled with all of the broker-dealers in the Broker-dealer Action except for Cowen. Thus, only Cowen moved for summary judgment, and the Court analyzed InterVest's claims as pertaining to Cowen as the sole remaining defendant.

The District Court examined the evidence summarized above. While the Court did not believe that any individual piece of evidence constituted direct evidence that Cowen engaged in an unlawful conspiracy, the Court looked at the evidence as a whole and concluded that:

a rational finder of fact might draw the series of inferences necessary to find that Cowen participated in a conspiracy with Bloomberg and other broker-dealers in order to maintain the existing closed bond trading system. The establishment and perpetuation of a closed system, the allusions by various broker-dealers, including defendant Cowen, to the `rules,' the universal refusal to deal with InterVest, the complaints to Bloomberg, and Bloomberg's subsequent resistance to buy side and transparent bond trading, all support an inference of conspiracy.

Intervest, 206 F.Supp.2d at 716.

Nevertheless, the Court determined that InterVest could not survive summary judgment because "all of this evidence supports equally well the inference that Cowen, other broker-dealers, and Bloomberg acted independently and rationally in light of their own interests and consumer demand." Id. at 716-17. The Court opined that "Cowen's desire to perpetuate a system in which it earned significant profits is, in and of itself, perfectly rational and legal," thus rendering Cowen's behavior ambiguous as to whether it participated in an illegal conspiracy. Id. at 717. The Court therefore concluded that "[b]ecause [InterVest] has presented only circumstantial evidence of illegal conspiracy and the conduct of [Cowen] is ambiguous, [InterVest] has not offered sufficient evidence of concerted action to survive Cowen's motion for summary judgment." Id. at 717. Since InterVest could not prove concerted action, the Court did not address Cowen's allegation that InterVest could not show antitrust injury. Intervest, 206 F.Supp.2d at 721. The Court also dismissed InterVest's tortious interference with contract claim because it concluded that Cowen's conduct was privileged. Id. at 721-23.

The District Court had jurisdiction pursuant to 15 U.S.C. § 15(a) and 28 U.S.C. §§ 1331, 1337 and 1367. We have appellate jurisdiction under 28 U.S.C. § 1291, and review the Court's decision to grant summary judgment de novo, employing the same legal standards the Court was required to use. Mass. Sch. of Law v. ABA, 107 F.3d 1026, 1032 (3d Cir. 1997).

II. The Antitrust Claim A. Section 1 of the Sherman Act

Section 1 of the Sherman Act provides that "every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce . . . is declared to be illegal." 15 U.S.C. § 1. Although the language in the statute is broad, the Supreme Court has explained that this provision covers only unreasonable restraints of trade. Business Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717, 723, 108 S.Ct. 1515, 99 L.Ed.2d 808 (1988). As is well known, courts utilize two modes of analysis to determine whether a restraint of trade is unreasonable. Under the "per se" standard, "conduct that is `manifestly anticompetitive' or `would always or almost always tend to restrict competition,' . . . is conclusively presumed to unreasonably restrain competition `without elaborate inquiry as to the precise harm [it has] caused or the business excuse for [its] use.'" Rossi v. Standard Roofing, Inc., 156 F.3d 452, 461 (3d Cir. 1998) (quotations omitted). In these cases, the plaintiff need only prove that the defendants conspired among each other and that this conspiracy was the proximate cause of the plaintiff's injury. Id. at 465.

By contrast, under the so-called "rule of reason" standard, the defendants' conduct is analyzed on a case-by-case basis whereby the fact finder weighs all the circumstances in a case to determine whether a particular practice amounts to an unreasonable restraint of trade. Id. at 464. In order to survive summary judgment in cases where this standard applies, the plaintiff must show concerted action, antitrust injury, evidence that the conspiracy produced "adverse, anti-competitive effects within the relevant product and geographic markets," and evidence "that the objects of and the conduct pursuant" to that conspiracy were illegal. Id.

The parties did not discuss in the District Court or before us which standard applies. In its decision, the District Court noted that Cowen "acknowledge[d] that the allegations warrant per se treatment," and thus applied that standard. Intervest, 206 F.Supp.2d at 711 n. 15. We will also apply the per se standard because under either standard, InterVest will need to show the existence of a genuine issue of material fact that (1) Cowen contracted, combined, or conspired with others, and (2) InterVest was injured as a proximate result of that conspiracy. Since we do not believe that InterVest presents sufficient evidence to survive summary judgment on these elements, the distinctions between the per se and rule of reason standards are immaterial.

B. The Summary Judgment Standard in Antitrust Cases 1. Proving the Conspiracy

In Alvord-Polk, Inc. v. F. Schumacher Co., 37 F.3d 996, 999 (3d Cir. 1994), we explained that "[t]he very essence of a section 1 claim . . . is the existence of an agreement," because "section 1 liability is predicated upon some form of concerted action." Unilateral activity by a defendant, no matter the motivation, cannot give rise to a section 1 violation. Rossi, 156 F.3d at 465. This is because a business may "deal, or refuse to deal, with whomever it likes, as long as it does so independently." Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 761, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984).

In order to prove the existence of concerted action among the defendant and other parties, a plaintiff may rely on either direct or circumstantial evidence. Direct evidence "is explicit and requires no inferences to establish the proposition or conclusion being asserted." In re Baby Food Antitrust Litig., 166 F.3d 112, 118 (3d Cir. 1999). Because direct evidence, the proverbial "smoking gun," is difficult to come by, "plaintiffs have been permitted to rely solely on circumstantial evidence (and the reasonable inferences that may be drawn therefrom) to prove a conspiracy." Rossi, 156 F.3d at 465.

Generally, the movant's burden on a summary judgment motion in an antitrust case "is no different than in any other case." Big Apple BMW, Inc. v. BMW of North America, Inc., 974 F.2d 1358, 1363 (3d Cir. 1992). Summary judgment shall be rendered when the evidence shows "that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c). When analyzing the sufficiency of the evidence, the court must view the facts and any reasonable inferences drawn therefrom in the light most favorable to the party opposing summary judgment. Eastman Kodak Co. v. Image Technical Servs., Inc., 504 U.S. 451, 456, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992). However, when a plaintiff relies solely on circumstantial evidence in an antitrust case, we must apply special considerations so that only reasonable inferences are drawn from the evidence. This is because "antitrust law limits the range of permissible inferences from ambiguous evidence in a § 1 case." Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 588, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986).

In Monsanto and Matsushita, the Supreme Court provided clear indications of what types of inferences may not be drawn from circumstantial evidence in an antitrust case. The Court cautioned in Matsushita that "conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy." 475 U.S. at 588, 106 S.Ct. 1348. To survive a motion for summary judgment, therefore, a plaintiff "must present evidence that `tends to exclude the possibility' that the alleged conspirators acted independently." Id. (quoting Monsanto, 465 U.S. at 764, 104 S.Ct. 1464).

When undertaking such an analysis of the evidence, a court must bear in mind that there is "often a fine line separat[ing] unlawful concerted action from legitimate business practice." Petruzzi's IGA Supermarkets, Inc. v. Darling-Delaware Co., 998 F.2d 1224, 1230 (3d Cir. 1993). Nevertheless, Monsanto and Matsushita do "not mean that antitrust defendants are entitled to summary judgment merely by showing that there is a plausible explanation for their conduct; rather, the focus must remain on the evidence proffered by the plaintiff and whether that evidence tends to exclude the possibility that [the defendants] were acting independently." Rossi, 156 F.3d at 466 (internal quotations omitted).

In sum, a court "must ascertain whether the plaintiffs have presented `evidence that is sufficiently unambiguous' showing that the defendants conspired." In re Baby Food, 166 F.3d at 124 (quoting Matsushita, 475 U.S. at 597, 106 S.Ct. 1348). When analyzing the evidence under this summary judgment standard, "a court is not to weigh the evidence or make credibility determinations"; these are tasks left for the fact-finder. Petruzzi's IGA, 998 F.2d at 1230. Further, "a court should not tightly compartmentalize the evidence put forward by the nonmovant, but instead analyze it as a whole to see if together it supports an inference of concerted action." Id.

2. The Direct Evidence Exception to the Monsanto/Matsushita Standard a. InterVest's Argument that the Direct Evidence Exception Applies

We explained in Rossi that "under our jurisprudence, the Matsushita standard applies only when the plaintiff has failed to put forth direct evidence of conspiracy." 156 F.3d at 466 (citing Petruzzi's IGA, 998 F.2d at 1233). This is because direct evidence obviates the fact finder's need to make inferences of a conspiracy, "and therefore the Supreme Court's concerns over the reasonableness of inferences in antitrust cases evaporate." Id. InterVest submits that the evidence it presented in this case qualifies for the direct evidence exception to the Monsanto/Matsushita summary judgment standard and consequently faults the District Court's application of these cases. Its central argument is that the Court misidentified the conspiracy that led to InterVest's alleged antitrust injury.

InterVest contends that its treatment by Cowen, the other broker-dealers, and Bloomberg "was merely one overt act to carry out" the broader conspiracy of "perpetuating a closed bond market that lacked price transparency and preserved secret inflated spreads and prices." In InterVest's submission, the District Court should not have focused on whether InterVest presented evidence that would allow a reasonable inference of a conspiracy to exclude InterVest specifically from the bond market, but instead should have concluded that the direct evidence of Cowen's conduct as a broker-dealer in the bond market "is anti-competitive on its face and therefore requires no limitation on the inferences that can properly be drawn from the circumstantial evidence that such conduct took place."

In order to support this type of analysis, InterVest cites to our decision in Petruzzi's IGA, supra. That case involved an allegation of a restraint of trade in the fat and bone rendering industry. 998 F.2d at 1228. The plaintiff contended that three rendering companies created a cartel that — through the use of various collusive agreements, including an agreement not to bid on existing accounts secured by other rendering companies and threats of punishment for companies that failed to abide by these "rules" — ensured that the price of raw materials for these companies would be artificially low. Id. at 1233. While we determined that Matsushita controlled our analysis of the defendants' motion for summary judgment, we noted that the concerns motivating the Supreme Court in Matsushita were not present. Id. at 1232.

In Matsushita, American makers of television sets alleged that Japanese manufacturers conspired to fix prices below market level so as to push the American companies out of the market. To support their case, the plaintiffs presented evidence that suggested that the Japanese companies earned monopoly profits in Japan, which they then used to finance a coordinated predatory export pricing scheme in the United States. 475 U.S. at 580-81, 106 S.Ct. 1348. The Supreme Court held that this evidence was insufficient to overcome summary judgment. It noted that predatory pricing schemes are generally unlikely to occur, especially when the defendants could not recoup the short-term loss of profits in the long-term. Id. at 588-89, 106 S.Ct. 1348. Because the Court was satisfied that the Japanese manufacturers were unlikely to recoup such losses, it concluded that the defendants lacked the motive to engage in the alleged conspiracy. Id. at 592-93, 106 S.Ct. 1348. The Court also determined that the defendants' behavior was equally consistent with lawful conduct because "cutting prices in order to increase business often is the very essence of competition." Id. at 594, 106 S.Ct. 1348.

In Petruzzi's IGA, we identified two "important circumstances" underlying the Court's Matsushita decision: "(1) that the plaintiff's theory of conspiracy was implausible and (2) that permitting an inference of antitrust conspiracy in these circumstances would have the effect of deterring significant procompetitive conduct." Petruzzi's IGA, 998 F.2d at 1232 (quotation omitted). We have also explained that the Matsushita Court was particularly concerned about allowing "mistaken inferences to be drawn from the defendants' price cutting policies [because] it would chill procompetitive behavior." Id. Therefore, "the acceptable inferences which can be drawn from circumstantial evidence vary with the plausibility of the plaintiff's theory and the dangers associated with such theories." Id.

The alleged conspiracy in Petruzzi's IGA, we concluded, did not present the circumstances or concerns identified in Matsushita for several reasons. First, we determined that the plaintiff's theory was plausible and made "perfect economic sense." Id. Unlike in Matsushita, where it was unlikely that the defendants would be able to make monopoly profits in the long term, the defendants in Petruzzi's IGA could make profits in both the short and long term (as a result of the alleged cartel). Id. Second, in contrast to the price cutting in Matsushita, which the Court characterized as "the very essence of competition," 475 U.S. at 594, 106 S.Ct. 1348, the collusive practices in which the defendants in Petruzzi's IGA allegedly engaged could hardly be described as procompetitive conduct. Finally, we held that the plaintiffs had presented very strong circumstantial evidence that the defendants engaged in consciously parallel behavior that could lead to the reasonable inference that the defendants participated in an unlawful conspiracy. Petruzzi's IGA, 998 F.2d at 1232, 1242-46.

b. Why the Direct Evidence Exception Does Not Apply in this Case

Distinctions similar to those are not present in this case. As we noted above, InterVest maintains that the District Court should have concluded that Cowen's conduct as a broker-dealer in the bond market "is anti-competitive on its face and therefore requires no limitation on the inferences that can properly be drawn from the circumstantial evidence that such conduct took place." Unlike the alleged cartel in Petruzzi's IGA, however, the entire bond market cannot properly be labeled anticompetitive. Although Cowen and other broker-dealers benefit from the relative lack of price transparency in the bond market, this market hardly resembles a collusive regime or an illegal conspiracy. As explained extensively earlier, see supra Section I.A., the lack of price transparency in the bond market benefits investors who wish to transact anonymously and thus reduce the market impact of their transactions. Further, broker-dealers provide the needed liquidity to investors who seek to deal in thinly traded bonds. Importantly, InterVest does not contend that there is anything in the structure of the bond market that prevents the entry of new broker-dealers. InterVest emphasizes, however, when Cowen and other broker-dealers told InterVest it would be more likely to succeed if the company "played by the rules," that this advice represented a conspiracy on the part of the broker-dealers.

We do not believe that this is direct evidence of a conspiracy for several reasons. First, unlike the "rules" in Petruzzi's IGA, which included an agreement not to bid on the other defendants' accounts and punishment for those who defected, the "rules" in this case are simply the operation of the bond market as a whole. There are thousands of broker-dealers in the bond market, trading various types of securities; it stretches credibility to suggest that they all agreed on "rules" in a manner approximating an illegal conspiracy.

Second, as the District Court ably explained in its memorandum and order, the evidence InterVest produces in this case does not approach the clarity of the direct evidence of a conspiracy that we found in other cases, including:

(1) a direct threat to the plaintiff from a competitor that if he went into business his competitors would do anything they could to stop him, including cutting prices or supplies, see [ Rossi, 156 F.3d] at 468; (2) advising distributors that a supplier would cut off access if the distributor failed to maintain a certain price level, see Monsanto, 465 U.S. at 765, 104 S.Ct. 1464; (3) a memorandum produced by a defendant conspirator detailing the discussions from a meeting of a group of alleged conspirators, see Arnold Pontiac-GMC, Inc. v. Budd Baer, Inc., 826 F.2d 1335, 1338 (3d Cir. 1987); and (4) a public resolution by a professional association recommending that its members withdraw their affiliation with an insurer, see Pa. Dental Ass'n v. Med. Serv. Ass'n of Pa., 815 F.2d 270, 273 (3d Cir. 1987).

Intervest, 206 F.Supp.2d at 713. Direct evidence "must be . . . explicit and require no inferences to establish the proposition or conclusion being asserted." In re Baby Food Antitrust Litig., 166 F.3d at 118. A vague reference to "rules" is insufficiently explicit and requires ample inferences of illegal actions that would constitute a conspiracy among Cowen and other broker-dealers in violation of section 1 of the Sherman Act.

Finally, and in a similar vein, we are mindful of the Supreme Court's warning in Monsanto that evidence must not be too broadly construed lest such a conclusion "create an irrational dislocation in the market." 465 U.S. at 764, 104 S.Ct. 1464. The Court further cautioned against drawing unreasonable inferences from evidence in such a manner that legal practices will be deterred and the antitrust statute will be applied overinclusively. Id. at 763-64, 104 S.Ct. 1464. InterVest in effect invites us to turn on its head the summary judgment standard articulated by the Supreme Court in Monsanto and Matsushita by holding that Cowen's role as a broker-dealer in the bond market is direct evidence of conspiracy. As we have explained above, however, Cowen's operations as a broker-dealer are not "facially anticompetitive and exactly the harm the antitrust laws aim to prevent," Eastman Kodak, 504 U.S. at 478, 112 S.Ct. 2072. We must therefore take "special care . . . in assigning inferences to circumstantial evidence," Alvord-Polk, 37 F.3d at 1001, of Cowen's participation in an alleged conspiracy to harm InterVest. Monsanto and Matsushita's concerns about the chilling effect on lawful conduct that would result from the unreasonable interpretation of evidence are particularly germane in this case because if we hold that the bond market is a conspiracy, every broker-dealer would be subject to antitrust liability. We must therefore analyze Cowen's motion for summary judgment under the standards articulated in those cases.

C. Application of the Monsanto/Matsushita Standard

In reviewing Cowen's motion for summary judgment, we must determine whether a fact finder could draw a reasonable inference from the evidence that Cowen participated in a conspiracy that resulted in an antitrust injury to InterVest. InterVest alleges two antitrust theories: (1) a conspiracy on the part of Cowen and other broker-dealers; and (2) a conspiracy between Bloomberg and Cowen. We will review each of these theories separately, but in doing so, we will take care to view the evidence as a whole and not compartmentalize individual pieces of evidence. See Petruzzi's IGA, 998 F.2d at 1230.

1. Concerted Action Among Cowen and Other Broker-Dealers a. Conspiracy on the Basis of Circumstantial Evidence

Cowen's principal defense is that any action it undertook with respect to InterVest was unilateral. It emphasizes that InterVest fails to proffer a single piece of evidence that reasonably suggests that Cowen communicated with other broker-dealers in regards to InterVest. This is in stark contrast to the evidence presented by plaintiffs in other antitrust decisions rendered by this court. See, e.g., Petruzzi's IGA, 998 F.2d at 1233-34 (evidence that defendants had discussed a "code" of price fixing strategies); Rossi, 156 F.3d at 468-70 (evidence of meetings between the defendants in which price fixing was discussed as well as specific evidence of other pressure tactics used by defendants); Alvord-Polk, 37 F.3d at 1013-14 (evidence of communications among defendants regarding actions they individually were undertaking against competitors); Big Apple BMW, 974 F.2d at 1367-68 (evidence of meetings among dealers about how they were opposed to a new franchisee and indicating continued efforts to pressure manufacturer).

InterVest's attempts to rebut this assertion are unavailing. It notes that Garcia told Fondren that Bloomberg's decision to suspend development of InterVest's municipal bond trading platform was due to pressure from Cowen and other "dealers" who had seen InterVest's advertisements. InterVest also points to an alleged comment by a Bear Stearns representative that the firm would share its concerns about InterVest with other broker-dealers, and the internal BONY memorandum claiming that Bloomberg ceased dealing with firms like InterVest and BondNet because of "pressure from Wall Street." While this evidence might indicate that Bloomberg received and responded to multiple complaints from various broker-dealers, it does not reasonably establish as a matter of evidentiary proof that the broker-dealers actually agreed to work together to harm InterVest or even communicated with each other about InterVest. At all events, InterVest does not proffer any evidence indicating that Cowen, in particular, spoke with other broker-dealers about InterVest.

InterVest points to comments by Joseph Cohen, the former chairman of Cowen, regarding what would happen to Cowen if it gave the prices and types of bonds it was trading, known as "pictures," to InterVest to place on its system. InterVest interprets Cohen's testimony in his deposition that he "would have lost [his] business" if he gave the pictures to InterVest as meaning that the broker-dealers in the bond market had "rules" that included punishment by boycott if a firm dealt with a company like InterVest. In InterVest's submission, this is evidence of a tacit agreement among broker-dealers.

Cowen argues that this is too much of a "stretch" to satisfy evidentiary standards. We agree. It also submits that this is not a reasonable interpretation of Cohen's statement. Cowen explains that because it operated both as an inter-dealer broker and a dealer, it had to maintain strict confidentiality between these two aspects of its business. If it did not, Cowen's use of such insider information would harm the other broker-dealers using its inter-dealer services or the investors using its broker-dealer services. As explained earlier, anonymity is often sought by investors in the bond market; if Cowen breached such confidentiality, no investor or broker-dealer would use its services, i.e., Cohen "would have lost [his] business." Therefore, in not dealing with InterVest, Cowen was exercising a unilateral and perfectly reasonable course of action.

InterVest counters that even if we believe Cowen's interpretation of Cohen's statement, this raises a genuine issue of material fact for the jury to decide. We disagree. In Petruzzi's IGA we explained that the Matsushita Court "stated that the acceptable inferences which can be drawn from circumstantial evidence vary with the plausibility of the plaintiffs' theory and the dangers associated with such inferences." 998 F.2d at 1232. In this case, InterVest would ask the jury to draw an inference from Cohen's statements that is in contravention of the workings of the bond market and would chill a lawful practice of broker-dealers who operate both regular dealer and inter-dealer operations.

The evidence that Cowen refused to share its pictures with InterVest is consistent with the Monsanto Court's statement that a business may "deal, or refuse to deal, with whomever it likes, as long as it does so independently." 465 U.S. at 761, 104 S.Ct. 1464. There are many reasons that a broker-dealer might independently choose not to partner with a fledgling start-up whose technology and business model remained unproven. As noted above, in the entire time that InterVest was on the Bloomberg system, only ten trades were executed, all of which had been pre-arranged over the telephone. Further, Cowen presented largely unrefuted evidence from its expert in computer programming that InterVest's trading platform contained design flaws, including a risk of data loss. Evidence of Cowen's decision not to do business with InterVest, therefore, does not "tend to exclude the possibility" that Cowen acted individually instead of conspiring with other broker-dealers. Id. at 764, 104 S.Ct. 1464.

b. Conspiracy on the Basis of Consciously Parallel Behavior

Without any evidence of communication between the broker-dealers or other reasonable inferences of concerted action, InterVest would need to show that the broker-dealers engaged in activity that approximates conscious parallelism. In Petruzzi's IGA, we explained that in order to establish a conspiracy on the basis of "consciously parallel behavior, a plaintiff must show: (1) that the defendants' behavior was parallel; (2) that the defendants were conscious of each other's conduct and that this awareness was an element in their decision-making processes; and (3) certain `plus' factors." 998 F.2d at 1242-43. These "plus" factors include: "(1) actions contrary to the defendants' economic interests, and (2) a motivation to enter into such an agreement," each of which must be present. Id. at 1242. We require these additional "plus" factors in order to allay the concerns expressed by the Supreme Court in Monsanto and Matsushita, "namely the desire not to curb procompetitive behavior." Id.

Even assuming that InterVest can fulfill the first two requirements, it cannot prove both of the "plus" factors. As for the first "plus" factor, InterVest cannot establish that Cowen's conduct with regard to InterVest was against Cowen's economic interests. The record is barren of evidence of price fixing or any other non-competitive behavior by Cowen that was against its economic interests. Cf. In re Baby Food, 166 F.3d at 132-38 (lack of evidence of price fixing or an agreement to preserve market share failed to fulfill "plus" factors needed to adduce a conspiracy). Rather, as described extensively above, broker-dealers benefit from the relative lack of price transparency in the bond market through their ability to charge high spreads on transactions. Cowen's decision not to deal with InterVest, therefore, was in alignment with its economic interests.

With regard to the second "plus" factor, while there is evidence tending to show that Cowen or other broker-dealers could have viewed InterVest as a threat to their business, that is not sufficient for any inference that Cowen had sufficient motive to conspire to boycott or otherwise harm InterVest. First, it is reasonable to conclude that Cowen simply chose not to partner with a new company with unproven technology that, in theory, could reduce the profits the firms earned. Further, assuming that Cowen took InterVest's advertised potential to refashion the bond market seriously, Cowen had a legitimate reason not to aid InterVest in its pursuit so that Cowen could maintain the level of profits it earned from the spreads on transactions. This motivation is not evidence of a conspiracy. See id. at 137 (stating that companies "have a legitimate understandable motive to increase profits" and that "[p]rofit is the essence of a capitalistic society"). Given the paucity of evidence of interaction between the broker-dealers regarding InterVest, we cannot infer that Cowen had a motivation to agree with other firms to injure InterVest. At all events, because the first "plus" factor is clearly not fulfilled, InterVest cannot survive summary judgment on a conscious parallelism claim.

InterVest correctly argues in its reply brief that "there is more here than the refusal of one dealer at a time to deal with InterVest" — there is also the possibility of collusion with Bloomberg. While this is not a response to Cowen's defense that it did not conspire with other broker-dealers, it is another possible antitrust theory.

2. Concerted Action Between Cowen and Bloomberg

Having failed to present evidence that reasonably suggests that Cowen conspired with other broker-dealers, InterVest attempts to save its case by arguing that Cowen conspired with Bloomberg. To support this contention, InterVest points to Fondren's recollection that Charles Garcia from Bloomberg told him that Bloomberg was getting pressure from broker-dealers, and specifically Cowen, because it was moving forward in its relationship with InterVest. In that same conversation, Garcia notified Fondren that Bloomberg was placing a moratorium on the development of InterVest's municipal bond trading platform. InterVest also relies on the internal BONY memorandum stating that Bloomberg wanted to stop dealing with firms like BondNet and InterVest because of "pressure from Wall Street." Finally, InterVest cites several internal Bloomberg communications, including an email recalling that "Mike," who likely is Michael Bloomberg, the founder of Bloomberg and the current mayor of New York City, "[said] no to firms like InterVest."

Looking at this evidence in the best light for InterVest, we cannot conclude that it establishes a reasonable basis for inferring concerted action between Cowen and Bloomberg. Even if the evidence showed that Bloomberg responded to complaints from Cowen or other broker-dealers, this fact would be legally insufficient to prove a conspiracy. Very little of the evidence provided by InterVest relates specifically to Cowen's conduct with respect to Bloomberg. At most, InterVest shows that Cowen complained to Bloomberg about its relationship with InterVest. Evidence of merely a complaint, without any other supporting evidence tending to show illegal pressure or a conspiracy, is insufficient to survive summary judgment because "[p]ermitting an agreement to be inferred merely from the existence of complaints, or even from the fact termination came about `in response to' complaints, could deter or penalize perfectly legitimate conduct." Monsanto, 465 U.S. at 763, 104 S.Ct. 1464.

InterVest ignores this legal standard, instead interposing arguments that call into question Bloomberg's proffered reason for terminating its relationship with InterVest: the high cost of developing and running the system compared with its minimal use. InterVest claims that since Bloomberg had an interest in the dissemination of information, it would be against Bloomberg's business interest to frustrate its relationship with InterVest, which aimed to open the bond market to price transparency. For this reason, InterVest submits, Bloomberg's explanation that it severed its relationship with InterVest because it was devoting an inordinate amount of resources to a little-used service must be pretextual.

Even assuming that this is a reasonable inference of Bloomberg's motivation and actions, InterVest's focus on Bloomberg's purported conduct does not suggest that Bloomberg did anything but act unilaterally, and therefore this conduct cannot be used to prove an antitrust violation on the part of Cowen. InterVest does not present any evidence of Cowen threatening Bloomberg, e.g., by no longer providing it with general information on its bond trading or by refusing to use its information services, nor does InterVest show an agreement between the parties to thwart it. Rather, the evidence that InterVest presents of Cowen's concerted action with Bloomberg is at best ambiguous and does not "tend to exclude the possibility" that Cowen and Bloomberg acted independently. Monsanto, 465 U.S. at 764, 104 S.Ct. 1464.

D. Allegation that the District Court Improperly Weighed the Evidence

Finally, InterVest urges us to reverse on the ground that the District Court stated that a reasonable jury could have found InterVest's conspiracy theory plausible but then "proceeded to weigh the evidence on either side of the conspiracy issue, to arrive at an outcome of exactly 50/50. . . ." InterVest is referring to the section of the District Court's opinion in which it stated:

a rational trier of fact might draw the series of inferences necessary to find that Cowen participated in a conspiracy with Bloomberg and other broker-dealers in order to maintain the existing closed bond trading system. The establishment and perpetuation of a closed system, the allusions by various broker-dealers, including defendant Cowen, to the `rules,' the universal refusal to deal with InterVest, the complaints to Bloomberg, and Bloomberg's subsequent resistance to buy side and transparent bond trading, all support an inference of conspiracy.

Intervest, 206 F.Supp.2d at 716.

The Court then went on to conclude that InterVest "fails to survive summary judgment because all of this evidence supports equally well the inference that Cowen, other broker-dealers, and Bloomberg acted independently and rationally in light of their own interests and consumer demand." Id. While these passages, viewed out of context, give the impression that the Court was improperly balancing the evidence, a job left to the fact-finder, what the Court clearly did was to conclude that the evidence InterVest presents, when viewed as a whole and under the standard established by Monsanto and Matsushita, could not lead to the reasonable inference by a jury that Cowen participated in an antitrust conspiracy, a conclusion with which we agree. Although the Court might have been more careful in its language, we do not believe it committed an error in its analysis.

III.

For all the foregoing reasons, the judgment of the District Court with respect to the antitrust claim will be affirmed. We will also affirm the Court's summary judgment order on the tortious interference with contract claim for the reasons stated in the margin.