HORIZONTAL AND VERTICAL AGREEMENTS: DIFFERENCES BETWEEN THE EUROPEAN UNION AND THE UNITED STATES

This article compares the European Union and the United States with respect to competition law and enforcement practices as it pertains to agreements among competitors in a market (horizontal) and agreements among firms in a supply chain (vertical). Regarding horizontal agreements, the primary difference in the law is the ability of the competition authority to bring a criminal case in the U.S. and a more subtle difference is the presence of concerted practices in the EU. Enforcement differs in the far more active role of private litigation in the U.S. The differences are greater when one turns to vertical agreements. Though the EU provides safe harbors for vertical agreements, something which is absent in the U.S., it is abundantly clear that the U.S.is more lenient in the law and in enforcement. Also provided is a discussion of some recent departures between the U.S. and EU.


Introduction
The relevant U.S. legislation pertaining to horizontal agreements is Section 1 of the Sherman Act 1 and Section 5 of the Federal Trade Commission Act. 2 The jurisprudence that has come out of these acts has provided a reasonably well-defined, though incomplete, statement regarding what types of arrangements between competitors in a market are in violation of the law. 3 Complementing this dicta are guidelines provided by the Antitrust Division of the U.S. Department of Justice (DOJ) and U .S. Federal Trade Commission (FTC). 4 With one exception, it is my opinion that there are not any fundamental differences between the European Union (EU) and the United States (U.S.) with regards to competition law as it pertains to horizontal agreements. There are, however, subtle differences which could impact which cases are brought and how they are resolved.
That one fundamental difference is that the DOJ can bring either a civil case or a criminal case under Section 1, while criminal cases cannot be pursued under Article 101. For those countries with criminal liability for violating their national competition laws (such as Germany and Ireland), the U.S. is unusual in that the competition authority (specifically, the DOJ) has the capacity to criminally prosecuted, while that capacity resides elsewhere in the government for these other countries, such as with the Federal Prosecutor's office. As a result, the DOJ can decide to bring either a civil case or a criminal case, while most competition authorities are restricted to the former. A criminal case brings with it a higher standard for proving liability but also a more severe individual penalty in the form of incarceration. It is also worth noting that, contrary to almost all other countries that have criminalized collusion, the U.S. routinely penalizes those individuals involved with prison sentences.
This review of the differences between the EU and the U.S. regarding horizontal agreements will first address some issues of liability and then turn to evidentiary rules.

Agreement
Though the word "agreement" does not appear in Section 1 of the Sherman Act, jurisprudence has established that it is an agreement which is in violation of the law. Subsequent competition laws, such as 2 "Unfair methods of competition in or affecting commerce, and unfair or deceptive acts or practices in or affecting commerce, are hereby declared unlawful." 3 It is incomplete in that certain practices have been prohibited through a consent decree between the government and firms but the Court has not yet ruled on their illegality. A consent decree is an agreement or settlement between two parties that does not involve an admission of guilt or liability. A consent decree appears similar to a European Commission's commitment. 4 In particular, "Antitrust Guidelines for Collaborations Among Competitors," Federal Trade Commission, U.S. Department of Justice 2000.
Article 101 of the TFEU, have gone on to use the term "agreement". The U.S. Supreme Court has developed the doctrine that an agreement to restrain trade is unlawful and has defined an agreement as or as requiring a "unity of purpose or a common design and understanding, or a meeting of minds" (American Tobacco Co. v. United States, 1946), "a conscious commitment to a common scheme designed to achieve an unlawful objective" (Monsanto Co. v. Spray-Rite Serv. Corp., 1984), and "mutual consent" (Esco Corp. v. United States, 1965). Similarly, the EU General Court has defined an agreement as or as requiring "joint intention" (ACF Chemiefarma, 1970) and a "concurrence of wills" (Bayer v. Commission, 2000). All of these phrases refer to the same state of the world: Competitors have reached a state of mutual understanding to restrain competition. By my reading, there is no difference between the EU and the U.S. regarding what is an unlawful agreement. As discussed below, there can be differences in practices (due to evidentiary rules) and then there is the matter of "concerted practices" which exists in the EU and, at least in that form, does not exist in the U.S.
For firms to have an unlawful agreement in the U.S., there must be some exchange of assurances. This could mean one firm expressly inviting another firm to raise price to some common level and the other firm expressly accepting that invitation. However, the exchange need not be so explicit: " [Firms] need not have exchanged promises of assurances of their actions; it is enough that they have communicated their intent to act and their reliance on others to do so." (Page 2009, 451). For example, acceptance of an invitation by a firm could take the form of raising price to the proposed common level without any verbal or written communication. Or a firm announces its plan to raise price in a trade meeting with competitors and subsequently all firms raise price to the same level. Thus, an exchange of assurances can be inferred from subtle messages and actions.
U.S. courts have recognized three forms of collusive conduct, not all of which are illegal. An explicit or express agreement involves direct communication between firms which, without the parties having to draw any inferences, involves an exchange of assurances to restrain competition. A tacit agreement involves non-express communication but where there is a distinct, identifiable action that facilitates achieving mutual understanding. For example, this could mean one firm announcing a price increase at a private gathering with competitors. Explicit agreements are per se illegal in that it is sufficient to establish that the firms engaged in the activity. (Though we will later explain that there are exceptions.) Tacit agreements may be per se illegal or may only be illegal after balancing the procompetitive and anticompetitive implications of the agreement (referred to as the rule of reason). The third category of collusive conduct is conscious parallelism whereby firms are coordinating their conduct and have done so without any communication (or at least there is no evidence of communication). The most common instance of conscious parallelism is where firms price in a parallel manner and there is no evidence that they agreed to coordinate their conduct. Conscious parallelism is a process not in itself unlawful, by which firms in a concentrated market might in effect share monopoly power, setting their prices at a profit-maximizing, supracompetitive level by recognizing their shared economic interests. 5 The European Commission recognizes the same three forms of collusive conduct. Though no agreement is per se illegal under Article 101, explicit agreements are, for practical purposes, as per se illegal in the EU as they are in the U.S. Jurisdictions will vary in terms of what it takes to conclude that there is an unlawful tacit agreement (just as much as U.S. courts will differ on the matter) but both the EU and U.S. recognize this category of agreement. (There will be more on this topic later when we discuss concerted practices.) As in the U.S., parallel pricing (and other forms of conscious parallelism) is not unlawful in the EU.

Concerted Practices
A possible point of departure between the EU and U.S. is on the matter of concerted practices. In the EU, concerted practices refer to co-ordination between undertakings which, without having reached the stage where an agreement, properly so called, has been concluded, knowingly substitutes practical cooperation between them for the risks of competition. 6 My own view is that the distinction is largely semantic. The U.S. has a more expansive view of "agreement" which to a significant (if not complete) extent accommodates those practices that are classified as concerted in the EU. For example, advance price announcements can be viewed as a concerted practice. Though there is no "proper agreement" between firms, their sharing of those price intentions allows them to coordinate and thus is a concerted practice. In the U.S., advance price announcements could be viewed as an agreement in that the firm that initially proposes to raise its prices by 10% come the first of the next month is inviting the other firms to coordinate their prices, and the other 5 U.S. Supreme Court, Brooke Group Ltd. v. Brown & Williamson Tobacco Corp, 509 U.S. 209 (1993). 6 CJEU, joined cases 40 to 48, 50, 54 to 56, 111, 113 and 114-73, Coöperatieve Vereniging "Suiker Unie" UA and others v Commission of the European Communities, ECLI:EU: C:1975:174, para. 173. firms are accepting that invitation when they respond with the same announcement. 7 A case which suggests that there may be some subtle distinctions between the EU and U.S. is corrugated containers. 8 In this case, a firm would contact a rival firm to learn its most recent price charged or quoted. It was typical for a firm to match that price and, as evidenced by their pricing conduct, there was an understanding not to undercut another firm's price. The Court inferred there was direct evidence of an agreement to share information but not direct evidence of an agreement to coordinate prices. Such information sharing was determined not to be per se illegal, but a violation was nevertheless found because it was shown that it had the effect of raising price. The U.S. Supreme Court concluded that the practices violated the Sherman Act: [T]he exchange of prices made it possible for individual defendants confidently to name a price equal to that which their competitors were asking. The obvious effect was to stabilize prices by joint arrangement ... I cannot see that we would be justified in reaching any conclusion other than that defendants' tacit agreement to exchange information about current prices to specific customers did in fact substantially limit the amount of price competition.
It is possible that the information sharing practice, without any evidence of effect, might be sufficient to establish it as a concerted practice. In this case, the Court used evidence of effect and left somewhat open the question of whether that was necessary.
In sum, many concerted practices could, in principle, be viewed as agreements under U.S. law. While concerted practices may result in Article 101 being more expansive than Section 1, the differences appear to be minor.

Invitations to Collude
Under Section 5 of the Federal Trade Commission Act, the FTC has brought a number of actions against companies for inviting competitors to collude. As there was no evidence that the invitation was accepted, these cases are not prosecuted under Section 1 of the Sherman Act on the grounds that there is no agreement (though such cases have been prosecuted under Section 2 of the Sherman Act as an attempt to monopolize).
These "invitation to collude" cases range from an express invitation made in private by one firm to a rival firm 9 (which, had it been accepted by the rival firm, would have been a per se violation) to public announcements that tacitly solicit coordination. An example of the latter is an executive during an earnings call announcing that it will stop an industry price war by raising prices but will continue it should competitors not follow. 10 These cases have typically resulted in a settlement whereby the firm agrees to discontinue the conduct. Thus far, they have not been tested in court. It is unclear to what extent this type of legal action exists in the EU. It is not an agreement under the more expansive U.S. interpretation -and thus would not seem to be an agreement in the EU -and it is not clear that it is a concerted practice because there is no "practical coordination."

No Poaching Agreements
In concluding this comparison of liability for horizontal agreements in the EU and U.S., let me mention a recent enforcement direction by the DOJ. It has been prosecuting agreements between companies to restrain competition in the labor market. Those firms may or may not be competitors in the product market. These agreements often take the form of "no poaching" which means that each company agrees not to try to hire ("poach") the employees of another company. For example, a case with leading high-tech companies as defendants (including Apple, Google, and Intuit) involved an express agreement not to try to hire each other's software engineers as well as other employees.
These cases do not involve a new interpretation of the law but just a wider application of existing jurisprudence. For if one were to replace "worker" with "consumer," the "no poaching agreement" would be a customer allocation scheme whereby each company agrees not to compete for another company's customers. That agreement is per se illegal in the U.S. and illegal by object in the EU. The DOJ has increased its activity in this area and I suspect there will be a steady stream of cases. It has also put out guidance for companies so that they are well informed of the law. 11 I am unaware of any such cases in the EU though "no poaching" agreements would seem to be illegal by object under Article 101.

Evidentary Rules
In the U.S., there are three standards for determining whether some conduct is a violation of antitrust law: 1) per se rule: 2) rule of reason; and 3) quick look rule. These rules differ in terms of where the initial burden lies -with the plaintiffs or the defendants -and to what extent there is a balancing of benefits and costs in coming to a judicial decision.
The rule of reason assesses whether or not, on net, the conduct is harmful to consumers. With the rule of reason, "the factfinder weighs all of the circumstances of a case in deciding whether a restrictive practice should be prohibited." 12 The plaintiffs carry the initial burden of proving that the practice is anticompetitive. Only if the plaintiffs succeed in doing so, does the burden shift to the defendants to establish procompetitive benefits. The Court must balance the two effects in deciding whether to prohibit the conduct.
The per se rule only requires that plaintiffs show the prohibited conduct is present; there is no need to prove that it has a harmful effect. Thus, there is no balancing by the Court. The per se standard recognizes that "some classes of restraints have redeeming competitive benefits so rarely that their condemnation does not require application of the full-fledged rule of reason." 13 Firms expressly communicating to raise their prices or allocate markets are per se offenses because they clearly facilitate coordination on a supracompetitive outcome and there is almost never a competitive rationale for such communications.
It is often said that there is no defense for conduct subject to the per se rule. That is not exactly right. The Court has recognized justifications for firms to expressly coordinate their prices or allocate markets. For example, "an agreement is per se illegal as price fixing only if it affects the price at which the parties will sell something, which they could have sold individually." 14 If the market did not exist but for collusion then collusion is not unlawful. In the sulfuric acid market, there was an express agreement between Canadian suppliers and American suppliers whereby the former would enter the U.S. market (and supply the latter) as long as the American suppliers agreed to stop supplying the U.S. market. 15 Judge Richard Posner found it defensible because the Canadian suppliers were lower cost and arguably would not have entered the U.S. market but for the agreement ensuring that the American suppliers would shut down. These cases are consistent with the perspective that the Court has not really identified per se offenses but rather per se unacceptable defenses (Krattenmaker 1988, 165-180). For example, an unacceptable defense for price-fixing is that the agreed-upon price was fair and reasonable.
The "quick look" rule applies when "per se condemnation is inappropriate, but at the same time, the inherently suspect nature of the restraint obviates the sort of elaborate industry analysis required the traditional rule-of-reason standard." 16 While it may be possible to identify procompetitive benefits, the presumption is that anticompetitive benefits are present so the initial burden is on the defendants to offer a competitive justification for the conduct.
Summing up, practices for which the presumption is one of anticompetitive effects and for which procompetitive effects are exceedingly unlikely are subject to the per se rule. Practices for which the presumption is one of anticompetitive effects but it is not implausible that there are procompetitive effects are subject to the quick look rule. And practices for which both anticompetitive and procompetitive effects are plausible are subject to the rule of reason.Turning to Article 101(1), an agreement can be unlawful by object or effect. Should it be found to restrain competition by object or effect, the conduct can be determined to be legal when it satisfies a set of conditions listed under Article 101(3). In brief, these conditions are that the conduct benefits consumers (in that it produces efficiencies, some of which are passed on to consumers) and competition is not significantly harmed. The burden of proof is on the firms to defend their conduct under Article 101(3).
In principle, no practice is per se illegal in the EU as, under Article 101 (3), it could always be justified if one can show offsetting benefits to any competitive harm. For example, consider two firms engaging in express communication to coordinate their prices to maximize joint profits. It is possible that price collusion could sufficiently intensify nonprice competition (such as with regards to product quality) that consumers are actually better off (Fersthman, Pakes 2000, 207236). However, in reality, there is a set of practices that are effectively treated as per se illegal and these practices largely coincide with those that are per se illegal in the U.S. With regards to succeeding with an Article 101 (3) justification for such practices, "there is very little existing jurisprudence which provides comfort to suggest that the presumption of illegality can ever be overcome." (Jones, Kovacic 2017, 281).
As discussed above, there are defenses in the U.S. for even the most egregious horizontal agreements. Many of those defenses could satisfy Article 101(3) and thus deliver similar exemptions in the EU. In practice, there does not seem to be much difference between the EU and the U.S. with regards to explicit agreements that constrain competition. Those that are per se illegal in the U.S. (which means the Court does not consider effect and only whether an acceptable defense applies) would be illegal by object in the EU, and Article 101 (3) is unlikely to change the outcome.
For those practices that fall under the rule of reason in the U.S., they would probably be evaluated under Article 101(1) by effect and, depending on the case, Article 101(3) would result in an evaluation along the lines of the rule of reason. Thus, cases in which Article 101 (3) is relevant are likely to look like cases considered under the rule of reason in the U.S. (though there might be exceptions). However, there may not be a counterpart to the quick look rule in the EU. Recall that the quick look rule puts the initial burden on the firms to establish pro-competitive benefits. With Article 101, the European Commission must first argue a violation by object or effect and, only then, does the burden shift to the firms to argue an exemption under Article 101(3).

Twombly
An important judicial ruling in 2007 raised the bar for civil cases to proceed in U.S. courts. Standard protocol is for a plaintiff to submit a complaint which a court may choose to dismiss for failure to state a legitimate claim (i.e., according to the plaintiff's own complaint, there is not an apparent violation of the law). If it is not dismissed then the plaintiff is allowed to engage in discovery which means collecting relevant documentary evidence from the defendants. Prior to 2007, it was sufficient to use economic evidence -such as parallel pricing -to effectively state a claim and move to discovery. While such economic evidence is insufficient for proving that firms have an illegal agreement, the hope of a plaintiff and the expectation of a court is that discovery might yield the required evidence.
With its decision in Bell Atlantic Corp. v. Twombly 550 U.S. 544 (2007), the U.S. Supreme Court erected a plausibility standard in order for a claim not to be dismissed. In pleading an antitrust claim, "an allegation of parallel conduct and a bare assertion of conspiracy will not suffice." 17 The plaintiff must present "enough facts to state a claim to relief that is plausible on its face." 18 Some Section 1 cases have been dismissed for failure to state a claim (under the Twombly ruling). Firms engaged in an unlawful agreement may avoid prosecution if the only trace that they leave in the public domain -and thus is accessible by plaintiffs prior to discoveryare the higher prices that the agreement delivers. There are then cases that would be dismissed in the U.S. which could survive and ultimately lead to a conviction in the EU.

Private Enforcement
The largest gap in enforcement between the EU and the U.S. is with regards to private enforcement. Private litigation related to Section 1 of the Sherman Act has been very active for a long time, and performs two key enforcement roles. First, some cases are brought by private litigants prior to any DOJ action (and, with some cases, the DOJ never brings an action). Thus, private litigants add to enforcement by expanding the set of cartels that are shut down. Second, private litigants impose corporate penalties through customer damages. For a publicly prosecuted case, these damages augment government fines (and jail sentences) and, for those cases not publicly prosecuted, damages (and also legal fees) are the only penalties imposed on firms.
The EU has recently encouraged customer damage suits and the recommended approach differs from the U.S. in terms of the magnitude of the damages and who can claim damages. In the U.S., customers are entitled to treble damages. That is, if firms are convicted then they must pay damages equal to triple the amount of harm imposed on customers (where harm is measured by the additional payments made for the units purchased). The norm in the EU is that firms are liable for single damages. In the U.S., only direct purchasers can sue in federal court (though in many states, indirect purchasers are allowed to sue). For example, if a cartel of manufacturers raised their wholesale prices to retailers then retailers can sue but final consumers cannot, even though they are likely to have been harmed due to the pass through of higher wholesale prices. In the EU, both direct and indirect purchasers have standing to sue.
In rationalizing these differences, the EU's policy is motivated by properly compensating those who were harmed. While compensation is also a consideration in the U.S., its system is also justified by deterrence. By allowing for treble damages, the penalty is higher and thus is more likely to deter collusion. By only allowing direct purchasers to sue, it is more likely that colluding firms will be prosecuted and convicted because those with the best information (direct purchasers) are highly incentivized to sue and, from a practical perspective, one avoids the challenge of allocating damages across different purchasers in the vertical chain.
In comparison to many EU countries, the U.S. legal environment makes it easier to engage in private litigation. Private litigants can engage in discovery which allows them to acquire crucial evidence towards obtaining a conviction (or a profitable settlement). In addition, it is relatively easy to pursue class action suits which means that cartels for which customers are large in aggregate but individually small will be prosecuted.
While private litigation is growing in the EU, it is still minor in terms of the additional penalties it levies and, to my knowledge, it is rare if not totally absent for any private litigation to occur without first achieving a conviction by a competition authority.

Concluding Remarks
While there are differences between the treatment of horizontal agreements in the EU and the U.S., those differences are minor. For the most part, what is illegal in the U.S. is illegal in the EU and vice versa. There may be some practices that are illegal in the EU as a concerted practice but are legal under the more expansive concept of agreement in the U.S., but they appear to be rare. Evidentiary rules do differ because of the hurdle erected in the U.S. from the Twombly ruling but, once a case surmounts that hurdle, it does not appear that there are substantive differences regarding the resolution of those cases. The most substantive distinctions are that the U.S. actively engages in criminal prosecution (and regularly incarcerates cartelists) and the much more active role of private enforcement.

Types of Vertical Restraints
Vertical restraints are agreements between firms that operate at different levels of the production and distribution chain. For example, a manufacturer that produces goods and a retailer that distributes those goods. There are many legitimate reasons for such firms to enter into an agreement and it is the challenge of competition law and enforcement to identify those agreements that have anticompetitive effects and then prohibit them. As opposed to horizontal agreements, for which the most egregious are kept private to the parties (for, once discovered, they are sure to be prosecuted), vertical restraints are generally out in the open for all to see. The challenge is not detecting them but rather determining whether they are harmful. That task has proven difficult in principle (as attested to by the challenges faced by economic research) and in practice (as attested to by the challenges faced by courts and competition agencies).
The EU and U.S. are common in the types of vertical restraints that can be found in violation of competition law. These vertical restraints are: 1) resale price maintenance; 2) territorial and customer restraints; 3) exclusive dealing; and 4) tying. As we will discuss later, differences lie in what it takes to violate the law (or for there to be a determination that there has been a violation) and the intensity with which enforcement occurs.
Resale price maintenance involves an upstream firm (typically, a manufacturer) restricting, in some manner, the prices that can be charged by downstream distributors (typically, retailers). Resale price maintenance can involve limiting the maximum price that retailers can charge, but the more serious concern is with specifying a minimum price for retailers. With minimum resale price maintenance, the risk is that it will lead to supracompetitive prices that can harm consumers, and it can possibly facilitate collusion among upstream firms. But it can also generate efficiencies by promoting non-price competition. (From hereon, RPM will refer to minimum resale price maintenance.) The other three restraints draw the attention of competition law because they can exclude competitors from serving consumers and this can (but need not) harm consumers. Given that competition is about a firm trying to sell to more consumers and thereby "exclude" rivals from doing so, there is an intrinsic challenge to identifying when such exclusion is harmful and when it is competition as ought to occur.
A customer or territorial restraint is an agreement between a supplier and a retailer that constrains to whom a retailer can sell or to whom a supplier can supply. A common territorial constraint gives a retailer the sole right to sell in a particular geographic area and with that sole right often comes the limitation that the retailer cannot sell outside of its designated area.
Exclusive dealing is a contract between an upstream supplier and a downstream buyer which specifies the latter will buy all of its supplies from that supplier. An exclusive dealing contract is a special case of a "contract that references rivals" (CRR) which can be exclusionary (see Scott Morton 2013, 7279). The defining feature of a CRR is that the terms of a contract between a buyer and a seller depend on the buyer's transactions with another seller. It could require that a buyer purchase a certain percentage of its inputs from a supplier (which is exclusive dealing when the percentage is 100) or just make it costlier to buy from another seller but not outright prohibit it.
Tying refers to the practice of a supplier agreeing to sell its customer one product (the tying good) only if the customer agrees to purchase all of its requirements for another product (the tied good) from that same supplier. A traditional concern with tying is that it can be a way to leverage market power in one product market to another product market. Whether that is a cogent argument depends on the particular situation.
The competition law challenge associated with vertical restraints is that there are possible efficiencies from such practices but also possible anticompetitive effects through reduced price competition or foreclosure of rival firms.
There is no simple conclusions as to whereby any particular type of restraint -territorial restrictions, tie-ins, vertical price restraints, etc. -always improves economic efficiency or reduces it. All types of vertical restraints, including both price and nonprice restrictions, may either increase or decrease efficiency, and they have different economic effects in different contexts. Thus, exclusive territories, resale price maintenance and exclusive dealing can all be used to solve free-riding problems in the provision of retail services. The same restraints can also be used to provide manufacturers with better incentives to invest in product quality. However, all of these restraints can also reduce interbrand competition (Comanor, Rey 1997, 38).
Given this reality, competition law should neither make these vertical restraints always lawful or always unlawful. Each instance of a vertical restraint is to be judged as to whether the procompetitive benefits exceed the anticompetitive costs and thus should be allowed, or the contrary is true and thus should be prohibited. While such an approach is appropriate in principle, implementation of it can be difficult and costly. As a consequence, competition law on paper may be quite distinct from what it is in practice. That point will be relevant when assessing the competition policies of the EU and U.S.

U.S. Law and Enforcement regarding Vertical Restraints
The relevant laws under which these practices are most commonly prosecuted in the U.S. are the Sherman Act's Section 1 (which prohibits unreasonable restraints of trade) and Section 2 (which prohibits monopolization and is relevant when vertical agreements are used to advance the market power of a dominant firm). However, some of these restraints have also been prosecuted under Section 3 of the Clayton Act (which makes it unlawful to sell goods while requiring that the buyer not purchase a competitor's goods when it would substantially lessen competition) and Section 5 of the Federal Trade Commission Act (which prohibits unfair methods of competition). In all cases, a plaintiff must establish that the vertical agreement is likely to deleteriously affect the competitive process in such a manner as to harm consumers.
The perspective of U.S. courts and the Antitrust Division of the U.S. Department of Justice (and the U.S. Federal Trade Commission) has changed radically over the last half century, and this change is generally attributed to the research and writings of economists. The traditional view in economics was that many of these vertical restraints were inherently anticompetitive because they foreclosed rivals from markets or restrained price competition. That view was tenuous because it neither had grounding in economic theory or supportive empirical evidence. The Chicago School of the 1950s and '60s provided a systematic theoretical examination of many of these practices and arrived at the conclusion that they are procompetitive. They first provided models of firm conduct which showed that practices like exclusive dealing and tying would not lead to more profits for a firm with market power. That analysis undercut the claims of anticompetitive effects. This left them with the challenge of explaining why firms would adopt those practices if they are not helping augment their market power. That led to the second contribution, which was to identify procompetitive justifications for these vertical restraints.
The views of the Chicago School proved highly persuasive to courts. Indeed, rarely has economic reasoning been so influential in the area of competition law. However, economic analysis proved to be subtler than revealed by the Chicago School. Starting in the 1980s with the use of game theory, these practices were re-examined with more sophisticated and more realistic models. It was found that there are conditions under which all of these vertical restraints are anticompetitive. The Chicago School had identified a special set of market conditions whereby these vertical agreements were procompetitive but there are other market conditions for which they are indeed anticompetitive.
Returning to our discussion of U.S. law, many of these vertical restraints were per se illegal until the 1970s (and prior to the intellectual impact of the Chicago School). That is, it was sufficient to establish that the vertical restraint was in place; there was no need to show harm. As stated above, there were no theoretical or empirical bases for such a legal approach. Abiding by the contributions of the Chicago School, one might have shifted to making many of these vertical restraints per se legal in that they showed them not to be harmful and that they could be beneficial. That did not happen but the courts were persuaded enough by the Chicago School to shift from per se illegality to the rule of reason. Fortunately, they did not go to per se legality because the later game-theoretic contributions showed that there are plausible conditions whereby these restraints do harm consumers but also plausible conditions whereby they benefit consumers. In such a situation, the rule of reason is appropriate.
With Continental Television v. GTE Sylvania, 433 U.S. 36 (1977), the U.S. Supreme Court ruled that all vertical restraints, except those that restrained prices, are to be judged by the rule of reason. In State Oil Co. v. Khan, 522 U.S. 3 (1997), the U.S. Supreme Court ruled that maximum resale price maintenance was to be judged under the rule of reason. The final restraint subject to the per se rule fell ten years later. In Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007), the U.S. Supreme Court decided that minimum resale price maintenance was not per se illegal because the "per se illegal" category is reserved for those practices, such as horizontal price fixing among competitors, that rarely have any procompetitive benefits to offset anticompetitive harm. However, in the case of RPM, many economic studies had established ways in which consumers can benefit from restraining intrabrand price competition among retailers through inducing more intense non-price competition. While forsaking per se illegality, the Court made clear that RPM was not per se legal and was to be judged by the rule of reason.
With the possible exception of tying (which, under certain conditions, is claimed to be per se illegal), 19 all vertical restraints in the U.S. are judged by the rule of reason. With the rule of reason, the burden is initially on the plaintiff to establish that the restraint harms consumers. If it succeeds in doing so, the burden shifts to the defendants to show that there are countervailing benefits to consumers. For the defendants to avoid the restraint being prohibited, it is not enough for the procompetitive benefits to exceed the anticompetitive harm. There must not be another practice that is better in the sense that it delivers those same benefits but with less harm.

Comparison of Competition Law and Enforcement
between the EU and U.S.
Our comparison of the treatment of vertical restraints in the EU and the U.S. will focus on four dimensions: 1) safe harbors and exemptions; 2) 19 "Although courts have been recently inclined to consider the business justifications for tie-ins and have analysed the economic effects of the tying arrangement, hallmarks of a rule-of-reason analysis, a tying arrangement may be treated as per se illegal (i.e., irrefutably presumed to be illegal without the need to prove anticompetitive effects) if the following elements are satisfied: 1) two separate products or services are involved; 2) the sale or agreement to sell one product/ or service is conditioned on the purchase of another; 3) the seller has sufficient market power in the tying product market to enable it to restrain trade in the tied product market; and 4) a substantial amount of interstate commerce in the tied product is affected." (Kazmerzak, Sandrock 2019, 1011). evidentiary standards; 3) dominant firms; and 4) intensity of enforcement. There are substantive differences though they are of a smaller order of magnitude then the differences between the U.S. today and the U.S. fifty years ago.

Safe Harbors and Exemptions
The EU's Block Exemption Regulation (BER) provides a safe harbor for firms regarding vertical restraints. If neither the upstream firm nor the downstream firm has a market share exceeding 30% then a vertical agreement is lawful. However, there are "hardcore" restrictions (such as RPM and certain territorial and customer restraints) that lack this safe harbor. U.S. competition law provides no safe harbor though, in practice, enforcement effectively provides it. There is little or no chance of a vertical agreement being prosecuted by the DOJ or FTC if the upstream and downstream firms are sufficiently small. Though it is not codified as in the EU, I believe that a market share not exceeding 30% would be small enough. Thus, the presence of safe harbors in EU law but not U.S. law belies the reality that the U.S. is more tolerant than the EU when it comes to vertical restraints.

Evidentiary Standards
In its near-universal use of the rule of reason, the U.S. makes a vertical restraint lawful if consumers benefit after considering all implications of the restraint. For example, RPM can raise prices to consumers but still be lawful if defendants can show there are non-price benefits to consumers (such as retailers providing better service) that, on net, means consumers are not harmed. Even if exclusive dealing is shown to significantly foreclose rivals, plaintiffs must go further and establish that it is detrimental to consumers such as through higher prices or reduced product variety. With regards to customer and territorial restraints, anticompetitive concerns are raised in the U.S. only if those restraints impair interbrand competition (as opposed to competition among retailers of the same brand) and it is then shown that consumers are worse off. Unfailingly, U.S. competition law requires that a vertical restraint is shown to harm consumers regardless of how much it might foreclose the market to competitors or raise prices to consumers. Turning to the EU, the European Commission (EC) must establish under Article 101(1) that the vertical agreement has as its "object or effect, the prevention, restriction or distortion of competition within the common market." If it establishes that claim, the defendants can turn to justifying that it is procompetitive by drawing on Article 101(3), and showing that there are efficiencies to offset the anticompetitive effects and that consumers receive their "fair share" so as not to be harmed. Despite its apparent similarity to the U.S.'s rule of reason, there is a difference. This framework would approximate the US rule of reason if the Commission's burden ... were to show likely adverse effects on consumer welfare. This, however, does not appear to be the case. The Commission's burden does not require an analysis of competitive effects of the sort undertaken in the US. Rather, EU case law suggests that it is enough for the Commission to show that the agreement in question restricted the "economic freedom" of either a party to the agreement or a third party, without regard to a likely effect on prices, output, or consumer welfare generally (Cooper, Froeb, O'Brien, Vita 2005, 298).
The U.S. religiously abides by the consumer welfare standard which means the plaintiff must show that consumers are harmed (e.g., showing foreclosure is insufficient) and the defendants can avoid a guilty verdict by establishing that there are benefits to consumers which exceed the harm identified by plaintiffs. In contrast, it appears that a vertical restraint can be found unlawful in the EU without establishing a harmful effect on consumers.
For a defendant to appeal to Article 101(3) in justifying a vertical restraint, it is necessary that the restraint not eliminate competition, either actual or potential. There is no parallel condition in U.S. competition law. A strict application of the rule of reason would imply that a vertical restraint could eliminate all competition and not be prohibited if the efficiencies it generated were able to offset any harm to consumers from the loss of competition. For example, if exclusive dealing were to eliminate all rival companies but, due to scale economies, result in lower prices to consumers so as to make them better off, that could be lawful in the U.S. However, the set of instances in which competition is entirely eliminated and consumers benefit is likely to be sufficiently sparse as to make this difference not meaningful.

Dominant Firms
A substantive point of departure between the EU and U.S. is with regards to market dominance. To begin, the EC seems more inclined to pursue Article 102 cases than the DOJ is to pursue Section 2 cases (which have been exceedingly rare for quite some time). In the context of vertical restraints, the BER does not apply to dominant firms. Furthermore, vertical restraints can be prohibited without showing effect; it is enough for a dominant firm to foreclose and exclude rivals.
Exemplifying the different approaches to market dominance in the context of vertical restraints, Virgin Atlantic brought a case against British Airways for agreements that the latter made with travel agents. These agreements provided higher commissions if a travel agency exceeded its previous year's sales of tickets on British Airways flights. Virgin Atlantic claimed that this was an abuse of dominance in that it encouraged travel agencies to put more of their business with British Airways. The EC concluded that it was a violation of Article 82 and its decision was affirmed by the Court of First Instance. 20 In contrast, the case failed in U.S. courts with the Second Circuit commenting that "even with monopoly power, a business entity is not guilty of predatory conduct through excluding its competitors from the market when it is simply exploiting competitive advantages legitimately available to it." 21 2.7. Intensity of Enforcement More generally, enforcement with regards to vertical restraints is distinctly more aggressive in the EU than the U.S. One can point to cases in which the EC prosecuted parties and the DOJ chose not to bring a case. Already mentioned is the British Airways-Virgin Atlantic case. As another example, the EC pursued a case against Coca-Cola for placing restrictions on downstream customers (such as restaurants and bars) regarding the extent to which they could carry competing brands. 22 Coca-Cola settled by agreeing not to enter into such vertical restraints. In contrast, such vertical agreements have had no problems in U.S. courts.
More recently, the EU has been active in pursuing cases involving vertical agreements in the form of most favored nation clauses and minimum advertising price agreements, while such practices have not been prosecuted in the U.S.
A most-favored nation clause (MFN) is a vertical agreement requiring a party to give a buyer (or a supplier) the same or a better deal as offered to other buyers (or suppliers). A class of MFNs have been relevant in recent years are price parity clauses associated with online booking sites. An online platform, such as Booking.com, would require that the prices offered by a hotel on its site are as low as it offers elsewhere. A narrow price parity clause pertains only to prices offered at direct booking channels, while a broad price parity clause pertains to all online listings including competing platforms and direct booking channels.
This has been a growing area of activity by the EC and national competition authorities in the EU (see cases cited in Verge 2018). While no consensus has yet emerged regarding when price parity clauses are anticompetitive, the most common view in the EU is that narrow price parity clauses are not anticompetitive by object, though they could be viewed as anticompetitive without having to prove harm to consumers (Schaeffer, Rinne, Coombs 2016). Regarding broad price parity clauses, there may be an emerging consensus that they "restrict competition between platforms on commission rates and may also prevent entry by innovative low-cost platforms." (Verge 2018, 6). In comparison, MFNs and price parity clauses are generally viewed as benign or procompetitive in the U.S. In particular, we have not seen cases based on price parity clauses adopted by online booking platforms in the U.S.
When a manufacturer has a minimum advertised price agreement (MAP) with retailers, there is a floor placed on the price which a retailer can advertise a good though a retailer is allowed to sell the good at a price below that floor. In principle, a MAP is distinct from RPM. The DOJ and FTC have not pursued cases against MAPs and, even prior to the adoption of the rule of reason for RPM in 2007, the pro-competitive benefits of MAP were recognized.
It has been quite a different story in the EU. There seems to be more concern with MAPs and, consequently, there have been some cases. Reflective of this view: "Restrictions on advertising prices below a certain level have been found to lead to de facto RPM in certain past cases on the basis that these restrict the ability of the reseller to determine its sales prices." 23 Like RPM, there seems to be an approach to MAP rooted in them being anticompetitive by object.

Concluding Remarks
In reading the law, one might infer that the EU is more tolerant of vertical restraints than the U.S. There are safe harbors encoded in the Block Exemption Regulation in the EU, while the U.S. has no such counterpart. In principle, Article 101(3) could provide as much opportunity for firms to defend a vertical restraint as being procompetitive as the rule of reason in the U.S. Jurisprudence and the conduct of competition authorities tell a different story. Vertical restraints that meet the safe harbor conditions under the BER are, in practice, extremely unlikely to be prosecuted in the U.S. In effect, there are non-codified safe harbors in the U.S. that are just as "safe" as those in the EU. Historically, U.S. courts have moved from some vertical restraints being subject to the per se rule to where now all (or almost all) vertical restraints come under the rule of reason. Furthermore, there have been so few cases involving vertical restraints pursued by the DOJ or FTC in recent years that one might be inclined to infer that there is de facto per se legality for some vertical restraints. I do not yet subscribe to that view but the evidence in support of it is mounting.
The disparity in law and its enforcement between the EU and the U.S. is exemplified by the treatment of minimum resale price maintenance (RPM). Since 2007, it has been evaluated under the rule of reason in the U.S. but what that means is unclear for there has not been a single case brought by the DOJ or FTC. In contrast, RPM continues to be a concern to the European Commission and is treated as anticompetitive by object. Though RPM can be defended by putting forth efficiencies under Article 101(3), this seems to be a more challenging task than with the U.S.'s rule of reason. As another example of the greater intensify of enforcement in the EU, minimum advertised price agreements have been viewed as "de facto RPM," and thus subject to scrutiny, while such agreements have not drawn the attention of the DOJ or FTC.