In the United States, a cartel can be prosecuted criminally by the federal government and civilly by the federal and state governments and private parties. A cartel (or conspiracy) is an understanding or agreement between two or more independent businesses to fix prices, restrict output, allocate markets, or otherwise limit competition.
The central law in the USA for challenging cartel conduct is Section 1 of the Sherman Act, 15 U.S.C. §§ 1 et seq. At the federal level, the Department of Justice’s (DOJ's) Antitrust Division and the Federal Trade Commission (FTC) enforce the Sherman Act.
Section 4 of the Clayton Act, 15 U.S.C. § 15(a), allows private plaintiffs to enforce Section 1 of the Sherman Act. Under Section 4 of the Clayton Act, successful plaintiffs are entitled to treble damages, as well as fees and costs.
There are two relevant federal public enforcement agencies: the Antitrust Division of the DOJ and the FTC. They have overlapping authority.
The Antitrust Division has the exclusive authority to bring criminal actions for violations of the Sherman Act. When it does so, the criminal prosecution addresses an agreement between/among horizontal competitors not to compete by, for example, agreeing to fix prices, allocate markets, or rig bids. The Antitrust Division can seek criminal sanctions, including fines and/or imprisonment, for such conduct. The Antitrust Division can also pursue civil actions against cartels, including a challenge to unilateral conduct, mergers and other less-co-ordinated behaviour, and seek civil remedies such as an injunction.
The FTC brings civil actions to enforce the FTC Act, 15 U.S.C. §§ 41-58, which prohibits unfair methods of competition. The FTC often challenges conduct that does not meet all of the elements of a Section 1 Sherman Act violation, such as invitations to collude.
State attorneys general also play a role in enforcing these laws. States may bring federal antitrust suits on behalf of individuals residing in their states as well as sue to enforce their own antitrust laws.
There are no conditions precedent to filing private actions in the USA challenging cartel conduct. Private suits often follow parallel DOJ criminal investigations. A single fact pattern may lead to class actions by a “direct purchaser” class, an “indirect purchaser class” and various direct action plaintiffs who sue on their own behalf. Pursuant to 28 U.S.C. § 1407, the Judicial Panel on Multidistrict Litigation may centralise these cases before a single federal judge for pre-trial proceedings.
Section 1 of the Sherman Act explains that “every contract, combination... or conspiracy, in restraint of trade” is illegal. The definition of these terms is judge-made. Not every restraint of trade is unlawful because nearly every contract restrains trade in some way. Only “unreasonable” restraints of trade are unlawful.
American courts have developed three tests to determine whether economic activity unreasonably restrains trade: the per se test, the “quick look” test and the rule of reason test.
Certain economic activity is per se unlawful because, regardless of the actual effects or intent, it typically produces anti-competitive effects. Cartel agreements among competitors – such as those to fix prices (or reduce output), allocate customers, boycotts, or rig bids – are evaluated under the per se standard. In a criminal cartel case, the government need merely prove the existence of the agreement involving an effect on interstate commerce (the latter almost never a disputed issue). In a civil cartel case, the plaintiff must prove the existence of the agreement, antitrust injury (eg, directly purchasing a price-fixed product or service) and damages caused by the cartel.
The “rule of reason” test evaluates whether a challenged restraint imposed by a firm with market power has anti-competitive effects (often in a relevant market depending upon the nature of the restraint; eg, horizontal or vertical) as measured by, for example, increased prices, reduced output, or excluded rivals. If a plaintiff presents this proof, then the defendant must present evidence that the restraint is pro-competitive; and if so, the plaintiff must demonstrate that the anti-competitive effects of the restraint outweigh the claimed pro-competitive benefits. The rule of reason test requires a rigorous evaluation of the market and the effects of the challenged restraint.
Certain conduct outside the per se boundaries have such anti-competitive potential that absent proof of a pro-competitive justification, it can be prohibited after a “quick look” without detailed market analysis.
In the USA, certain industries are exempt from the laws prohibiting cartels, including, in whole or part:
Other congressional exemptions exist as well.
For private suits, the Clayton Act has a four-year statute of limitations. In cases of continuing unlawful acts, such as a conspiracy to fix prices, the clock starts anew for each sale at an artificially inflated price, and the plaintiff has four years from the date of that sale to bring a timely action.
There are three main exceptions that pause the running of the statute of limitations:
In criminal cartel cases, the statute of limitations is five years. For statute of limitations purposes, a conspiracy does not end until the conspiracy succeeds or is abandoned.
The Foreign Trade Antitrust Improvements Act (FTAIA) creates a general rule that the Sherman Act does not apply to conduct involving trade or commerce with foreign nations. However, the FTAIA has two major exceptions: the “import exclusion” exception and the “domestic effects” exception.
Under the “import exclusion” exception, the Sherman Act applies to conduct related to imports. For example, if a foreign defendant conspires abroad and sells a price-fixed product directly to companies in the USA, then the defendant and its conspirators are subject to the Sherman Act.
Under the second exception, the Sherman Act applies to foreign commerce if it has a “direct, substantial, and reasonably foreseeable effect” on US trade, and that effect “gives rise to” the plaintiff’s claim.
In order for a foreign company to be a defendant in a cartel case, the court must have subject matter jurisdiction over the case and personal jurisdiction over the defendant. A court will exercise personal jurisdiction over a foreign defendant only if that defendant has sufficient “minimum contacts” with the USA. There is a substantial body of law analysing the requirements for exercising personal jurisdiction over foreign defendants.
Federal courts may refrain from hearing cases under a judge-made doctrine called “international comity.” Under this doctrine – which is vague and often difficult to apply – courts may refrain from exercising jurisdiction in cases in which other sovereigns may have greater interests.
Courts usually look to a non-exhaustive list of factors relating to the relationship between US and foreign courts to determine whether to refrain from hearing a case.
At both the Antitrust Division and the FTC, an initial investigation may start in a variety of ways. The agencies will evaluate confidential complaints from market participants, typically customers or competitors aggrieved by some potentially illegal conduct. In other instances, an investigation will originate from academic articles evaluating potentially anti-competitive conduct or from informal discussions with other law enforcement or regulatory agencies.
The agencies often begin an investigation by interviewing market participants and requesting voluntary productions of documents or data from firms with information relevant to an evaluation of the conduct. Agency staff also may meet with the potential target(s) and their counsel to discuss the business justification for the conduct at issue. At this early stage, potential targets may voluntarily elect to disclose business information, typically, when they believe the information will allow staff to recommend against opening a formal investigation.
Unannounced searches of premises by enforcement authorities are colloquially referred to as “dawn raids.” In the USA., dawn raids are conducted by the Antitrust Division, “armed” with subpoenas, in conjunction with the Federal Bureau of Investigation (FBI) and at times in a co-ordinated effort with international authorities. For example, simultaneous raids have occurred at the offices of multinational firms with offices in the USA and Europe, Canada, South Korea, Japan and elsewhere in order to maintain the element of surprise. The FBI may serve business executives with warrants to search their personal residences as part of these raids.
In the USA, when the government has a valid search warrant, it does not have to wait for the arrival of counsel before executing the search, although FBI agents may agree to wait. Any statements that employees make to the agents may be attributed to the company. Individual employees have the right not to speak to agents during the search.
Dawn raids have been particularly productive in uncovering evidence of cartel activity, which typically occurs in secret and can be hard to detect and prosecute.
Before conducting a “raid”, the enforcement authority must obtain a search warrant from a judge upon a showing of probable cause that a criminal violation has occurred. The search warrant request must be supported by sworn statements, as well as detail the place and persons and things to be seized as provided under the Fourth Amendment of the US Constitution.
The Fourth Amendment and Rule 41 of the Federal Rules of Criminal Procedure impose limits on search warrants. A search warrant must be reasonable under the circumstances, including the location, scope and type of information to be seized. A proper search warrant will describe the location (eg, specific office) and set forth the information to be seized (eg, documents relating to the pricing of “X” product, including all communications between the target company and its competitor). The government must execute the warrant within 14 days of issuance, and searches must be conducted during daylight hours unless expressly authorised otherwise. An investigating officer must prepare and verify an inventory of property seized in the presence of another officer or at least one credible person.
In some instances, a search warrant permits authorities to search incidental locations and detain occupants of the premises being searched. Based on the willingness of individuals to co-operate with authorities, an individual may be asked to contact competitors and have those communications recorded. Under certain conditions, Rule 41 allows authorities to remotely access, search and seize electronic media outside the jurisdiction of the warrant. Rule 41 further provides that electronically stored information (ESI) may be seized and subject to later review by authorities.
Searches and seizures are not without boundaries. If authorities exceed the scope of the warrant – eg, by searching for information not delineated in the warrant – the seized materials may be inadmissible in later criminal proceedings.
“Spoliation” refers to the act of destroying evidence or allowing it to be destroyed. Depending on the nature of the conduct and jurisdiction involved, spoliation can have significant consequences.
In the criminal context, it can result in charges of evidence tampering or obstruction of justice separate and apart from any charges based on anti-competitive conduct. The key element of these chargeable crimes is the element of intent – that is, knowingly affecting the outcome of a law enforcement investigation.
In the civil context, sanctions for a party’s failure to preserve evidence, depending on whether the party acted in bad faith and the prejudice to the other party, can include:
Officers and employees of the search target cannot be required to respond to agents’ substantive questions during the raid, and if they do choose to speak with the agents, then they have a right to counsel.
The search warrant defines the scope of what agents can search and seize. If presented with a valid search warrant, then the company cannot stop the agents from executing the warrant. If permitted, someone from the company can follow and observe the authorities throughout the search, keeping notes of what the agents take and say. The employee should respond to agents’ administrative questions, but they have no obligation to respond to substantive questions.
All officers and employees have a right to counsel, and all employees have the right to decline to be interviewed.
In connection with a dawn raid or antitrust investigation, counsel for the company likely will be confronted with the decision of whether to represent an officer or employee of the company as well. The most significant concern relates to whether a conflict of interest exists between the company and the officer or employee that would preclude the attorney’s joint representation. It is often difficult to determine early in the investigation whether such a conflict exists.
Counsel should proceed with extreme caution in such situations and avoid the potential for a conflict if the attorney has any reason to believe that a current or former employee may possess facts or assert positions that could conflict with the company’s position. In most circumstances, counsel can represent only the company. See 2.12 Attorney-Client Privilege.
As noted above, the interests of a current or former employee of a company that is the target of a dawn raid or antitrust investigation may conflict with the company’s interests and require the individual to obtain his or her own counsel. For example, an employee who has participated in criminal conduct that is the subject of the investigation likely has legal concerns of his or her own and may be willing to co-operate with an enforcement authority against the company. That employee’s interests are clearly in conflict with the interests of the company, and counsel could not ethically represent both the company and the employee in that situation.
In the event of a dawn raid, counsel should engage in the investigation process as early as possible, including participating in investigative interviews and, depending on the circumstances, assisting with securing counsel for employees prior to interviews. In addition, counsel may be involved in evaluating the reasonableness and scope of an authorisation or warrant and provide guidance to the client during the seizure process. If the enforcement authorities attempt to seize privileged, confidential or trade secret information, then counsel may be instrumental in preventing such seizure or ensuring that the materials are properly sequestered until the confidentiality concerns are resolved.
Counsel should also oversee the inventory of all materials and information seized and consider conducting post-seizure interviews of employees who witnessed or participated in the seizure. During this initial phase, counsel’s objective should be to develop a complete understanding of the scope of the seizure and who said what to whom.
Following a dawn raid, US enforcement authorities typically use grand jury proceedings to continue their investigation. As part of that process, authorities issue subpoenas for both documents and witness testimony. Based on this evidence, the grand jury may issue criminal indictments against businesses and individuals for engaging in anti-competitive activity.
Information presented before grand juries is confidential and not discoverable absent a court order in a civil proceeding, pursuant to the rules of criminal procedure, or waiver by the person who testified. There is no prohibition against speaking with witnesses who testified before the grand jury, assuming they agree to be interviewed.
Law enforcement agencies can ask to speak with potential witnesses about matters under investigation. Victims of cartel behaviour may voluntarily co-operate with law enforcement agencies and provide price and bid information about, and communications with, cartelists that occurred in the ordinary course of business.
Beyond voluntary co-operation, grand jury subpoenas are an effective means of compelling testimony under penalty of perjury from those individuals with relevant knowledge of an illegal cartel.
Enforcement authorities also use grants of immunity to companies and people in an effort to obtain valuable information and testimony in cartel investigations.
In response to a request for production under Rule 34 of the Federal Rules of Civil Procedure, a responding party must produce relevant, non-privileged documents or ESI that is in the party’s “possession, custody, or control”, including information that the responding party has the legal right to obtain, such as by contract or by virtue of the nature of the relationship (eg, accountant/client, corporation/officer or customer/bank).
In 2018, the US Congress passed the Clarifying Lawful Overseas Use of Data Act (the “CLOUD Act”). This law allows federal law enforcement to compel US data and communication companies to provide requested data stored for a customer or subscriber on any server that they own and operate, whether stored in the USA or a foreign country. A company can challenge a request that it believes violates the privacy rights of the foreign country where the data is stored.
The attorney-client privilege is one of the most litigated issues in US jurisprudence. The privilege belongs to the client and, simply stated, is the client’s right to refuse to disclose, and to prevent others from disclosing, confidential communications between the client and the attorney. For the privilege to attach, a communication must be made to an attorney for the purpose of securing legal advice. If the communication was made in the presence of individuals who were neither attorneys nor fellow clients, or was disclosed to others outside their circle, the privilege is generally waived. Similarly, if it was made for business purposes, or for the purpose of committing a crime or fraud (even if the attorney is not aware of the purpose), then the privilege does not protect the communication.
In cases brought under federal question jurisdiction (such as federal antitrust cases), the courts apply federal common law to resolve issues of attorney-client privilege. In diversity cases, the court will look to the law of the relevant state to determine questions of privilege.
Under Upjohn v United States, 449 U.S. 383 (1981), the attorney-client privilege extends to communications between corporate counsel and a wide group of corporate employees. When interviewing corporate employees, counsel are strongly advised to give “Upjohn warnings” to the employees explaining that the privilege belongs solely to the company, which may choose to protect or waive the privilege.
In the in-house counsel context, the “client” is the legal corporate entity and not the entity’s individual officers, directors, shareholders, or employees. The subject matter of the communication is particularly important in determining whether the attorney-client privilege applies to a corporate communication. While legal advice need not be the only reason for the communication or why the attorney is included in it, legal advice must be one of the primary reasons for the communication. Merely copying an attorney on a communication does not render it privileged.
In addition to the attorney-client privilege, the work product doctrine protects material prepared in anticipation of litigation from discovery. The work product protection extends to materials prepared by persons other than the attorney so long as they were prepared with an eye toward the realistic possibility of impending litigation. In the cartel context, this might include signed statements executed by individuals who are no longer employed by the subject company. Unlike the attorney-client privilege, the work product protection may be overcome by a showing of necessity.
Communications between parties and between their counsel may be protected from disclosure if the parties have entered into a joint defence or common interest agreement. Because cartels by definition involve multiple parties, this protection often arises when the defendants are contesting the assertion that a cartel even existed. Parties who intend to assert that their communications are protected by a common interest agreement should reduce their agreement to writing executed by all participants.
Because cartel activities have potential criminal sanctions, absent a grant of immunity, the Fifth Amendment privilege against self-incrimination is always in play, even in civil proceedings (in which an individual (but not a company) may assert his or her Fifth Amendment right not to answer questions when testifying). An exception may exist if the statute of limitations has run on a potential criminal violation. A witness can assert the Fifth Amendment privilege only to protect against disclosing information that the witness reasonably believes could be used, or could lead to other evidence that could be used, in a criminal prosecution against the witness. “Taking the Fifth” has consequences. In a civil case, it may lead to an “adverse inference” instruction to the jury, which allows them to presume that the answers, had they been given, would have been harmful to the witness’s interests. In addition, a party who invokes the Fifth Amendment likely will be barred from offering evidence on that issue at trial.
While the Fifth Amendment may be used to protect information in the hands of the target, a target entity cannot assert it to protect information in a third party’s possession or control. Of course, third parties may be able to assert the other privileges described above.
It is worth noting that a witness who fears prosecution in a foreign country, as opposed to the USA, cannot invoke the Fifth Amendment.
Whether and to what extent a company should resist a request for information from a government enforcement agency depends heavily on the context and scope of the request. Potential victims of cartel activity have an incentive to co-operate with the government. A request may be so broad and intrusive, however, that it would be unreasonably time-consuming or disrupt the everyday business activity of the recipient by directing resources away from business-related tasks for prolonged periods.
In some cases, the enforcement agency’s request may seek information that is protected from disclosure (eg, information subject to contractual confidentiality provisions, personal or private customer information). In these situations, a party may be obligated to resist the request for certain information altogether or resist producing the information without taking affirmative steps to protect it, such as redaction.
In producing materials to enforcement agencies, individuals and entities may be able to negotiate confidentiality agreements with the enforcement authorities. Typically, such agreements provide for the return or destruction of materials at the conclusion of the investigation.
Defence counsel employ various strategies when a client is under investigation for participating in a cartel. These strategies depend on the client’s level of participation, how extensive and to what level company employees engaged in the cartel activities, and the amount of proof that the enforcement agency has at the time counsel is retained.
One universal tactic is to attempt to shrink the timeline of the client’s participation in the cartel by conceding limited participation in a set timeframe and denying any participation outside that timeframe. An upfront concession, coupled with an attempt to negotiate a reasonable penalty, can discourage a scorched earth investigation.
Defence counsel often resort to prolonging the investigation for as long as possible by, among other things, engaging in lengthy negotiations over document productions, delaying deadlines, raising problems with the client’s ability to collect relevant documents, and moving slowly to contact and co-ordinate with former employees.
Depending on the circumstances, a client may conduct an internal investigation, terminate any bad actors, and attempt to show that the conspiracy was limited to these individuals. This tactic can backfire if the participation is more pervasive and extensive than the individuals identified. Indeed, those individuals may attempt to trade their knowledge about company activities in exchange for more lenient treatment.
Companies with minor roles in the cartel activity sometimes co-operate fully with the enforcement agency so that the agency focuses its attention on other, more significant players.
Another defence strategy, if applicable, is to demonstrate that the client has a low market share and therefore no ability to participate meaningfully in an industry cartel.
The Antitrust Division’s Leniency Program is one of its most important tools for detecting cartel activity. Corporations and individuals who report their cartel activity and co-operate in the cartel investigation can avoid criminal charges, fines and prison sentences if they meet the requirements of the Program. For a corporation to qualify for leniency under the Program before an investigation has been started (Part A), the corporation generally:
Alternatively, for corporations that cannot meet all of the above conditions, the corporation may obtain leniency (part B) if:
If a corporation qualifies for leniency under Part A, all directors, officers and employees of the corporation who admit their involvement in the illegal activity will not be charged criminally so long as they act with candour and provide complete and continuing assistance to the Antitrust Division. If a corporation does not qualify for leniency under Part A, directors, officers and employees who come forward will be considered for immunity on the same basis as if they approached the Antitrust Division individually.
Even when the corporation does not seek leniency, an individual may do so in his or her individual capacity. For an individual to qualify:
A grant of leniency by the Antitrust Division applies only to criminal charges. The applicant may still be liable civilly to those they have injured. However, a party who receives leniency and provides full and complete co-operation in any civil proceeding as provided by the Antitrust Criminal Penalty Enhancement and Reform Act of 2004, Pub. L. No. 108-237, 118 Stat. 665 (2004) (ACPERA) is not subject to joint and several liability and treble damages under the Sherman Act.
While investigating authorities would prefer to receive unbridled answers from the employees of an entity under investigation, that situation rarely occurs. Once a company is aware that an investigation is ongoing, its counsel may inform the investigating authority that company counsel also represents the company’s current officers and employees (collectively, “employees”). In that situation, company counsel is in a position to negotiate the scope and terms of any employee interviews, limit access and attend any interview that the authorities conduct of the employee.
Concomitant with an assertion of joint representation by the company’s counsel, employees of a company under investigation are typically instructed as a matter of practice not to speak with an investigative agency and to notify the company’s attorneys at once if any attempt is made. This helps to ensure that the company’s attorneys are aware of where the investigation is headed.
As set forth above, the employee may be represented by independent counsel in connection with such an investigation. Separate representation creates a barrier between the investigative agency and the employee. At the same time, the company under investigation loses some amount of control.
The Antitrust Division and the FTC have administrative compulsory process referred to as Civil Investigative Demands (CIDs). The issuance of CIDs does not require “probable cause”, but there must be a factual basis to believe that there may be a violation of the antitrust laws.
CIDs are similar to subpoenas in civil litigation in that CIDs can compel the production of documents or data, answers to interrogatories, or testimony from live witnesses. A CID can also require production of the record in any judicial or administrative litigation, including depositions, documents and interrogatory answers.
Both agencies may issue CIDs themselves, and the recipient may petition a court, or the Commission in the case of the FTC, to narrow or quash the CID. If there is a dispute later over a recipient’s compliance with a CID, both agencies must turn to a federal court to enforce it.
The Antitrust Division and the FTC have formal and informal co-operation agreements with their counterpart agencies around the world. The formal agreement, called Bilateral Mutual Legal Assistance Treaties (MLATs), provides for treaty members to assist each other in criminal enforcement matters. MLATs create a channel for obtaining a broad range of foreign investigative assistance, including taking testimony from witnesses, providing documents and other physical evidence, or executing searches. Additionally, the agencies may seek letters rogatory, or requests from a US court to a foreign court for assistance in obtaining discovery materials from the foreign jurisdiction.
In cartel investigations, the Antitrust Division co-operates and shares information with the state attorneys general and their umbrella organisation, the National Association of Attorneys General (NAAG). The NAAG Antitrust Task Force, which is comprised of staff attorneys responsible for antitrust enforcement in their respective states, often organises inter-agency co-operation in complex investigations on behalf of all participating states.
Section 4F of the Clayton Act (15 U.S.C. § 15f) requires the Antitrust Division to provide state attorneys general with information concerning potential violations of the federal antitrust laws. Thus, the Antitrust Division must notify state attorneys general when it believes that the states may be entitled to bring an action under the Clayton Act based substantially on the same violation of the antitrust laws alleged in a civil or criminal antitrust prosecution filed by the Antitrust Division.
The Antitrust Division’s policy is to disclose materials to aid state attorneys general in fulfilling their antitrust enforcement responsibilities. However, several non-disclosure rules limit what information the Antitrust Division can provide to the states. Categories of protected information include grand jury matters, confidential sources, confidential business information and materials obtained from other federal agencies (such as the Internal Revenue Service) or from antitrust law enforcement agencies in other jurisdictions (such as the Directorate-General for Competition of the European Commission (“DG-Comp”)).
In 2015, the Antitrust Division implemented a protocol requiring it to refer certain criminal cases to the states when the interests and impacts are particularly local. The new policy provides that any transfer of prosecutorial responsibility under this protocol must occur at the earliest practicable point in the development of the matter.
The Antitrust Division and the FTC have formal and informal co-operation agreements with their counterpart agencies around the world, MLATs (see 3.3 Obtaining Information from Entities Located Outside this Jursidiction).
A second set of formal bilateral antitrust co-operation agreements exist with Australia, Brazil, Canada, Chile, the European Commission/DG-Comp, Germany, Israel, Japan and Mexico that permit sharing of confidential information in certain circumstances.
The Antitrust Division and the FTC also have bilateral memoranda of understanding with Chinese, Indian and Russian competition agencies, which memoranda provide a framework for co-operation concerning shared enforcement interests. For these jurisdictions, the Antitrust Division or FTC may request waivers from relevant parties to facilitate the sharing of confidential information, but those parties are allowed to withhold consent.
Consistent with the general DOJ policy, Antitrust Division attorneys recommend charging the most serious, readily provable offences. The Antitrust Division typically proceeds against corporations by indictment. One exception is the Antitrust Division’s policy that leniency is available to the first corporation to make full disclosure to the government.
Staff will inform defence counsel when it is seriously considering recommending indictment. Counsel for defendants are usually granted an opportunity to meet with staff and the relevant section or office chief to present arguments as to why their client should not be indicted.
If staff and the chief decide to proceed, then they prepare indictment recommendations to the Assistant Attorney General, the Criminal Deputy and the Director of Criminal Enforcement. The indictment recommendation memorandum typically describes the proposed defendants and summarises the charged offences and the government’s and defence’s cases. When senior staff approve a request, it will issue letters of authority authorising a grand jury proceeding.
The authorised investigation is conducted with the assistance of the local US Attorney’s office by a grand jury in a judicial district where venue lies for the underlying offence, such as a district where conspiratorial communications occurred or price-fixed goods were sold.
The FTC and Antitrust Division have different procedures for issuing complaints in civil antitrust cases. The FTC is an administrative agency with its own dedicated administrative tribunal. In its administrative adjudicative process, the Commission determines whether a practice violates the law. In contrast, the Antitrust Division can sue only in a federal district court. Both agencies must seek preliminary injunctive relief in district courts.
At the Antitrust Division, the decision to sue rests with the Assistant Attorney General for the Antitrust Division, following an investigation by staff, meetings with counsel for prospective defendants, and memoranda evaluating the merits of the case, including evidence and enforcement options. If the Assistant Attorney General decides to authorise enforcement, the matter will be returned to staff with the approval papers, signed pleadings and any other information required to file the complaint. At that point, staff will file the complaint on behalf of the United States.
For information related to the filing of a Complaint before the FTC, see 1.2 Public Enforcement Agencies and Scope of Liabilities, Penalties and Awards.
Because Section 1 of the Sherman Act outlaws understandings and agreements that restrict competition, cartel prosecutions inherently involve multiple parties. While prosecutions of Section 1 violations may proceed against a single defendant, the government typically prosecutes all conspirators in a single case.
Defendants, however, often seek separate trials pursuant to Fed. R. Crim. P. 14 on the basis that they will be prejudiced by evidence introduced against their alleged co-conspirators. Defendants also may move for severance in cases in which other defendants are charged with related crimes, such as mail or wire fraud, tax evasion, or perjury. On occasion, when the risk of prejudice is particularly clear and hard to resolve by other methods, courts will order separate trials of the charges against one or more defendants.
The government must prove a criminal Sherman Act cartel violation “beyond a reasonable doubt”, which is the highest burden of proof in the US legal system.
The burden of proof in a civil antitrust case is the standard civil burden of “preponderance of the evidence”. This standard applies regardless of whether the plaintiff is a government agency or a private party, or whether it is a jury or bench trial.
In criminal prosecutions, the trial judge instructs the jury on the elements of the crime, and the jury decides whether the prosecutors have proven an antitrust violation based on the facts presented at trial. In civil cases, depending on the plaintiff’s requested relief, the case may be tried either to a judge or a jury. Private plaintiffs typically sue for damages and seek a jury trial, although injunctive relief is a remedy that is decided by the court.
In both criminal and civil cases, the jury’s verdict is final unless it is reversed by the trial judge or a court of appeals. A court of appeals is more likely to overturn a verdict if it finds that the judge’s instructions to the jury misstated the applicable law.
In civil enforcement actions, the Antitrust Division and FTC typically seek an injunction to prevent future violations of the antitrust laws or to restore competitive conditions to the marketplace. Because these are cases for equitable relief, the judge acts as the finder of fact and issues an opinion after trial reciting those findings. Similarly, in cases tried in the FTC’s administrative system, the Commission ultimately issues an opinion, which can be appealed to the Courts of Appeals.
Evidence obtained in one proceeding typically can be used in other proceedings. Section 6(f) of the FTC Act, 15 U.S.C. § 46, authorises the FTC to share confidential information with other appropriate enforcement agencies, subject to certain confidentiality assurances. This allows the FTC and other agencies to minimise duplication in investigations. In related civil litigation, private plaintiffs may subpoena information provided to a government agency from the producing entity, including a defendant.
It should be noted, however, that evidence that might be admissible in an administrative hearing may not meet the standards for admissibility under the Federal Rules of Evidence applicable in a federal court. For example, the fact that a witness asserted the Fifth Amendment privilege is not admissible in a criminal proceeding, but it may be admissible in a civil proceeding and can lead to an adverse inference against the party asserting the privilege.
The Federal Rules of Evidence apply in all federal civil and criminal cases, including cartel and other antitrust cases. When the judge acts as the finder of fact in a “bench trial,” courts may take a more relaxed approach and admit evidence that they would not have allowed a jury to consider.
The FTC Rules of Practice establish standards for admissibility of evidence that are more permissive than the Federal Rules of Evidence. For example, the FTC permits the introduction of hearsay evidence, provided that it meets standards of materiality, reliability and relevance.
The Economic Analysis Group (EAG) of the Antitrust Division and the Bureau of Economics (BE) at the FTC employ economists to provide economic analysis in antitrust matters. These are career positions and, as a group, agency economists have deep experience with the types of analysis typically performed in antitrust matters, including statistics and econometrics.
In matters involving economic issues of substance, agency economists identify the economic issues involved in an investigation or case, assist in the development of the theory of the case, identify and present data, assist in the development of trial strategy relating to the economic issues, and sometimes testify as expert witnesses in litigation.
In addition, as with private litigants, the agencies employ independent or “outside” economists to serve as testifying experts at trial. In complex cases, they may also employ outside experts from other fields, such as industry experts.
Under Fed. R. Civ. P. 26(a)(2), which governs cases in federal court, the parties must disclose a written report, prepared and signed by the testifying expert, containing a complete statement of all opinions to be expressed at trial and all supporting data and analysis. The FTC Rules of Practice concerning testifying expert disclosures are similar to Rule 26(a)(2).
The same legal privileges discussed in 2.12 Attorney-Client Privilege and 2.13 Other Relevant Privileges apply in antitrust cases. These include the attorney-client privilege and work product doctrine, which are the most commonly asserted privileges in antitrust litigation. In the litigation context, lawyers on the same side who have signed a joint prosecution or defence agreement may use the agreement to protect their communications and work product from discovery.
In the USA, it is commonplace to have multiple proceedings involving the same nucleus of facts or the same conduct. There are several reasons for this, including the overlapping authority between federal and state governments and the existence of both public and private causes of action.
The DOJ has a formal policy, however, of not prosecuting conduct as a federal crime after completion of a state proceeding unless the state proceeding left substantial federal interests “demonstrably unvindicated”. This policy applies whenever there has been a prior state proceeding resulting in a plea bargain, a conviction and reasonable sentence, or an acquittal. Even when there is an acquittal in state court, a federal prosecution is not allowed if the state prosecutors offered essentially the same evidence that would be presented in a federal case, and there is no reason to believe that the verdict reflected anything but a good faith reasonable doubt on the part of the judge or jury.
Neither the FTC nor the Antitrust Division has the authority to impose sanctions without an adjudicative proceeding. This is true even if the agency has negotiated a resolution with the subject entity or individual. Any negotiated resolution is in almost all instances reduced to a consent order that must be approved by the court or the commissioner having jurisdiction over the matter.
Traditionally, an administrative complaint brought by the FTC is heard in front of an independent administrative law judge with the FTC attorneys acting as prosecutors. After the administrative law judge issues its findings of fact and conclusions of law, the case is reviewed by the full FTC Commission. A respondent against whom an order is issued may seek review with the appropriate United States Court of Appeals.
Civil and criminal actions brought by the DOJ are filed in the federal district court where the subject entity resides or conducts business or where the activity that is the subject of the action occurred. Except in the case of criminal acquittals, the decision of the court is appealable to the Court of Appeals for that district court.
A company or individual charged, or who will be charged, with a criminal or civil violation of the US antitrust laws can choose to voluntarily negotiate with the DOJ (criminal or civil case) or FTC (civil case) to plead guilty (criminal case) or resolve (civil case) the government enforcement action filed or to be filed. The scope of a guilty plea or civil settlement is subject to negotiation between the parties.
In a criminal case, once the negotiations conclude, the DOJ typically will file an Information (or Criminal Complaint) that alleges the criminal conduct. Occasionally, although not always, if strategic reasons countenance otherwise, the DOJ will file a plea agreement along with the Information. The DOJ may seek to file the plea agreement and Information under seal at first to protect aspects of the ongoing criminal investigation.
In civil cases, the parties may negotiate settlement terms at any time, including before, during, or even after a trial. If the settlement occurs in a federal court case, then the court will typically review the settlement, including hearing from affected persons. A settlement with the FTC outside of court requires the Commission’s approval following a notice period affording an opportunity for public comment. In either instance, the terms of the settlement will be public.
Under Section 5(a) of the Clayton Act, 15 U.S.C. § 16, a final judgment or decree rendered in any civil or criminal proceeding brought by the federal government under the antitrust laws that states that a defendant has violated the laws shall be prima facie evidence against that defendant in any related civil proceeding. For a more in-depth discussion of the collateral effects of a criminal guilty plea on subsequent related civil proceedings, see 5.4 Admissibility of Evidence Obtained from Governmental Investigations/Proceedings.
There may be other consequences to a defendant who is found guilty of violating the federal antitrust laws. If the defendant participates in government contacting, then the guilty plea may disqualify the company from that business for a defined period, or subject to reinstatement.
A guilty plea also will be unsettling for a defendant’s supply chain customers, who had counted on the company to compete vigorously and fairly for their business. Customers, even sophisticated ones, rarely, if ever, know that their supply chain had been cartelised. This is because cartelists are able to use changes in market conditions that are perceptible to customers to explain price changes and, in so doing, create the illusion of competition when, in fact, it did not exist.
Under the Sherman Act, 15 U.S.C. §§ 1-3, entities found guilty of illegal antitrust conduct are subject to imprisonment for up to ten years and fines of up to USD1,000,000, or both. In some circumstances, the court can impose fines of up to twice the gain or loss involved. The DOJ can also seek to impose fines and imprisonment for other, related crimes, such as perjury, obstruction of justice, and mail and wire fraud. The penalties for such crimes may be added to those for violating the antitrust laws.
The Robinson-Patman Act, 15 U.S.C. § 13A, provides for criminal sanctions of not more than one year of imprisonment and up to a USD5,000 fine for price discrimination for the purpose of destroying competition or eliminating a competitor. Criminal prosecution under the Robinson-Patman Act is extremely rare.
A court’s discretion in sentencing for antitrust violations is informed by the US Sentencing Commission Guidelines. A major factor in calculating the level of the offence for the purpose of sentencing is the volume of affected commerce caused by the violation. The Sentencing Guidelines expressly address the seriousness of bid rigging. According to USSG § 2R.1(b)(1), “if the conduct involved participation in an agreement to submit non-competitive bids”, the offence is increased by one level of severity in the Guidelines matrix.
At sentencing, the DOJ typically proposes what it deems an appropriate sentence, and a defendant generally counters a proposed sentence that is lower. The court has the discretion, guided by the Sentencing Guidelines, to accept either side’s proposal or reject them and impose the sentence the court deems appropriate.
The remedies for civil violations of the federal antitrust laws are as broad as any equitable remedy that a court has the power to issue. For example, courts can prohibit certain types of conduct, mandate restitution, impose reporting requirements, impose disclosure requirements or, in extreme cases, break up a business into competing parts under separate ownership. These equitable remedies can apply to both past and future conduct.
Despite the array of equitable remedies that the DOJ may seek for unlawful cartel activities, it has no power to seek civil fines. Fines are available only in a criminal proceeding subject to a statutory limit.
While there is no statutory law barring a federal enforcement agency from bringing a civil action against an individual for a violation of the antitrust laws, such actions are rare. Where there is sufficient evidence of illegal conduct by identifiable individuals within the offending companies, criminal as opposed to civil charges are typically brought against the culpable individuals in those circumstances.
A private plaintiff may also sue to enforce the antitrust laws. In a private action, the victim of the wrongful anti-competitive conduct may seek treble damages, as well as injunctive relief. Moreover, private plaintiffs may include individuals within the company as defendants in order to obtain benefits under procedure and evidence rules, and possibly to add leverage to their claim.
In July 2019, the Antitrust Division published an “Evaluation of Corporate Compliance Programs in Criminal Antitrust Investigations”. The Evaluation was drafted as an internal document intended to assist Antitrust Division prosecutors in their exercise of prosecutorial discretion. The Antitrust Division considers the nature and quality of a firm’s antitrust compliance programmes as a significant factor prosecutors must consider at the charging and sentencing phases of an investigation. The Evaluation provides guidance as to what the Antitrust Division considers to be an effective compliance programme and as to how the agency will weigh corporate defendants’ compliance efforts in prosecutions of cartel conduct.
Through the inherent equitable powers of the federal courts, government proceedings can extend to mandatory consumer redress. At their core, the antitrust laws are designed to maximise consumer welfare. To further this goal, federal district courts, subject only to an abuse of discretion standard, are fully authorised to issue mandatory injunctions, such as the disgorgement of unjust enrichment, restitution for injury suffered by consumers, or other appropriate equitable remedies.
A final judgment in a civil enforcement action brought by the DOJ is appealable as of right by the losing party to the Courts of Appeals having jurisdiction over the district court entering the final judgment. An adjudication of guilt in a criminal proceeding is similarly appealable as of right to the Courts of Appeals having jurisdiction over that district court. In either of these situations, discretionary review may be sought by the Supreme Court.
A final decision by the FTC is appealable as of right by any respondent against whom the order is issued by filing a petition for review with any Court of Appeals within whose jurisdiction the respondent resides, does business, or where the challenged practice was used. If the Court of Appeals affirms the Commission’s order, then the Court enters its own order of enforcement. The party losing in the Court of Appeals may seek discretionary review by the Supreme Court.
Section 4 of the Clayton Act, 15 U.S.C. § 15(a), provides private plaintiffs (only direct purchasers) with a claim for damages to enforce Section 1 of the Sherman Act. Under this section, successful plaintiffs are entitled to treble damages and attorneys’ fees and costs, and defendants face joint and several liability. Thus, a single defendant could be responsible for the total damages of the entire cartel (trebled), plus fees and costs.
Pursuant to 28 U.S.C. §§ 1331 and 1337(a), federal courts have exclusive jurisdiction over claims brought under the federal antitrust laws. State antitrust claims may be heard in state courts, but they may also be brought in federal courts under 28 U.S.C. § 1367 if the state claim is so related to the federal antitrust claims that they form part of the same case or controversy.
Indirect purchasers can sue under the federal antitrust laws only for injunctive relief. They will often sue in federal court asserting an injunctive relief claim under federal law and a damages claim under state law. The injunctive relief claim provides the hook that allows them to sue in federal court.
In civil antitrust cases, plaintiffs must prove their cases by a “preponderance of the evidence”. This is a lower standard than the “proof beyond a reasonable doubt” standard that the government must overcome in criminal antitrust proceedings.
In the USA, the class action is a common tool for bringing collective antitrust claims against cartels. Courts have certified both direct purchaser and indirect purchaser classes claiming damages for price-fixed products or services.
In appropriate circumstances, another potentially attractive (or more attractive) option for a plaintiff-victim is to bring a direct antitrust action (as opposed to participating in a class action). In a direct action, a company “opts out” of a class and brings its own lawsuit against cartel participants. This option allows for several benefits for the plaintiff, including control of the litigation, the ability to negotiate a resolution privately as opposed to the public character of any class action settlement, and to negotiate settlement terms that are more favourable than the terms negotiated or available in the class action.
Determining whether to remain in an antitrust class action or “opt out” and pursue a direct action depends upon a variety of factors, with no two companies being alike. Common factors that are considered in deciding what to do include the volume of affected commerce, the impact of the challenged conduct on the company, the importance of the supply chain or product or service to the company, and the company’s commitment to prosecuting the case. As a general rule, in the US system, passive class members do not have discovery obligations, whereas direct action plaintiffs do have to participate in discovery as part of prosecuting their direct action.
It is not uncommon for class and direct actions that involve the same nucleus of facts to be centralised before a single federal district judge to administer all pre-trial proceedings.
Under Supreme Court precedent, only direct purchasers (ie, entities that buy price-fixed goods or services directly from cartelists and their co-conspirators) can bring damages claims based on federal antitrust law (see Illinois Brick Co. v Illinois, 431 U.S. 720 (1977)). An indirect purchaser (ie, a purchaser that buys a price-fixed good or service from a third party who itself buys the good or service directly from the price-fixer) does not have standing to sue for damages under US law, even if the direct purchaser passed on the overcharge to the indirect purchaser.
Under federal law, the fact that a direct purchaser passed on any portion of an overcharge to its customers is not a defence (see Hanover Shoe, Inc. v United Shoe Machinery Corp., 392 U.S. 481 (1968)). This means that a direct purchaser’s estimated damages in a cartel claim are 100% of the overcharge, even if the direct purchaser plaintiff passed on all of the overcharge to its customers. (Lost profits are another measure of damages cognisable in a Sherman Act Section 1 claim.) After the Supreme Court ruled in Illinois Brick that indirect purchasers lack standing to sue for damages under US antitrust laws, a number of states enacted laws, called “repealer statutes” (as in Illinois Brick repealers), that allow indirect purchasers to sue for damages under state antitrust law.
An issue that arises frequently is whether and to what extent private plaintiffs can use criminal guilty pleas in civil proceedings.
Under Section 5(a) of the Clayton Act, 15 U.S.C. § 16, a final judgment or decree rendered in any civil or criminal proceeding brought by the federal government under the antitrust laws that states that a defendant has violated the laws shall be prima facie evidence against that defendant in any related civil proceedings.
In private suits related to cartels, Section 5(a) is most frequently invoked when plaintiffs seek to use a criminal conviction in a related civil proceeding. There is no real dispute that a criminal conviction following a guilty plea qualifies as a final judgment within the meaning of Section 5(a). Thus, plaintiffs can use such guilty pleas to estop defendants from litigating certain facts and issues that were decided in parallel criminal proceedings.
Whether a civil court will grant preclusive effect to a criminal antitrust guilty plea depends on whether the facts admitted and issues in the criminal case are substantially similar to the issues present in the civil case. Courts will closely examine the record of a criminal case to determine whether the facts and issues are subject to estoppel in the parallel civil case.
Figures show that the Antitrust Division and the FTC have settled more than 90% of the civil cases that they have brought in the last 20 years. The percentage of private actions that have been resolved through settlement is slightly lower, but the overwhelming majority of actions alleging cartel activity are resolved through settlement. A recent article co-authored by Herbert Hovenkamp points out that between approximately 70 and 88% of these cases are settled.
The American Antitrust Institute has reported that a modest percentage (approximately 15%) of antitrust cases are resolved in less than two years. About half of the cases are resolved in two to six years. Another 25% are resolved in six to eight years. The rest, approximately 10%, take more than eight years. All of these figures are heavily dependent upon the judge in the case and the defendants’ ability to use the discovery process and interlocutory appeals to prolong the proceedings.
Section 4 of the Clayton Act, 15 U.S.C. § 15(a), allows a prevailing plaintiff to recover a “reasonable attorney’s fee”. A plaintiff is considered a “prevailing” party when it is “awarded some relief by the court” (Buckhannon v West Va. Dep’t of Health and Human Services, 532 U.S. 598 (2001)). Usually, this means victory at a trial. However, a plaintiff may prevail without obtaining a final judgment, such as when it receives a preliminary injunction that provides all necessary relief. When a plaintiff is awarded treble damages at trial, a court must award attorney’s fees, although the amount is left to the court’s discretion.
In addition to attorney’s fees, a prevailing civil plaintiff is also eligible to recover costs. Most courts have adopted the position that costs of suit recoverable under Section 4 of the Clayton Act (15 U.S.C. § 15) are limited to those recoverable pursuant to 28 U.S.C. § 1920, which contains a limited list of items that are taxable as costs.
When antitrust class actions settle short of trial or judgment, the court must review and approve the settlement terms and rule on the attorneys’ fee award. Often, class counsel seek a fee in the range of approximately 20-30% of the gross recovery. In cases in which the volume of commerce and resulting settlement recovery is so large as to generate an unreasonably large attorneys’ fee when a 20-30% range is applied, courts will award fees based on a substantially lower percentage of the recovery.
In a direct action, the terms of the attorneys’ fee are set forth in private terms of engagement between the direct action plaintiff and the lawyer. These terms are subject to any number of possible permutations, and neither the private settlement, nor the private engagement terms, are reviewable by the court.
A plaintiff who does not succeed on its federal antitrust claim against a defendant is not obligated to pay the defendant’s attorneys’ fees, but must pay its costs as defined by Section 4 of the Clayton Act, as noted above.
Private litigants who sue under the federal antitrust laws can appeal decisions of the district courts to federal appellate courts.
The general rule is that only “final judgments” are appealable. Final judgments are decisions that dispose of a case completely. There are exceptions to the final judgment rule. For example, under 28 U.S.C. § 1292, a party can appeal an order that (i) involves a controlling question of law (ii) about which there is substantial ground for difference of opinion, and (iii) for which an immediate appeal from the order may materially advance the ultimate termination of the litigation. In antitrust cases, orders granting or denying class certification are often appealed pursuant to 28 U.S.C. § 1292.
In terms of final judgments, antitrust cartel cases rarely make it to trial, although some do. As described above, they often settle provided they survive motions to dismiss or later motions for summary judgment. If a court dismisses the case, then plaintiffs will often appeal. The federal appellate courts apply three different standards on review: decisions on questions of law are reviewed “de novo”, decisions on questions of fact are reviewed for “clear error”, and some decisions (like evidentiary decisions) are reviewed under an abuse of discretion standard.
Antitrust cases implicate antitrust law and economics. Even in a cartel case, in which the existence of the cartel may not be in dispute as a result of criminal enforcement actions, a firm grounding in antitrust law and economics is necessary. These disciplines are necessarily implicated when, for instance, a plaintiff sues a company for a conspiracy that is broader in scope than the company’s guilty plea. Such broader pleading and proof often occurs because (i) a guilty plea is a negotiated instrument that defines the minimum parameters of a cartel, and (ii) conduct that is not covered by a criminal guilty plea may still be actionable in a private action under Section 1, and provable due to the lower burden of proof (preponderance of the evidence (civil) rather than beyond a reasonable doubt (criminal)). As such, experienced antitrust lawyers with a plaintiff or defence practice know that a cartel case requires genuine commitment. The volume of ESI involved in discovery will be substantial, the evidentiary and practical challenges of building the case will be ongoing, proving and estimating impact with the rigour mandated by US law will involve intellectual and resource investment, and seeing that justice is done will require dedication by counsel.
The Antitrust Division publishes several policy statements concerning its Leniency Program, which affords protections to the first cartel participant to self-report the illegal cartel conduct.
The COVID-19 pandemic is having an unprecedented impact on our society. Courts have postponed jury trials, hearings generally occur via telephone or web-based videoconference, and many businesses have shuttered. Normal supply and demand have been knocked off course. Significantly, travel has come to a near halt and most, if not all, trade shows have been postponed or reduced to operating virtually. These factors presently place limitations, some potentially severe, on the ability of cartel participants to conduct secret activities in person. Meeting at trade shows or “bumping” into one another on the road provide cover for cartel exchanges. Whereas phone calls, emails and videoconferences leave a trace, in-person meetings under seemingly innocent circumstances do not necessarily raise suspicions. Thus, one possible side effect of the pandemic is a reduction in in-person cartel activities, because some of the circumstances under which collusive meetings occur have largely been shut down for the time being.
Despite the restrictions imposed on in-person meetings as a result of the pandemic, there is good reason to believe that cartel conduct will continue and perhaps be more frequent in the future. This is because the economic slowdown caused by the pandemic, and perhaps continued or exacerbated by a recession that may follow it, may cause firms in a variety of industries – but particularly those hard hit by one or both of these events – to co-ordinate in their production, pricing, marketing or sale of products or services in order to survive.
Modern history demonstrates that a great many reported cartels are a result of perceived necessity caused by economic hardship due to, for example, war, market entrants, recession, obsolete products or technology, or market innovation. Many of these confirmed cartels occurred after government enforcement agencies levied enormous fines on companies for earlier-in-time cartels. In other words, companies cartelised markets for their economic survival despite knowing the substantial risks that they faced if caught.
In the internet age, there are any number of ways that cartelists can communicate and reach an understanding or agreement not to compete without meeting physically in person, although, as noted above, almost any method of communicating electronically will leave a trail.
How firms will cartelise markets in the post-pandemic age remains uncertain (although, as noted, the means available for them to do so are well known). Whether they will do so seems certain if history is a guide. As the noted economist Adam Smith stated: “People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or some contrivance to raise prices.”
This chapter discusses trends and developments in two areas that have seen significant public and private antitrust enforcement in recent years: so-called two-sided-market cases and pharmaceutical generic-exclusion cases. Although the legal issues are different, they concern the ability of a 130-year-old statute to adapt to new areas of commerce, new technologies and new forms of anti-competitive conduct.
Disclosure: Kenny Nachwalter has represented plaintiffs in cases raising the issues discussed below, including some of the cases cited. It also submitted amicus briefs to the Supreme Court in both the Amex and the Actavis cases.
Ohio v American Express Co.: An Early Status Report
On 25 June 2018, the Supreme Court voted 5-4 to affirm the ruling of the Second Circuit that the government had failed to prove that the no-discrimination provision in Amex’s merchant acceptance agreements violates Section 1 of the Sherman Act (see Ohio v Am. Express Co., 138 S. Ct. 2274 (2018) (“AmEx”)). Two years on, lower-court decisions have struggled with how to implement the Court’s mandate to define a single product market for transactions platforms given the often starkly different competitive conditions on opposite sides of the platform market.
The government sued American Express for entering into and enforcing a contract with merchants that prohibited merchants from favouring a competing network card (Visa, MasterCard, or Discover) by offering their customers discounts or other incentives at the point of sale (POS). Amex calls this the non-discrimination provision, or NDP. The NDP prevented any expression by merchants with words (eg, “we prefer MasterCard”) or actions (eg, merchant loyalty points) to signal merchants’ preference for a different, lower-cost payment means.
The government claimed that the NDP unreasonably restrained price and non-price competition among the four competing credit card networks. The NDP insulates Amex and its competitors from having to compete on almost any metric at the POS for merchants’ credit card payment volume. The government argued that it could meet its initial burden of proof by demonstrating harm in a “one-sided” relevant market for the sale of acceptance services to merchants.
Amex did not dispute that its NDP effectively restrained competition in the sale of acceptance services to merchants. The parties also agreed that Amex is a “platform” in the sense that it serves two distinct classes of customers: merchants and cardholders. In the Supreme Court, the primary issue was whether and how to account for competition on the cardholder side of Amex’s business. While there are only four credit card networks, numerous banks issue credit cards and vie for usage by cardholders, including by rewarding points to affluent cardholders. Amex argued that cardholder-side competition was an essential part of the relevant market notwithstanding the different participants and competitive conditions.
Justice Thomas’ majority opinion holds that, because Amex is what he describes as a “transactions platform”, “courts must include both sides of the platform – merchants and cardholders – when defining the credit-card market” (id at 2286). Under this view, “because the product that credit-card companies sell is transactions, not services to merchants... the competitive effects of a restraint on transactions cannot be judged by looking at merchants alone” (id at 2287). The majority concluded that the government failed to meet its burden because it did not address the cardholder side in its market definition.
The majority justified its approach with reference to “indirect network effects”, or INE, which “exist where the value of the two-sided platform to one group of participants depends on how many members of a different group participate” (id at 2280). As Justice Breyer explained in dissent, INE means that “shoppers want a card that many merchants will accept and merchants want to accept those cards that many customers have and use” (id at 2297). He went on to ask whether the majority’s new market definition applies to a “farmers’ market” because “[t]he more farmers that participate (within physical and esthetic limits), the more customers the market will likely attract, and vice versa” (id at 2299). As Justice Breyer’s question illustrates, any firm that enjoys some discretion in setting prices to both its suppliers and its customers arguably has the attributes of a two-sided platform.
After AmEx, antitrust defendants have every incentive to contend that they operate as two-sided platforms. Within weeks of the AmEx decision, the National Collegiate Athletic Association submitted an expert report opining that the relevant market for its collegiate athletes was “a multisided market for college education in the United States” in which colleges balance their pricing to many constituencies. While the court rejected the opinion, it was at least as plausible as Justice Breyer’s farmers’ market.
FTC v Surescripts
On 24 April 2019, the Federal Trade Commission filed a monopolisation case against Surescripts, LLC, a provider of ePrescribing services. Pharmacies, pharmacy benefit managers and prescribers use its platform to exchange payment and eligibility information. Surescripts is the first post-Amex government case to address a two-sided platform market. The FTC’s Section 2 theory is that Surescripts intentionally extended its monopoly position over an emerging service by using loyalty discounts, exclusive agreements and other tactics to block potential rivals.
The FTC argued that AmEx’s two-sided market test does not apply while also alleging harms (overcharges and reduced quality and innovation) in two separate markets corresponding to the pharmacy and prescriber sides of the platform. The FTC also alleged a super-competitive total price comprised of prices charged to both pharmacies and prescribers. The complaint, however, seeks to establish separate harms to two sets of customers without having to adjust for the rewards paid to prescribers who agreed to use Surescripts exclusively.
The district court denied Surescripts’ motion to dismiss. While it declined to decide whether AmEx applies to the Surescripts platform, the district court found that the FTC could satisfy that test for harm in a single, two-sided market notwithstanding factual allegations that were not conspicuously different from the trial court’s findings of fact in AmEx. Fed. Trade Comm'n v Surescripts, LLC, No CV 19-1080 (JDB), 2020 WL 264147, at *8 (D.D.C. 17 January 2020).
US Air v Sabre
On 11 September 2019, the Second Circuit issued US Air Inc. v Sabre Holdings Corp., 938 F.3d 43 (2d Cir. 2019), the first appellate decision to apply AmEx on a fully developed record.Sabre andAmEx were chronologically and substantively intertwined. The Sabre trial started just after the Second Circuit’s decision in AmEx and concluded well before the Supreme Court’s decision. Like AmEx, Sabre concerns a “transaction platform” – Sabre’s transactions being airline reservations as opposed to credit card transactions. Airlines and travel agents constitute the respective sides of the Sabre platform.
The Second Circuitvacated the jury verdict for US Air and remanded the case for a second trial on US Air’s claim for restraining trade in violation of Section 1 of the Sherman Act. This remanded claim challenges Sabre’s “full content” rules applied to airlines that list flights on Sabre’s global distribution system or GDS. US Airalleged that the rules requiring it to provide the exact same flight information it provides elsewhere, including its own website, prevent it from fostering competition to Sabre’s GDS in a way closely analogous to the NDPs at issue in AmEx. For example, Sabre’s rules prevent the airline from offering lower-cost alternatives to customers through other channels to avoid fees that it otherwise must pay Sabre.
The Second Circuit held that, under AmEx, all “transaction platforms” are two-sided as a matter of law, even in the face of economic evidence to the contrary (id at 57). Accordingly, the jury verdict finding Sabre liable in a one-sided market was invalid. A second consequence was that any damages awarded to US Air for overcharges must be reduced to reflect the value of incentive payments Sabre made to travel agents to entice them to use its reservation system exclusively or almost exclusively. Thus, the court held, as a matter of law, that “[t]wo‐sided damages must... be lower than one‐sided damages would have been” (id at 59).
On cross appeal, US Air received reinstatement of its claims for monopolisation and conspiracy to monopolise a relevant market for just Sabre’s own reservation service (and excluding its two rivals, Travelport and Amadeus). The Second Circuit remanded these Section 2 claims for further proceedings, including a possible trial. Specifically, it found that intentional design features of Sabre’s travel agent user interface prevented travel agents from using multiple reservation systems, and this lock-in effect was plausibly reinforced by Sabre’s financial incentives “setting a threshold number of required bookings and tying the magnitude of incentives to the volume of travel agentsʹ activity on Sabreʹs platform" (id at 65).
Of course, these travel agent incentives are the same payments the Second Circuit held were an integral part of the two-sided price and an offset to damages under AmEx. Its first holding views them as a competitive benefit while the second finds them to be a barrier to effective competition. The court never reconciled these two interpretations.
United States v Sabre
On 8 April 2020, a district court in Delaware refused the government’s request for a preliminary injunction prohibiting Sabre from acquiring Farelogix, an information technology company with services that allow airlines to communicate flight information and offers to any third party outside of GDSs, including over the public internet (United States v Sabre Corp, 2020 WL 1855433, No. 1:19-cv-01548-LPS (D. Del. Apr. 7, 2020)). The government had alleged that the merger is Sabre’s plan to “kill” these new technologies. The decision shows the difficulties trial courts confront piloting through AmEx.
For example, the district court found that “Sabre and Farelogix view each other as competitors” and “the record reflects competition between Sabre’s and Farelogix’s direct connect solutions for airlines” (id at *15). However, the court reasoned that this competition is not cognisable under AmEx. Because Farelogix is a one-sided service on only the airline side of the Sabre platform and “the Sabre GDS is a two-sided platform, Sabre and Farelogix do not compete in a relevant market” (id at *32-33).
Defendants may try to argue that this holding creates a judicially crafted exemption to the antitrust laws based on a reading of Justice Thomas’ decision: “Only other two-sided platforms can compete with a two-sided platform for transactions” (AmEx, 138 S.Ct at 2287). It is another example of the difficulty courts will have accounting for differing competitive conditions on each side of a single, two-sided transactions market. Here, an innovative rival to a platform monopolist has a markedly different business model. Against the government’s argument that this difference is cause for heightened concern, the district court chose to excise the rival entirely from the market in which the parties acknowledge they compete.
What has become clear in the two years since Amex was decided is that the Supreme Court majority’s decision created the type of uncertainty that Supreme Court decisions normally resolve. In particular, a review of the cases to date shows that the lower courts have struggled with three interrelated questions.
Early in 2020, it appeared that there were opportunities on the horizon for courts to try to answer these difficult questions. But the Third Circuit will not review the Sabre merger decision after the parties abandoned their deal. US Air and Sabre have asked the trial court to stay their case due to COVID-19. This leaves the Surescripts litigation as the next vehicle for the lower courts to clarify the law.
No-AG Agreements: Market Allocation or Reverse Payment?
In FTC v Actavis, Inc., 570 U.S. 136, 148 (2013), the Supreme Court held that “reverse payment” settlement agreements in which a brand manufacturer pays a generic competitor to delay the launch of its product “tend to have significant adverse effects on competition” and are subject to the rule of reason. The agreement at issue in that case involved cash payments by Solvay, the manufacturer of branded AndroGel, a testosterone medication, to two would-be generic competitors seeking FDA approval to market generic versions of AndroGel. Both the district court and the Eleventh Circuit ruled that those agreements were immune from antitrust scrutiny because the entry date agreed to by Solvay and its competitors was before the expiry of Solvay’s (disputed) patent on AndroGel – a doctrine referred to as the “scope-of-the-patent test.” The Supreme Court ruled that such agreements are not immunised from antitrust scrutiny and reversed.
While Actavis involved old-fashioned cash, the most popular form of payment from brand companies to generic competitors has not been cash but “no-authorised-generic” or “no-AG” agreements, whereby the generic company agrees to delay the launch of its generic version of the branded drug and, in return, the brand company agrees not to market an “authorised generic” in competition with the generic company during the generic company’s 180 days of Hatch-Waxman exclusivity (or sometimes longer) (see In re Lipitor Antitrust Litig., 868 F.3d 231, 247 (3d Cir. 2017); In re Loestrin 24 Fe Antitrust Litig., 814 F.3d 538, 546 (1st Cir. 2016); King Drug Co. of Florence, Inc. v SmithKline Beecham Corp., 791 F.3d 388, 394 (3d Cir. 2015); In re Glumetza Antitrust Litig., 2020 WL 1066934 (N.D. Cal. March 5, 2020); In re Novartis and Par Antitrust Litig., 2019 WL 3841711 (S.D.N.Y. Aug. 15, 2019) (“Novartis”); In re Opana ER Antitrust Litig., 162 F. Supp. 3d 704, 713 (N.D. Ill. 2016); United Food & Comm. Workers Local 1776 v. Teikoku Pharma USA, Inc., 74 F. Supp. 3d 1052, 1063 (N.D. Cal. 2014) (“Lidoderm”); In re Niaspan Antitrust Litig., 42 F. Supp. 3d 735, 744 (E.D. Pa. 2014); Federal Trade Commission, Authorized Generic Drugs: Short-Term Effects and Long-Term Impact vi (“Settlements in which the brand-name company agrees not to compete with an AG have become... common in agreements with pay-for-delay provisions”) (August 2011)).
The reason for that popularity is not a mystery: no-AG agreements are more valuable to the generic manufacturer than they are costly to the brand manufacturer. Because launching an AG drives down generic prices, a brand manufacturer that enters into a no-AG agreement only gives up the profits it would earn from selling an AG at the lower prices that would result from that increased competition. However, the generic receives both the additional sales that would otherwise go to the AG and the higher prices it is able to charge as the only generic manufacturer on the market. While cash payments like the ones in Actavis come entirely out of the pocket of the brand manufacturer, the compensation to the generic in a no-AG case has two components: additional sales and higher generic prices. The additional sales are contributed by the brand manufacturer, while the higher generic prices are contributed by purchasers of the drug.
Some antitrust scholars have concluded that no-AG agreements can be “more anti-competitive” than cash payments, and for the same reason: they maintain higher prices at the expense of purchasers even after the agreed-upon generic entry date (XII P. Areeda & H. Hovenkamp, Antitrust Law ¶ 2046f2 at 411 (2019)). Cash payments delay the commencement of generic competition, but their direct anti-competitive effects end once the agreed-upon entry date is reached. By contrast, no-AG agreements delay the commencement of generic competition and continue to restrict competition after that date by ensuring that the first generic entrant will not face competition from an AG during its 180 days of exclusivity – in most cases, the only form of competition it can face during that time. Unlike a cash payment, which is simply a transfer from the brand company to the generic, a no-AG agreement “compensates the generic with something far more troublesome – namely, a second market division that serves to keep prices higher during the 180-day period when other generic firms are unable to enter the market” (id).
A no-AG agreement is essentially an agreement under which a generic company G agrees not to compete with brand company B for some period of time and, in return, B agrees not to compete with G for some subsequent period of time. Students of antitrust law will recognise that this is simply a temporal version of a horizontal market-allocation agreement, geographic varieties of which have been per se unlawful for many years (see Palmer v BRG of Georgia, Inc.,498 U.S. 46 (1990) (territorial market-allocation agreement was illegal per se); United States v Topco Assoc., Inc., 405 U.S. 596 (1972) (same)). There does not appear to be any principled reason to treat temporal market-allocation agreements differently from geographic ones. The result of both is the same: each conspirator will face no competition from the other conspirator in particular geographic areas or during particular periods of time. And, while federal courts seldom look beyond legal authorities for guidance, we have known for more than 100 years that space and time are interchangeable (see A. Einstein, “Zur Electrodynamik bewegter Körper"(On the Electrodynamics of Moving Bodies), Annalen der Physik 17:891 (1905) (theory of special relativity)).
Notwithstanding this straightforward analogy, the three district courts that have considered the issue have rejected the proposition that no-AG agreements should be condemned as unlawful per se (see Novartis, 2019 WL 3841711, *4 (recognising that a no-AG agreement involves temporal rather than geographic market division but seeing “no basis for extending the per se rule to the former context”); in re Zetia (Ezetimibe) Antitrust Litig., 400 F. Supp. 3d 418, 423-26 (E.D. Va. 2019); Lidoderm, 74 F. Supp. 3d at 1075). They have cited two primary reasons for that conclusion: per se treatment is (i) foreclosed by Actavis and (ii) would prevent the courts from considering the brand company’s patent rights as a justification for the payment. However, both rationales rest on questionable foundations.
First, as noted, the AndroGel case that resulted in the Actavis opinion involved cash payments, not a no-AG agreement. The Court was therefore not called upon to consider the appropriate level of antitrust scrutiny to be applied to agreements in which the payment takes a different and more insidious form. Nothing in Actavis suggests that antitrust courts should apply the rule of reason even if a particular kind of reverse-payment agreement falls under a longstanding and clearly applicable per se rule. Moreover, aspects of the Court’s reasoning appear to be limited to cash payments. For example, the Court noted that the payment “may reflect compensation for other services that the generic has promised to perform”, 570 U.S. at 156, as the defendants contended in Actavis. But no-AG agreements do not implicate a cash-for-services defence because they involve neither cash nor services. In no-AG agreements, the generic company simply agrees to delay its entry into the market and, in exchange, the brand company agrees not to launch an AG to compete with the generic when the generic ultimately launches and for some period of time thereafter. Neither cash nor services changes hands.
Second, there is no reason to believe that the Court chose the rule of reason in Actavis rather than a more stringent standard of scrutiny because of Solvay’s disputed patent rights. As the Court pointed out, “[t]he patent here may or may not be valid, and may or may not be infringed” (570 U.S. at 147). The last section of the opinion, in which the Court rejected the “quick look” standard proposed by the FTC (an intermediate standard in-between per se analysis and the rule of reason), does not rely on Solvay’s patent rights at all (see Actavis, 570 U.S. at 158-60). The Court chose to apply the rule of reason “because the likelihood of a reverse payment bringing about anticompetitive effects depends upon its size, its scale in relation to the payor’s anticipated future litigation costs, its independence from other services for which it might represent payment, and the lack of any other convincing justification” (id at 159). These considerations apply only tangentially, if at all, to no-AG agreements. While it is possible to calculate the “size” of a no-AG payment – ie, its cost to the brand and its benefit to the generic (which will be different) – that figure is a way of determining whether the agreement “reflects traditional settlement considerations”, id at 156, and an argument can be made that no-AG agreements cannot possibly reflect such considerations since they involve only restraints on competition.
Moreover, in rejecting the Eleventh Circuit’s scope-of-the-patent test, the Court cited a number of prior cases in which it had found that “patent-related settlement agreements” violated the antitrust laws (id at 147-151 (citing United States v Singer Mfg. Co., 374 U.S. 174 (1963); United States v New Wrinkle, Inc., 342 U.S. 371 (1952); United States v Line Material Co., 333 U.S. 287 (1948); and Standard Oil Co. (Indiana) v United States, 283 U.S. 163 (1931))). The Court in those cases had no difficulty applying per se rules (see New Wrinkle, 342 U.S. at 380 (“An arrangement was made between patent holders to pool their patents and fix prices on the products for themselves and their licensees. The purpose and result plainly violate the Sherman Act”); Line Material, 333 U.S. at 314 (“when patentees join in an agreement as here to maintain prices on their several products, that agreement, however advantageous it may be to stimulate the broader use of patents, is unlawful per se under the Sherman Act”) (emphasis added)). Intellectual property rights may in some cases immunise the challenged conduct altogether (as the lower courts concluded in Actavis), but they do not typically affect the decision to apply per se rules or the rule of reason, a decision that ultimatelyrests on the likelihood of anti-competitive harm. Indeed, the existence of a patent increases rather than decreases the likelihood of such effects (see Actavis, 570 U.S. at 161 (Roberts, J., dissenting)).