United States District Court, E.D. Virginia, Alexandria Division
December 24, 2014
UNITED STATES OF AMERICA ex rel. Kurt Bunk and Daniel Heuser, Plaintiffs/Relators,
BIRKART GLOBISTICS GmbH & CO., et al., Defendants. UNITED STATES OF AMERICA ex rel. Ray Ammons, Plaintiff/Relator,
THE PASHA GROUP, et al., Defendants.
ANTHONY J. TRENGA, District Judge.
Following remand from the Fourth Circuit, this case was tried before a jury beginning on July 21, 2014. On August 1, 2014, the jury returned a verdict against Defendants Gosselin World Wide Moving, N.V. and Marc Smet with respect to two provisions of the False Claims Act, 31 U.S.C. §§ 3729(a)(1) and (a)(3). Presently before the Court is Defendants Gosselin World Wide Moving N.V. and Marc Smet's Renewed Motion for Judgment as a Matter of Law and Alternative Motion for New Trial [Doc. Nos. 1321 and 1322] ("the Motion"), on which the Court held a hearing on September 4, 2014.
Gosselin, located in Europe, provided services to American carriers who contracted with the United States to move the household goods of military personnel to and from Germany, known as the ITGBL program. The United States claims that Gosselin engaged in a fraudulent course of conduct that inflated rates the United States paid to American carriers under every ITGBL program contract it awarded during the 2001-2002 period, even those to carriers that did not use Gosselin's services. That fraudulent conduct, as described by the government, consists of "a scheme to eliminate competition from the ITGBL bidding process with the intent and effect of inflating prices that DOD paid for moves." See United States and Relators' Memorandum in Opposition to Defendants' Renewed Motion for Judgment as a Matter of Law and Alternative Motion for New Trial (hereinafter "Government's Brief'), Doc. No. 1325, at 4. See also Doc. No. 1298 ("[A]ll claims submitted during these rate cycles were false or fraudulent, because all moves for which the Department of Defense paid were the subject of an anticompetitive agreement that eliminated competition and inflated rates in what was an explicitly competitive program."). There was no evidence, however, and the government does not contend, that Gosselin made any false statements or certifications, express or implied, or that it failed to comply with any contract provisions, statutes, or regulations that were a term or condition for payment under or Gosselin's participation in the ITGBL program. Rather, the government contends that Gosselin's anticompetitive conduct alone is sufficient to impose False Claims Act liability, even if it does not constitute an anti-trust violation. See Doc. No. 1304, Tr. at 1169:6-16; 1191:2-5, 1200:25-1201:1-25 (arguing that any type of collusion that affects the ITGBL program constitutes fraud). This case therefore involves whether and to what extent the False Claims Act liability may be imposed on companies and individuals who do not (1) enter into any contracts with the federal government; (2) submit any claims to the federal government; (3) receive any funds directly from the federal government; (4) make any misrepresentations or fraudulent non-disclosures, express or implied, to the federal government or anyone who contracts with the federal government; (5) violate any contractual provisions, laws, regulations or statutes that constitute terms or conditions of payment to those who contract with or provide services to the federal government;(7) engage in anticompetitive conduct that violates any statutory or other prohibitions; or (6) collude or conspire with anyone who does any of the foregoing.
For the reasons discussed below, the Court concludes that the government's theory of liability is both unprecedented and untenable. There was no evidence that Gosselin engaged in any deceptions or misrepresentations and the evidence was therefore insufficient, as a matter of law, to support the jury's finding of liability under instructions that required the jury to find, in order to impose liability, that Gosselin engaged in conduct that "knowingly deceived" the United States and "knowingly caused" the government to enter into an ITGBL contract. For similar reasons, discussed below, the Court finds, in its capacity as fact finder on the issue of materiality, that Gosselin did not engage in conduct that was "material" to the government's awarding ITGBL contracts or making payments thereunder since Gosselin did not engage in any conduct that pertained to any term or condition of payment to the American carriers that submitted claims to the United States for payment. The Court also concludes that the evidence was insufficient for the jury to find that Gosselin caused a specific, identifiable false claim to be presented to the government for payment, or the total numbers of such claims, or to award damages. Finally, the Court concludes that if the Court's decision to enter judgment in favor of the Gosselin defendants is vacated or reversed on appeal, a new trial is warranted on all issues. For these reasons, the Court GRANTS the defendants' motion for judgment as a matter of law and also CONDITIONALLY GRANTS defendants' motion for a new trial pursuant to Fed. R. Civ. Pro. 50(c)(1).
The lengthy procedural history and facts of this case are set forth in detail in this Court's previous orders and memorandum opinions. Briefly summarized, these consolidated actions were originally filed in 2002 by Relators Kurt Bunk and Ray Ammons but remained under seal until May 19, 2008. See Doc. No. 97. On July 18, 2008, the United States intervened as to all claims relating to the ITGBL program, but not as to certain other claims relating to a contract awarded to Gosselin in 2001 under the Direct Procurement Method (DPM) program. The case was tried beginning on July 18, 2011, and on July 28, 2011, following the close of the United States' case in chief, the Court dismissed the ITGBL claims other than those that related to two sets of transportation channels, referred to in this litigation as the Cartwright and Covan Channels, based on the antitrust immunity provision of the Shipping Act of 1984, 46 U.S.C. §§ 40301-40307. See Doc. No. 1032, Trial Tr. at 1031-64; Doc. No. 1072 (Memorandum Opinion). On December 19, 2013, the Fourth Circuit reversed this Court's ruling on immunity under the Shipping Act, vacated this Court's order based on its contrary ruling, and remanded the case for further proceedings. See Doc. No. 1167.
The retrial began on July 21, 2014. On July 28, 2014, following the close of the United States' case in chief, the defendants moved for judgment as a matter of law. The Court reserved on the motion. The defendants then renewed their motion at the close of the evidence. At that time the Court granted, without opposition, the motion as to the United States' common law claims and 31 U.S.C. § 3729(a)(2) claim and did not grant the motion as to claims under 31 U.S.C. § 3729(a)(1) and (3), submitting those claims to the jury. On August 1, 2014, the jury returned its verdict in favor of the United States and against Defendants Gosselin World Wide Moving, N.V. and Marc Smet on both claims. It also found that those defendants knowingly caused to be submitted 58, 950 false claims and awarded a total of $33.6 million in damages. Defendants Gosselin World Wide Moving, N.V. ("Gosselin") and Marc Smet now move for judgment as a matter of law or a new trial.
II. STATEMENT OF FACTS
The evidence at trial was in most material respects undisputed and essentially consisted of the same evidence presented at the first trial and discussed in this Court's Memorandum Opinion dated August 26, 2011. See Doc. No. 1072. Briefly summarized, that evidence, with disputed factual issues viewed most favorably to the government, established the following:
For decades the United States has transported the household goods of military personnel posted in Germany through the International Transportation Government Bill of Lading program, or ITGBL program. As the ITGBL program operated during the relevant time period, the Department of Defense ("DOD"), acting through the Surface Deployment and Distribution Command ("SDDC"), on a biannual basis, solicited bids for moves from American owned freight forwarding companies, also called carriers or Transportation Service Providers ("TSPs"). There were therefore two cycles per year, summer and winter, designated, for example, as IS01 for International Summer 2001 and IWO1 for International Winter 2001. The claims in this case involve the bids that were submitted by American carriers during the four cycles that occurred in 2001 and 2002: ISO1, IW01, IS02, and IW02.
Overall, there were one hundred and four channels between the United States and Germany, corresponding to different transportation routes: fifty-two westbound and fifty-two eastbound. The carriers submitted a separate bid, in the form of a dollar price per hundred weight, for each of the fifty-two channels, for as many or as few channels as they chose. During the relevant years, carriers filed their initial bids in November for the IS cycle, which started on April 1 in the following calendar year, and in May for the IW cycle, which started on October 1 of that same calendar year. The bidding took place pursuant to a two-step process. After carriers submitted their initial round of bids, the DOD published the five lowest rates. The lowest bid for a particular channel was known as the "prime rate, " and the carrier that submitted the lowest bid was guaranteed at least 10% of the total volume for that channel. After the publication of the five lowest rates, other carriers who had submitted bids in the initial round had the opportunity to match, or "me-too", those prime rates and thereby be guaranteed a particular percentage of the total volume for the channel.
The carriers adopted a variety of bidding strategies. Some would intentionally file a bid intended to set the prime rate. Others would file an "administrative high" bid that simply preserved their opportunity to file a subsequent bid in the me-too round. It was generally understood, and generally the case, that only those carriers that either set or me-too'ed the prime rate would receive a significant number of shipments in any given rate cycle. Indeed, it would appear from certain of the government's exhibits that, in a given rate cycle, less than five percent of the moves would ship at a rate above the prime rate. See, e.g., Gov. Ex. 157 for IS01.
Carriers generally did not perform all aspects of the moves themselves, but rather subcontracted certain portions of the move. The inland portions of the moves within Germany, i.e., the packing or unpacking and line haul services in Germany, were performed by "local German agents, " essentially local moving companies. The moves also sometimes involved "general agents" that would act as intermediaries between the American carriers and the German local agents and other agents. Gosselin provided services as both a local agent and as a general agent. Only the American carriers contracted directly with the United States. Only the American carriers received money from the United States and the United States paid directly to an American carrier the entire cost of a move pursuant to a voucher that presented for payment a government bill of lading ("GBL") particular to each move. The American carriers, in turn, paid the local and general agents they used to perform their transportation obligations to the United States.
Under the government's regulations, carriers were required to file what were called "compensatory bids." A "compensatory bid" was not defined, but it was generally understood that a compensatory bid was a bid that covered a carrier's costs and provided a reasonable profit. This requirement, however, does not appear to have been monitored or enforced by the government and as discussed below, certain carriers appear to have either set prime rates or metoo'd prime rates at prices that were below their overall costs in order to obtain a guaranteed percentage of total tonnage.
By the IWOO cycle, prime rates had dropped to historically low levels. There was uncontradicted evidence presented at trial that certain American carriers set prime rates or metoo'd prime rates that were for those particular carriers non-compensatory. As a result, certain carriers received payment for the ITGBL services that were insufficient to cover all their costs and return a reasonable profit. Overall, these bidding practices, and the ITGBL rates they established, negatively affected the ITGBL program in several ways. For example, the local agents in Europe would often go unpaid or only partially paid by the American carriers; and as a result, the local agents, out of fear of non-payment, cut corners on such items as packing materials that affected the quality of the moves. This downward spiral of prime rates proved unsustainable for certain carriers; eleven carriers, including two that set a significant number of prime rates and were known for filing non-compensatory rates, went bankrupt in the 1999 to 2000 period. The SDDC was concerned with the quality of the moves affected by non-compensatory rates, and began to consider changes in the structure of the ITGBL program to eliminate its policy of accepting prime rates solely on the basis of price. The local agents had other complaints about how they were being treated by the American carriers, including what they believed was the practice of certain American carriers to reduce payments to the local agents based on unsubstantiated claims that they were responsible for damage to goods that occurred at some point during the transportation process.
Against this backdrop, on November 14, 2000, Gosselin and five other local agents met in Sonthofen, Germany and entered into an agreement that became known as the Sonthofen Agreement. Under the Sonthofen Agreement, the signatories agreed to work under a "landed rate" for the transportation of military household goods in the ITGBL program. The Agreement provided in its entirety: "We, the undersigned companies agree that as of 4/1/2001 we will be using landed rates' for the U.S. Military business." Gov. Ex. 14.
A landed rate was a single price offered to a TSP for that bundle of services necessary to deliver household goods to or from an American port and a military family's residence in Germany) Gosselin had been offering landed rates to American carriers since the 1990s and the government does not challenge the use of a bundled rate, as such. Rather, the government objects to the Sonthofen Agreement on the grounds that, as understood by the signatories, it required the local agents to work as of April 1, 2001 exclusively under a landed rate, and not work under any contracts directly with an American carrier. As a practical matter, this agreement was understood to mean that the carriers could not contract directly with the local agents, who would work only through a general agent signatory, either Gosselin or ITO. While the evidence did not establish that the local agents agreed at Sonthofen on particular prices for their services, the signatories understood that the purpose of the Sonthofen Agreement was to raise the prices they would receive for their services as local agents in Germany. The signatories to the Sonthofen Agreement constituted approximately 70% of the market for local agents available for ITGBL services and some other non-signatory local agents supported it as well.
In order to implement the Sonthofen Agreement, Gosselin proposed separately to each of the local agents who signed the Sonthofen Agreement a written rate agreement contract which set forth the specific prices Gosselin would pay for their services. See, e.g., Gov. Ex. 24. Under that contract, Gosselin guaranteed that it would pay the local agent it hired within thirty days, thereby eliminating the credit risk associated with dealing with American carriers, and established a fund from which it would pay the carriers' claims for damaged goods.
Following the Sonthofen Agreement, the local agents informed the carriers of their intent to work only under the landed rate system. Gosselin also informed its American carrier customers that rates had increased and that the local agents would no longer be handling business at lower rates. Following the Sonthofen Agreement, Gosselin and ITO entered into landed rate agreements with certain, but not all, American carriers; and carriers using landed rates set the majority of prime rates for the four cycles that comprised the alleged conspiracy period.
Prime rates increased in all channels for the ISO1 cycle over the previous year; and certain signatories attributed that increase to the Sonthofen Agreement. They also acknowledged as a general proposition that an increase in local agent rates would generally result in an increase in the landed rate. The evidence was also sufficient to establish that once the Sonthofen Agreement went into effect, American carriers no longer had the same ability to negotiate rates directly with local German agents and that carriers would generally pass along through higher rates any increases in costs,  although representative of two carriers testified that they did not always incorporate price increases into their bids to reflect higher costs. There was, however, no evidence concerning the actual price impact the Sonthofen Agreement had on any particular prices, either for local agent services, Gosselin's general agency services, or any particular carrier's bids. More specifically, there was no evidence concerning what local agents, landed rate providers, or carriers would have charged in the absence of the Sonthofen Agreement; and the history of the ITGBL program showed repeated instances of sharp upward spikes in prime rates after the kind of prolonged decline in prime rates that had occurred in the years before IS01.
There was also uncontradicted evidence that, despite the Sonthofen Agreement, certain of its signatories did in fact, on occasion, contract directly with American carriers. For example, both signatories ITO and Christ offered services outside of the landed rate during IS01-IW02. There was also no evidence that any American carrier who wanted to contract directly with a local German agent was unable to do so; and certain carriers did obtain local agent services without using a landed rate. The carrier Cartwright International Van Lines, in particular, which set a significant number of prime rates during the alleged conspiracy period, consistently operated outside of the landed rate system, and in fact set bids without first consulting with local agents regarding their prices. There was also uncontradicted evidence that, regardless of the Sonthofen Agreement, carriers wanted to use the landed rate system because of its benefits, as they had before the alleged conspiracy period, because the landed rate lowered their transactional costs and eliminated certain currency risks.
III. LEGAL STANDARD
Under Fed. R. Civ. P 50, the Court may grant a motion for judgment as a matter of law on a particular issue if the Court concludes that "a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue, " that is, that the jury's findings on that issue are not supported by substantial evidence. See Wilhelm v. Blue Bell, Inc., 773 F.2d 1429 (4th Cir. 1985). Under Rule 59, "[t]he court should grant a new trial only if 1) the verdict is against the clear weight of the evidence, 2) is based on evidence which is false, or 3) will result in a miscarriage of justice, even though there may be substantial evidence which would prevent the direction of a verdict." Dennis v. Columbia Colleton Med. Ctr., Inc., 290 F.3d 639, 650 (4th Cir. 2002).
In order to impose False Claims Act liability, the government was required to show that Gosselin "knowingly presented, or caused to be presented, to the government, a false or fraudulent claim for payment or approval" or conspired to do so. The government contends that the evidence is sufficient to sustain the jury's verdict on liability because the evidence showed that Gosselin engaged in a fraudulent course of anticompetitive conduct that inflated the prime rates paid by the government above what they otherwise would have been. See Government's Brief, Doc. No. 1325, at 4-5 (Gosselin engaged in "anticompetitive conduct" consisting of "a scheme to eliminate competition from the ITGBL bidding process with the intent and effect of inflating prices that DOD paid for moves."). See also Doc. No. 1304, Tr. at 1169:6-16 (where the government agrees that in order to establish liability it must prove that the carriers filed prime rates higher than they would otherwise absent defendant's anticompetitive conduct).
The government does not contend that Gosselin engaged in any express or implied misrepresentations or non-compliance with any rule, regulation or statute; in fact, it contends that under its theory of "fraudulent conduct" no such showing is necessary. See Doc. No. 1307, Tr. at 698 (government confirming that it is not proceeding on any theory that Gosselin made express or implied misrepresentations or violated any rule, regulation or statute.) Rather, the government contends that it was enough to show that Gosselin's conduct "directly undermined the integrity of the price that [the government was] charged." Id. In advancing this theory of liability, the government makes a distinction between False Claims Act liability that is based on claims that are "legally false" and those that are "factually false." Here, the government claims that even though the claims it paid were not "legally false, " because Gosselin did not make misrepresentations or engage in conduct that did not comply with any term or condition of payment, the claims paid by the government were "factually false" because the claims were based on a prime rate that was "artificially inflated beyond a competitive price" by Gosselin's conduct. Id. at 700. The government further contends that Gosselin's conduct caused all prime rates to be higher than what they otherwise would have been and that every claim for payment by the carriers during the alleged conspiracy period was therefore a "factually false" claim, even those filed by carriers that did not use the landed rate or have any dealings with Gosselin or any other signatory to the Sonthofen agreement. See Doc. No. 1304, Tr. at 1179:12-18 (where the government agrees that it is claiming "not only are all the rates of the Gosselin carriers that they filed inflated, the rate of every carrier that was filed as a prime rate or a me-too rate was inflated even if they never dealt with Gosselin and they never used the landed rate.") Critical to the government's theory of liability or for the imposition of penalties is that it need not show that any particular prime rate paid by the government was, in fact, inflated because of Gosselin's alleged anticompetitive conduct. The dispositive issues therefore reduce to whether False Claims Act liability can be imposed based on anti-competitive conduct alone, without any false statement or non-compliance with any term or condition of payment and if so, without any showing that a specific, identifiable claim for payment by a specific carrier was inflated because of Gosselin's alleged conduct.
The Court concludes that in order to impose False Claims Act liability under the facts of this case the government was required to show that Gosselin engaged in some misrepresentation, express or implied, or failed to comply with some term or condition for payment. The Court further concludes that the government was required to show that a specific, identifiable claim was paid based on a prime rate that was inflated beyond what it would have otherwise been absent Gosselin's alleged anticompetitive conduct. Because the evidence was insufficient to make those showings, the jury's verdict is set aside and judgment will be entered in favor of the defendants.
The government bases its theory on the line of cases, beginning with the seminal case of Marcus v. Hess, 317 U.S. 537 (1943), that impose False Claims Act liability based on bidrigging or other anti-competitive conduct. All of those cases, however, involved an element of deception that is absent in this case and the government's version of the fraudulent inducement theory to impose False Claims Act liability in this case goes far beyond any reported application of that theory, as well as the jurisprudential basis for that theory.
Hess involved a conspiracy among electrical contractors to rig their bids on contracts with local governments for Public Works Administration projects. Specifically, all of the bidders on a series of contracts entered into an agreement whereby they would average their bids; and in rotating fashion, one among them would submit the average bid as the low bid on a specific contract, with the others submitting higher bids, thereby giving the appearance of competitive bidding. While the contractors did not contract directly with the United States, the United States provided most of the money used to pay the contractor to whom the contract was awarded and did so by placing the funds in a joint account administered by the local governmental unit that sponsored the work, thereby making the local governments essentially the paymaster for the United States. The Court assumed, without deciding, that the bid rigging scheme was fraudulent, see 317 U.S. at 540, and held that it came within the prohibition of the False Claim Act, even though the United States was not actually a party to the fraudulently induced contracts. As the Court in Hess explained, "[t]he bidding itself was a federal requirement; all bidders were fully advised that these were P.W.A. projects; and many if not most of the respondents certified that their bids were genuine and not sham or collusive.' Id. at 543. In that regard, the Court observed that "the jury and both [lower] courts found that the contracts were obtained by a successfully executed conspiracy to remove all possible competition from competitive bidding'. Id. The Court also noted that "[t]he government's money would never have been placed in the joint fund for payment to [the contractors] had its agents known the bids were collusive." Id. Thus, "[b]y their conduct, the [contractors]... caused the government to pay claims of the local sponsors in order that they might in turn pay respondents under contracts found to have been executed as the result of the fraudulent bidding." Id
The government contends that Hess stands for the proposition that False Claims Act liability may be based on "any conduct that corrupts a competitive procurement process and artificially inflates prices[, ]" even without any accompanying misrepresentations or false certifications. See Government's Brief, Doc. No. 1325 at 6. See also Doc. No. 1307, Tr. at 698:7-16 (acknowledging that the government was not proceeding on a theory that Gosselin misrepresented anything or violated an implied representation). However, while the Court in Hess did not rely on the false written certifications that most of the contractors filed, the Court's decision has been generally understood as imposing liability based on an implied representation that the presented bids were competitive and non-collusive, as reflected in the other cases relied on by the government. For example, in Murray & Sorenson v. U.S., 207 F.2d 119 (1st Cir. 1953), a purchasing agent for contractors working on a Navy project provided inside information to a company bidding on supply materials for the project, all in derogation of his duty to solicit competitive bids. As a result of the inside information received in breach of an explicit duty, the company supplying the materials bid $5.00 rather than $4.25 on each faucet, and those prices were charged to the United States on a cost-plus-fixed-fee basis. The First Circuit imposed liability on the basis of an implied representation theory and concluded that the scheme included "an element of falsehood comparable to that in Hess." Id. at 23. The First Circuit also observed that "in Hess there was an implied false representation that the bids were competitive, " while in Murray & Swenson, "there was an implied false representation that the bids were at a figure which the corporate defendant would have submitted in competition instead of at a somewhat higher figure suggested by the contractors' purchasing agent." Id. at 124; see also United States ex rel. Weinstein v. Bressler, 160 F.2d 403, 405 (2d Cir. 1947) (describing Hess as turning on the "fraudulent misrepresentation" that the contractors submitting the high bids "hoped to secure the contracts on which they bid"). See also Harrison v. Westinghouse Savannah River Co., 176 F.3d 776, 788 (4th Cir. 1999) (where the Fourth Circuit characterizes Hess as involving a contract that "was originally obtained based on false information and fraudulent pricing.")
Likewise in United States v. Dynamics Research Corp., supra, also relied on by the government, liability was imposed where an Air Force contractor responsible for identifying "best value" computer equipment suppliers engaged in self-dealing in violation of an explicit conflict-of-interest provision of its contract. And in Hooper v. Lockheed Martin Corp., 688 F.3d 1037 (9th Cir. 2012), the defendant knowingly misrepresented its projected costs in bidding on a cost-plus contract. See also United States ex rel. Longhi v. Lithium Power Technologies, Inc., 575 F.3d 458 (5th Cir. 2009) where the defendant submitted grant proposals containing false statements. It would appear that the government's reliance on an expansive application of Hess is even more untenable within the Fourth Circuit, see Harrison v. Westinghouse Savannah River Co., supra, 176 F.3d at 788 ("Our decision in Berge, 104 F.3d at 1461, holding, in part, that there can be no False Claims Act liability for an omission without an obligation to disclose, also make questionable an implied certification claim in the Fourth Circuit.")
Here, unlike in Hess, Murray & Sorrenson, Dynamics Research Corp or any other case cited by the government, Gosselin did not make false statements, provide false information or breach any duty in connection with its offered landed rates. In fact, Gosselin and the other signatories openly and explicitly informed the carriers of their agreement to work only under the landed rate. See Gov. Ex. 15, 17 & 23.
Ostensibly recognizing that some element of falsehood is embedded in False Claims Act liability, the Government claims that the Sonthofen Agreement and related activities were "fraudulent" because Gosselin was on notice of its "expectations" of "a fair and competitive ITGBL program free of collusion." See Government's Brief at 4-5; see also Doc. No. 1304, Tr. at 1191:2-5 ("the government expects a fair and competitive program, and if there's any type of collusion that is affecting the program, that is fraud."); see also Doc. No. 1201, Tr. at 29 ("What we have is a government program that requires competition to establish the rates. The requirement that competition among everyone that participates in the program was violated.") This "expectations" theory fails as a matter of law under the facts of this case.
First, the government's "expectations" as to Gosselin's conduct are not embodied in any contractual, regulatory, or statutory term or condition for payment to a carrier or Gosselin's participation in the ITGBL program. Nor is there any evidence that the government otherwise actually conveyed to Gosselin either the "expectations" themselves or that compliance with such expectations was a term or condition of payment or participation. Rather, the government, in effect, contends that Gosselin should have gleaned from the nature of the ITGBL program what those expectations were and that it was required to act consistently with such expectations. At trial, Gosselin disputed that its conduct was inconsistent with what was expected of it; but in any event, False Claims Act liability cannot be premised on the government's post-hoc, general, and vague declarations of what was necessary to preserve the "integrity" of the procurement process.
Civil liability under the False Claims Act can be financially devastating, even where there is no financial harm to the government; and it is incumbent upon the government to be clear as to precisely what is expected of those involved in the procurement process. Such participants should not have to guess at their peril. Had the government wanted to impose as terms and conditions what it now claims were its expectations, the SDDC could have easily required Gosselin, as it did the TSPs, to certify that it had not engaged in any conduct that affected competition. Likewise, the SDDC could have required its carriers to certify more broadly that that their vendors or subcontractors, such as Gosselin, had not engaged in any conduct that could be described as anti-competitive. If the government's expectations were broader than the scope of its Certificate of Independent Pricing (CIP) required of the American carriers, those expectations and conditions should have been made explicit in order to eliminate any uncertainty as to what was expected and what was not, particularly in programs that cross international borders and involve non-American participants.
Second, even if Gosselin could be charged with knowledge of the government's expectations, and failed to conform to them, Gosselin did not make any misrepresentations, expressed or implied, or engage in any deceptions, knowingly or otherwise, with respect to its conduct. As mentioned above, Gosselin openly disclosed the Sonthofen Agreement to carriers. The United States complains that Gosselin did not affirmatively disclose the Sonthofen Agreement to the government but Gosselin was under no obligation to do so and therefore did not engage in fraudulent conduct by failing to disclose something it was under no obligation to disclose. See Harrison, supra, 176 F.3d at 787, n. 8 (referencing the Court's holding in Berge that "there can be no False Claims Act liability for an omission without an obligation to disclose.") See also Doc. No. 1334, Tr. at 101 (Government concedes that Gosselin had no obligation to disclose Sonthofen Agreement).
Third, the government's expectations notwithstanding, there was no evidence from which a jury could reasonably find that Gosselin knew that the Sonthofen Agreement and related activity "caused" the government to enter into ITGBL contracts or that the government, in deciding to award ITGBL contracts, in fact, relied on the absence of any anti-competitive conduct on the part of anyone involved in the ITGBL program, other than the American carriers with whom it contracted and which had filed CIPs. In fact, there was no evidence concerning what the SDDC decision-makers, in fact, considered, assumed, relied on, or thought material with respect to its decision to award ITGBL contracts.
Ultimately, the government premises liability on the notion that any anti-competitive activity within the context of a government procurement necessarily equates to "fraudulent" anti-competitive activity, regardless of whether that conduct was accompanied by any false statements or representations, express or implied. That theory fails as a matter of law and on the facts of this case. As a matter of law, actionable False Claims Act conduct must contain an element of falsehood to be fraudulent. Anti-competitive conduct, in and of itself, does not necessarily; and absent some affirmative misrepresentation, expressed or implied, the government's "expectations" or Gosselin's anti-competitive conduct, standing alone, cannot supply the false statement or fraudulent conduct necessary to impose False Claims Act liability. As a factual matter, and regardless of its anti-competitive effect, which was never established with any specificity, Gosselin's conduct did not have embedded in it an element of deception or false statement that made it inherently fraudulent, as the government, in essence, contends.
Even were the evidence sufficient to show that Gosselin engaged in a fraudulent course of conduct, the evidence is nevertheless insufficient to impose liability since there is no evidence that any specific, identifiable carrier, in fact, presented a claim for payment based on a prime rate that was, in fact, inflated because of Gosselin's alleged conduct. In reaching this conclusion, the Court rejects the government's position that it was sufficient for this purpose to present general testimony, not specific to any particular carrier, prime rate or claim for payment, that higher costs in the supply chain are typically passed onto the government through higher prime rates, that the cost for local agent services and prime rates increased substantially over the previous cycle, or that the signatories to the Sonthofen Agreement thought that their actions had resulted in higher prime rates. In short, the Court rejects the government's position that liability may be imposed without any actual proof that any particular voucher was, in fact, higher than it would have been, absent Gosselin's conduct.
This presentment issue was effectively embedded in the Fourth Circuit's decision in United States ex rel. Nathan v. Takeda, 707 F.3d 451 (4th Cir. 2013), where the Court affirmed the dismissal of a False Claims Act case on the grounds that the relator failed to allege specific false claims but rather relied on inferences that false claims were likely presented to the government for payment. In reaching that decision, the Court specifically rejected as insufficient for the purposes of stating a claim the kinds of inferences that the government relies on in this case. See Id. at 456, 458-61.
In Takeda, a relator alleged that a pharmaceutical company had engaged in a fraudulent marketing scheme, under which sales representatives misled physicians about the proper dosages of a certain drug and made misrepresentations concerning available dosages of that drug. It was further alleged that as a result of this fraudulent marketing scheme, physicians made off-label prescriptions of that drug, not reimbursable under any government program, that were nevertheless presented to the government for payment. The relator failed to identify any specific non-reimbursable prescription that was presented and paid by the government.
The Court first made clear that "... the critical question is whether the defendant caused a false claim to be presented to the government, because liability under the [False Claims] Act attaches only to a claim actually presented to the government for payment." Id. at 456. The Court then rejected the relator's position that it "need only allege the existence of a fraudulent scheme that supports the inference that false claims were presented to the government for payment." Id Rather, the Court adopted the defendant's position that False Claims Act liability requires pleading "facts plausibly alleging that particular, identifiable false claims actually were presented to the government for payment." Id. In short, the Fourth Circuit concluded that there can be no False Claims Act liability "in the absence of an assertion that a specific false claim was presented to the government for payment." Id. "[W]hen a defendant's actions...could have led, but need not necessarily have led, to the submission of false claims, a relator must allege with particularity that specific false claims actually was presented to the government for payment." Id at 457.
It necessarily follows from Takeda for the purposes of this case that to impose liability and damages under the False Claim Act, the United States must prove what is required to be alleged with particularity, viz., that a carrier in fact presented a specific identifiable false claim. That showing in this case requires evidence from which a jury could conclude that a specific, identifiable carrier received payment based on a specific, identifiable prime rate that was in fact higher than it otherwise would have been, absent Gosselin's anticompetitive conduct.
The only evidence presented as to any specific claims was Government's Exhibit 157, which lists all ITGBL moves, by date, channel, GBL number and prime rate, performed between the United States and Germany from IS01 to IW02, and one actual public voucher for payment. Doc. No. 1288, Tr. at 210:4-19; 211:22-25; Def. Ex. 382 (voucher). Based on this evidence, the government argues that each of the 65, 513 moves listed in Exhibit 157 corresponds to a false claim, even though it concedes that there was no evidence that any particular carriers' bid, or any particular prime rate paid by the United States, was inflated because of defendants' conduct. Alternatively, the government argues that at least four hundred false claims were submitted, based on testimony that Cartwright International Van Lines, Inc. submitted at least one hundred claims per rate cycle. As a final alternative, the government argues that at least one false claim must have been submitted, based on the one public voucher actually submitted into evidence.
The evidence is far from sufficient to allow the inference that all prime rates were necessarily inflated because of the Sonthofen Agreement. Approximately thirty percent of the local agents were not signatories to the Sonthofen Agreement, and the government acknowledges that it was possible for carriers to contract for local agent services outside of the landed rate. Even some signatories to the Sonthofen Agreement dealt with carriers directly; and some nonlanded rate carriers, in fact, set some of the prime rates. Landed rates were negotiated with individual carriers and often lowered over the course of a bidding cycle. There was no evidence that all carriers had the same cost structure or bid based on the same profit margins or bidding strategy. Not even all tonnage moved at a prime rate, with some moves, albeit involving a small percentage, priced at rates higher than the prime rate.
The government points to evidence that during the alleged conspiracy period prime rates filed by non-landed rate carriers were comparable to those filed by landed rate carriers. But, without more, this evidence does not allow the reasonable inference that all rates were inflated because of the Sonthofen Agreement; an equally strong inference is that none of the rates were inflated. The government also points to evidence that when Three Star, an Italian local agent with no connection to the Sonthofen Agreement, entered the German market with Cartwright International Van Lines' help in IS03, an event the government points to as leading to the end of the conspiracy period and the effects of the Sonthofen Agreement, Tri Star's prices were equal to or only slightly less than those the existing local agents had been charging; but that evidence raises inferences as harmful as helpful to the government's contentions. Perhaps most dispositive on this issue is testimony from two TSPs that they did not always pass on higher costs through higher prime rates, but rather sometimes made the strategic pricing decision to absorb higher costs by reducing their "margins", i.e., profit.
The uncertainty associated with whether and to what extent prime rates were inflated is underscored by the lack of any testimony, fact or expert, concerning what any particular carrier, or any other participant in the ITGBL procurement process, would have charged or paid in the absence of the Sonthofen Agreement or to what extent any increase in costs caused by the Sonthofen Agreement were passed on to the government through a higher prime rate. The evidence showed that before the alleged conspiracy period the ITGBL prime rates were volatile and changed from year to year, sometimes dramatically. In the cycles immediately preceding IS01, prime rates had decreased to historic low levels. Eleven carriers went bankrupt and exited the market shortly before the alleged conspiracy period, including two major carriers that had set prime rates at levels widely regarded as non-compensatory rates. At least one general agent unconnected to any conspiracy, who acted on behalf of multiple American carriers to secure local agent services, made known within the industry, separate and apart from the Sonthofen Agreement, that beginning with the IS01 cycle, local agents were no longer willing to work for the low wages associated with those historic low prime rates. See Doc. No. 1307, Tr. at 725 (testimony of Lynn Appleton). Even during the alleged conspiracy period, rates had not increased beyond what they had been 5 years earlier.
The alleged cost increases in local agent rates were also three levels of competition below the contractual relationships entered into by the government. There was no evidence, or contention, that the American carriers who submitted bids to the SDDC failed to engage in the level of open and collusion- free competition required under the CIPs. There was no evidence of any collusive anti-competitive conduct as between the American carriers themselves or as between an American carrier and the general agents or landed rate carriers with whom they contracted, including Gosselin, who actively negotiated the landed rate with American carriers individually, and the government does not contend otherwise. The pricing dynamics of the ITGBL market were complex and opaque; and no reasonable jury could trace without completely speculating the extent to which any cost increases caused by Gosselin's alleged anticompetitive conduct filtered through to prime rates. Overall, no reasonable fact finder could conclude that every prime rate used by every carrier to obtain payment during the alleged conspiracy theory was inflated as a result of Gosselin's conduct.
In summary, the government presented no evidence from which the jury could reasonably conclude that any particular prime rate was in fact higher than it would have otherwise been, had Gosselin not entered into the Sonthofen Agreement. For the same reason, the evidence is insufficient to establish 58, 950 false claims, as determined by the jury, or four hundred false claims, based on the Cartwright filings or even one false claim, based on the one public voucher admitted into evidence.
Liability under the False Claims Act requires a false statement or claim that is "material." Harrison, supra, 176 F.3d at 785 ("Liability under each of the provisions of the False Claim Act is subject to the further, judicially-imposed, requirement that the false statement or claim be material."). The issue of materiality is to be decided by the Court. United States ex. Rel. Berge v. Board of Trustees of the Univ. of Ala., 104 F.3d 1453, 1459-60 (4th Cir. 1997). The test for materiality is "whether the false statement has a natural tendency to influence agency action or is capable of influencing agency action." Id. at 1460 (internal citation omitted).
The materiality requirement underscores the centrality of a false statement and a specific false claim in order to impose False Claims Act liability, and the mismatch between the government's theory of liability and the elements of and legal requirements for a successful claim under the False Claims Act. The materiality requirement must be satisfied with reference to a false statement or a false aspect of a claim that would influence the government's decision whether to pay the claim. See cases cited in Plaintiff's Post-Trial Memorandum Addressing Materiality, Doc. No 1323 at 2-3, including United States ex rel. Longhi v. Lithium Power Tech, Inc., supra, 575 F.3d at 470 ("All that is required under the test for materiality... is that the false or fraudulent statements have the potential to influence the government."); United States ex rel. Harrison v. Savannah River Co., 352 F.3d 908, 916-917 (8th Cir. 2012) (materiality "focuses on the potential effect of the false statement when it is made, not on the actual effect of the false statement when it is discovered.") (emphasis added in all quotes); see also Harrison, supra, 176 F.3d at 788 ("any time a false statement is made in a transaction involving a call on the U.S. fisc, False Claims Act liability may attach.").
Here, the government seeks to satisfy the materiality requirement, not with reference to a false statement or a false aspect of a specific claim, but rather based on Gosselin's anti-competitive conduct, had it been disclosed, without more. See Plaintiff's Post-Trial Memorandum Addressing Materiality, Doc. No. 1323 at 3 ("Here, the materiality analysis is focused on whether Gosselin's collusive conduct, which undermined the integrity of the competitive bidding process and resulted in inflated rates, was material. Based on the evidence adduced at trial, the Court should conclude that Gosselin's anti-competitive, rate-inflation conduct had the potential to influence DOD decision-making.") But the government made the decision not to require the disclosure of such conduct or to obtain representations or certifications that allowed it to assume such conduct did not exist; and in this sense, the government itself has defined, though the scope of its required certifications, what competition related information was material for its purposes. Likewise, as discussed as above, the government has not identified any particular claim whose prime rate was, in fact, inflated because of any Gosselin conduct.
Based on the evidence at trial, and the reasonable inferences to be drawn from that evidence, the Court finds, in its capacity as the trier of fact on the issue of materiality, that Gosselin did not engage in any conduct that was "material" for the purposes of imposing liability under the False Claims Act.
To be sure, the False Claims Act is to be construed broadly. It "is intended to reach all types of fraud, without qualification, that might result in financial loss to the Government..." and it is not to be subjected to a "rigid, restrictive reading." Harrison, 176 F.3d at 788, quoting United States v. Neifert-White Co., 390 U.S. 228, 232(1968). But it must also be remembered that the False Claims Act is not an anti-trust statute. It is not even a general, all purpose anti-fraud statute. See Id. at 785, citing United States v. McNinch, 356 U.S. 595, 599 ("The Supreme Court has cautioned that the False Claims Act was not designed to punish every type of fraud committed upon the government."); see also Allison Engine Co. v. United States ex rel. Sanders, 553 U.S. 662, 672 (2008) (False Claims Act is not "all-purpose antifraud statute"). Rather, it is a fraud-based statute that requires deception or deceit in connection with obtaining a specific, identifiable claim for payment under a government contract. The government's theory of liability extends the reach of that statute far beyond anything previously recognized as within its reach. It would allow liability to be imposed on anyone who supplies goods or services, at any level, within the supply chain or manufacturing process of any product or service whose end customer is the government. It is completely standardless and otherwise untethered to any misrepresentation or false statement or term or condition for payment of a claim or participation in a government procurement program.
For the above reasons, the Court finds and concludes that the evidence is insufficient to impose liability under either the First Cause of Action (False Claims Act) or the Third Cause of Action (conspiracy to defraud the United States) of the Government's Complaint in Intervention. See Doc. No. 110. The Court therefore will therefore set aside the jury's verdict and enter judgment in favor of the Gosselin and Smet and against the government.
In order to facilitate complete appellate review, the Court has also considered whether the evidence was sufficient to sustain the jury award of compensatory damages in the amount of $33.1 million.
The only evidence presented as to damages was the testimony of the government's damages expert, Dr. Robert Marshall, an economist, who opined that the government paid approximately $41.5 million more during the alleged conspiracy period, 2001-2002, than what the government would have paid absent Gosselin's anticompetitive behavior. His opinion was based on the prime rates that his economic model predicted for the alleged conspiracy period based on the cost and demand factors that best correlated to and predicted prime rates established during the a 24 year base period that both preceded (1979-2000) and post-dated (2005-2007) the conspiracy period, referred to as the benchmark period. In constructing his model, he considered the pricing of 18 "explanatory variables" as the relevant cost and demand factors that affect prime rates. He then reduced those variables to those that best positively correlated to the prime rates that were set during the benchmark period. Because a single prime rate operated during the entire 6 month summer cycle beginning April 1 and for the entire six month period winter cycle beginning October 1, but the cost and demand factors change month-to-month, Dr. Marshall "estimated six separate regressions" (one for each month of the rate cycle) and then averaged the results to create a but-for westbound price index. Doc. No. 1307, Tr. at 748:23-749:12; 750:7-25; 766:1-16. Dr. Marshall then created an eastbound price index based on the statistical relationship between the eastbound and westbound price indices in the benchmark period. Dr. Marshall calculated damages for the alleged conspiracy period by subtracting the but-for prime rates he calculated from the actual prime rates for IS01 to 1W02 and multiplying that figure by the amount of tonnage moved during that period. Under his analysis, damages were based on the difference between the prime rates predicted for the conspiracy period and the prime rates actually set during the conspiracy period.
Having now considered that testimony based on all the evidence in the case, the Court concludes that Dr. Marshall's testimony should have been excluded under Daubert. For that reason, the Court concludes that even were the evidence sufficient to impose liability, without Dr. Marshall's testimony, the evidence was insufficient to award any damages. The Court therefore sets aside the jury's damage award as well.
Under Daubert, the trial court is under an obligation to ensure that expert testimony "is not only relevant, but reliable." Daubert, 509 U.S. at 589; see also Kumho Tire Co., Ltd v. Carmichael, 526 U.S. 137 (1999) (extending Daubert's holding regarding the admissibility of scientific testimony to all other expert testimony). In fulfilling this obligation, the Court must determine "whether the reasoning or methodology underlying the testimony is scientifically valid and... whether that reasoning or methodology properly can be applied to the facts in issue." Id at 592-93. While a regression analysis that uses fewer than "all measurable variables" may be sufficiently reliable to pass muster under Daubert, the analysis must take account of the major factors. See Smith v. Va. Commonwealth Univ., 84 F.3d 672, 676 (4th Cir. 1996) ( en banc ).
Overall, defendants object to Dr. Marshall's testimony on the grounds that this model did not reliably predict what prime rates would have been during the conspiracy period in the absence of any collusion. In support of this position, defendants argue, inter alia, that his model underestimated certain prime rates during the benchmark period, particularly spikes in prime rates that followed deep multiple cycle declines, that his model's selection of explanatory variables is economically incoherent, that the model did not include or adequately account for major factors that affected prime rates, including specifically ocean rates that constitute approximately 30% of the ITGBL prime rate and that the six regression equations used to average prime rates predict widely different results, some of which have no relationship to the actual prime rates during the benchmark period and therefore a model that simply averages those divergent results cannot be deemed to be reliable. See Doc. No. 1223 at 3-5.
The Court does not question Dr. Marshall's high level of expertise or the well-recognized and accepted underlying principles of multiple regression analysis he used, which can, in appropriate cases, reliably determine the effect of independent, or explanatory variables, on a dependent variable, as Dr. Marshall purports to do in this case. Rather, at the most basic level, the Court questions whether, given the available data, regression analysis, as a methodology, can be reliably applied to predict prices set under the opaque, unusual and complex price setting mechanism for ITGBL prime rates. Even if the methodology was suitable for this case, the Court could not determine as to certain aspects of the model whether Dr. Marshall used data necessary to make it reliable. Finally, the Court concludes that the multiple regression analysis used did not adequately account for certain major factors affecting ITGBL prime rates during the benchmark period, specifically ocean transport rates, and the Court cannot therefore conclude that the model was sufficiently reliable to predict prime rates during the alleged conspiracy period. See Smith v. Virginia Commonwealth Univ., supra, 84 F.3d at 681 (Luttig, J., concurring in part) ("the Court [in Bazemore v. Friday, 478 U.S. 385 (1986)] left no question whatsoever that a regression analysis that does not control for at least major variables may be altogether inadmissible."); Eastland v. Tennessee Valley Auth., 704 F.2d 613, 621 (11th Cir.) reh'g denied and opinion modified, 714 F.2d 1066 (11th Cir. 1983) ("The probative value of a multiple regression analysis depends in part upon: (1) the inclusion of all the major variables likely to have a large effect on the dependent variables; and (2) the validity of the assumption that the remaining effects (the influences included in the random disturbance term) are not correlated with the independent variables included.").
Central to determining whether regression analysis is a suitable methodology under the facts of this case is that ITGBL prime rates were set under a system that had distinctive, if not unique, aspects that encouraged the bidding, and acceptance, of prime rates by carriers based on, it would appear, considerations other than those that were captured in the cost and demand factors Dr. Marshall used to create his model, or could be captured by any determinable cost and demand factors. For example, there was no evidence of any established or determinable "market prices" for the services provided by the TSPs, external to the ITGBL process itself, that carriers could accept or reject. Rather, a prime rate for any particular channel was effectively set, not directly by broad market forces, but by one buyer, the U.S. government, and one seller, the TSP that filed the lowest rate for a particular channel. The evidence at trial established that carriers would adopt various strategies to take advantage of this aspect of the ITGBL bidding process. Some would submit bids intended to establish the prime rate while others would file an intentionally high bid only for the purpose of allowing it to participate in the me-too bidding.
Among those carriers who hoped to establish prime rates during the benchmark period, particularly in the several cycles immediately before the alleged conspiracy period, were TSPs that filed bids that set prime rates that were considered "non-compensatory" by certain other carriers, that is, prime rates that did not cover all the internal costs of certain other carriers. These other carriers therefore needed to make the economic decision whether to me-too a prime rate that covered some, but not all of their costs or simply not obtain any revenue from the ITGBL business. Although the government required carriers to file bid that were "compensatory, " there was no evidence that the government selected prime rates on any basis other than the lowest price. See Doc. No. 1288, Tr. at 205:13-19; 223:12-19 (testimony of D. Martinez).
As it turned out, a number of carriers went bankrupt in the cycles leading up to the alleged conspiracy period, including some who had established prime rates, strongly suggesting that the prime rates they set were non-compensatory in the sense that the rates were lower than needed to cover all their fixed and operating costs and remain in business. In fact, from 1999-2000, immediately preceding the period of alleged collusion, eleven carriers went bankrupt. See Doc. No. 1307, Tr. at 819; Def. Ex. 731. Two of those carriers, A-Olympic and Emerald City, were significant prime rate setters. For example, in IS98, Emerald City set primes in 38.9% of the eighteen most volume-heavy westbound channels, which channels accounted for 75% of the total tonnage moved; Emerald City and A-Olympic collectively filed prime rates in 44.4% of those eighteen channels. Similarly, in IS99, Emerald City set primes in one third of those channels. A-Olympic and Emerald City set primes in 63.5% of all westbound channels in IS98 and 50% of all westbound channels in IS99. See generally Doc. No.1307, Tr. at 819-22 and Doc. No. 1331.
Based on this evidence, it appears that a single carrier had the ability to set a prime rate that did not necessarily have any particular relationship to its internal costs or the cost and demand factors used by Dr. Marshall. Likewise, a me-too carrier had the ability to accept a prime rate that did not have any particular relationship to its internal costs or cost and demand factors. The prime rate was not reflective of some broader market price established external to any particular carrier, but rather specific to a company's bidding strategy based on the where it was willing to peg its prime rate bid relative to its costs. For example, it could very well have been the case, and in light of the bankruptcies that did occur, likely was the case, that certain prime rate setters were setting rates below their costs and below the costs of some of the carriers that me-too'd that prime rate in order to get revenue that would cover some but not all their costs. In effect, the selected explanatory values best correlated to prime rates during the benchmark period that were set not just by determinable cost and demand factors but also the particular decision-making calculus of a prime rate setter relative to his internal costs. For that reason, the substance of that particular decision making by a particular prime rate setting carrier in a particular cycle may be different than the decision making calculus of a subsequent prime rate setter relative to those same cost and demand factors. Some of the cost and demand factors that the model used were no doubt influencing those decisions, and it was adequately demonstrated that certain combinations of explanatory variables best correlated to specific prime rates, but without knowing how a prime rate related to the internal costs of a prime rate setter, it would seem impossible to reliably predict future prime rates unless it were assumed that the future prime rate setter pegged a prime rate at the same level relative to its internal costs. There was no evidence that the data needed to make those assessments were considered or even available.
Dr. Marshall dismissed these concerns on the grounds that it was reasonable to assume that carriers acted rationally and that they would not me-too a prime rate unless it was "compensatory" in the sense that it "maximized their profits." But that explanation does not address that aspect of a price setting mechanism that sets prices based solely based on what a particular carrier decides his costs allow, if only over the short term. Specifically, it does not explain how a model that correlates to a prime rate set during a benchmark period by a specific carrier based on that particular carrier's bidding strategy relative to his internal costs can be a reliable predictor of a prime rate set by a different carrier during the alleged conspiracy period, particularly after the earlier prime rate setter has left the market. There may be economically and analytically sound answers that dispose of these issues, but the evidence and the explanations provided at trial were insufficient for the Court to conclude that the multiple regression analysis used in this case was a suitable and reliable methodology to predict what prime rates would have been during the alleged conspiracy period, absent any anticompetitive conduct on the part of Gosselin.
Even were the Court to accept the multiple regression analysis as a suitable methodology, the Court cannot conclude that its use in this case accounted for all of the major factors affecting the setting of prime rates. First, as an initial observation, it is not at all clear whether Dr. Marshall used the most appropriate data to find the explanatory variables that best correlated with prime rates during the benchmark period. In that regard, because the data pertaining to any particular explanatory variables varied from month to month, sometimes substantially, Dr. Marshall averaged results of six separate regression equations, one for each month of any particular cycle. Through computerized calculations, each of those six regression equations incorporated that combination of explanatory values (out of an array of 18 possible explanatory values) that best correlated to the prime rate that had been set for that particular cycle. The values attributed to the explanatory variables for that purpose were based on the data corresponding to the month in the six month cycle for which the equation was constructed. As mentioned above, because the data used for the 18 explanatory variables changed from month to month, the combination of explanatory variables that best predicted the prime rate for that benchmark period cycle changed from month to month, and the prime rates predicted by each of those six regression equations often differed quite dramatically, which necessitated their averaging. But prime rates for the IS cycle beginning on April 1 were submitted by the carriers in November of the preceding year, followed by me-too bids in December; and the prime rates for the IW cycle, beginning on October 1, were submitted in May, with me-too bids in June. See Expert Report of Robert C. Marshall, Ph.d., Doc. No. 922-2 at 9. Completely unexplained at trial is whether, and if so, why, the model was based on regression equations that used data that postdated a carrier's decision making by at least 4 months, without including data that was more proximate to the time period when carriers were in fact making rate decisions. It is not clear whether Dr. Marshall used any data from the period when bids were prepared, but to the extent that such data were not used and given the sometimes dramatic differences in predicted prime rates over the six month cycle, it would appear that prime rates predicted by data corresponding to economic conditions when the carriers actually prepared and submitted their bids would likely differ substantially from prime rates predicted by data corresponding solely to economic conditions during the cycle itself, several months later.
In any event, Dr. Marshall's model also did not adequately account for ocean shipping rates, which account for approximately 30% of the TSPs' costs. Ocean shipping rates are fixed by the Trans-Atlantic America Flag Liner Operators ("TAAFLO") and account for about thirty percent of the cost of a move. Doc. No. 1307, Tr. at 835. Dr. Marshall did not include TAAFLO rates as a potentially useful explanatory variable because he was not given TAAFLO rate data predating 1995. While Dr. Marshall testified that the variables he included served as proxies for TAAFLO rates, he also did not include a variable that directly represented ocean transport costs; and he was unable to identify the combination of specific variables that served as proxies, or how exactly they did so. Doc. No. 1295, Tr. at 859 (agreeing that it is difficult to say which variables would have accounted for an increase in ocean rates "because the variability in the variables is a very complicated process in terms of how they all are moving together to account for various effects"). Dr. Marshall's only explanation for his confidence that ocean rates were accounted for through proxies is that the "few" TAAFLO rates he had were positively correlated with the fuel variables, and that the westbound premium in the TAAFLO rates was positively correlated with the trade deficit, that is, the fact that more traffic moved west than east. Doc. No. 1307, Tr. at 835-36; Doc. No. 1295, Tr. at 859-60. However, he did not perform any analysis to test whether the factors he employed were in fact capable of predicting changes in ocean rates.
Other aspects of the model also raise substantial doubts as to its reliability. For example, Dr. Marshall's model predicted during the benchmark period substantially lower prime rates than the actual prime rates that spiked after the kind of long decline in prime rates that preceded the cycles constituting the alleged conspiracy period. See, e.g., Def. Ex. 730; Doc. No. 1307 at 780-781(For IS87, Dr. Marshall predicted an index of 64.40, when the actual index was 70.75) Likewise, the model predicted substantially lower prime rates than those that were actually set during the four cycles after the alleged conspiracy period ended. See Def. Ex. 730. Dr. Marshall dismissed any criticisms based on prime rates set during these cycles on the grounds that during this period, prime rates reflected the "lingering effects" of the conspiracy. But he admitted that he did not analyze whether there were, in fact, any lingering effects and excluded that period because it "may have" or "potentially" involved lingering effects, relying generally on his experience in cartel pricing cases. See Doc. No. 1307 at 762, 784.
Similar concerns exist with respect to Dr. Marshall's opinion that the model adequately accounted for the effect on prime rates by the reduction in available carriers, including some of the largest, because of bankruptcies that occurred in the cycles shortly before the conspiracy period. Dr. Marshall testified that the effect of these bankruptcies had been accounted for "through the number of factors of the eighteen variables that account for fluctuations in the economy." Doc. No. 1307, Tr. at 763. Likewise, he testified that the eighteen variables he chose "are accounting for fluctuations in the prime rate, and those variables are picking up the changes in the economic activity that's going on throughout this time period." Doc. No. 1295, Tr. at 934. He also testified that there was "nothing systemic" about the bankruptcies that had occurred over the thirty-year period preceding the collusion period, in that prices sometimes went up after a bankruptcy, sometimes went down, and sometimes remained essentially the same. Doc. No. 1307, Tr. at 763. Further, Dr. Marshall testified, he could not include a factor directly accounting for carrier bankruptcies because he could not include in the model any factors that might be tainted by the potential effects of the conspiracy, and to the extent bankruptcies occurred during the collusion period they could have been affected. Id. at 763-64, 823. But the Court found persuasive the testimony of defendants' expert Dr. Keith Ugone on this issue, who testified that one would expect the large number of bankruptcies that occurred shortly before the conspiracy period to drive up prices; and nothing in Dr. Marshall's explanations satisfies the Court that the model adequately accounted for the upward pressure on prime rates one would expect.
Overall, the Court is left with a firm conviction that the model is not reliably predictive of rates within the alleged conspiracy period and should have been excluded by the Court under Daubert. Since the government did not produce at trial any other evidence that would allow the jury to calculate damages, the Court must grant the defendants' motion for judgment as a matter of law as to damages and set aside its damage award.
For the reasons discussed above, the Court grants defendants' motion for judgment as a matter of law as to liability, the number of false claims, and damages and enters judgment in favor of defendants as to the ITGBL claims set forth in the First and Third Cause of Action set forth in the United States' Complaint in Intervention [Doc. No. 110] and also conditionally grants a new trial on all issues pertaining to those claims if the Court's judgment is vacated or reversed on appeal.
An appropriate Order will issue.