ECONOMETRIC EVIDENCE TO TARGET TACIT COLLUSION IN OLIGOPOLISTIC MARKETS

ECONOMETRIC EVIDENCE TO TARGET TACIT COLLUSION IN OLIGOPOLISTIC MARKETS Abstract Tacit collusion may reduce welfare comparably to explicit collusion, but it remains mostly unaddressed by antitrust enforcement that greatly depends on evidence of explicit communication. We propose to target specific elements of firms' behavior that facilitate tacit collusion by providing quantitative evidence that links these actions to an anticompetitive market outcome. We apply our approach to incidents on the Italian gasoline market, where the market leader unilaterally announced its commitment to a policy of sticky pricing and large price changes that facilitated price alignment and coordination of price changes. Antitrust policy must distinguish such active promotion of a collusive strategy from passive, best-response, alignment. Our results imply the necessity of stronger legal instruments that target unilateral conduct that aims at bringing about collusion. I. INTRODUCTION In most markets, firms quickly realize that they can earn supracompetitive profits by coordinating their market conduct. In response, antitrust policy seeks to foster “effective competition” by targeting collusive activities. The current legal framework to accomplish this goal has mainly evolved around communication as a means to reach a collusive agreement. In contrast, purely tacit collusion remains largely unaddressed by antitrust law, though it may bring about the same negative welfare effects. We argue that a crucial step forward in targeting tacit collusion could be taken through the forensic use of econometric evidence that may reveal collusive strategies. Theoretical and empirical findings on collusive behavior provide a basis for deriving clear test hypotheses to distinguish lawful oligopolistic interdependence from tacit collusion. Thus, econometric analyses may provide quantitative evidence that firms strategically use specific elements of market conduct to tacitly collude. Antitrust remedies should in turn take up such instances of market behavior to tackle tacit collusion. The paramount significance of evidence of explicit communication entails fundamental problems for the fight against cartels.1 Communication is not a necessary condition to collude. At the heart of collusion lies the incentive of firms to cooperate rather than to compete.2 In oligopolies, firms can exercise their unilateral market power to facilitate anticompetitive coordination without engaging in communication. As firms weigh the costs and benefits of explicit collusion, antitrust law's focus on communication incentivizes them to concentrate on tacit means of collusion. Legal instruments to counter collusion, the effectiveness of which depends on evidence of explicit communication, are least effective in concentrated industries3—that is, precisely in those industries where the cartelization rate is presumably the highest and communication is least needed to sustain collusion.4 Any economic approach to support the enforcement of antitrust law5 is challenged by a legal significance of evidence of communication. Economists can use observable variables, such as prices and their knowledge of the strategies employed by firms to infer collusion,6 but have no instruments to prove whether firms collude with or without communication. From an incentive-based perspective, illegal communication appears to be of relative unimportance; while non-enforceable communication might facilitate coordination on a particular collusive equilibrium,7 “talk is cheap” in the absence of effective enforcement mechanisms.8 This does not, however, mean that economic theory regards communication as uninformative. On the contrary, there is a large amount of literature on the conditions under which costless communication is informative.9 The point is that firms collude tacitly or explicitly because it is in their best interest to do so. If antitrust law is confined to addressing communication that facilitates collusion, or through which firms collude, firms will shift to other means of coordination that potentially have similar effects on welfare. It is, however, not out of economic naivety that antitrust law concentrates so much on evidence of communication in its struggle against collusion. First, this reflects skepticism about whether instances of tacit collusion may be distinguished from oligopolistic competition with a degree of precision that suffices for forensic purposes. This concern may be associated with the so-called “indistinguishability problem,” as put forward by Louis Phlips.10 He suggested that game theoretic arguments, combined with the unavailability of some key data, can make an economic based proof of collusion very difficult, as something that looks like collusion might stem from a multiplicity of indistinguishable equilibria. Thus, the application of any legal instrument that addresses tacit collusion faces the challenge to prevent an unacceptable high number of false positives. Secondly, for purposes of antitrust enforcement it does not suffice to show that an observable market outcome emerged as the result of a collusive strategy. Antitrust remedies may not straightforwardly tackle firms because they charge “collusive”—that is, supracompetitive—prices but must address specific elements of firms' market conduct that may be characterized as collusive. Without taking into account these issues, antitrust enforcement that tackles tacit collusion risks either unduly restricting market operators' leeway to compete or ultimately amounting to an instrument of price control. In the following, we outline an approach that addresses both these concerns, and thus provides the basis for an expansion of the law's ambition to tackle tacit collusion. Oligopolistic interdependence as such and oligopolistic collusion are conceptually distinct. Tacit collusion arises from decisions endogenous to the market by one or several firms that aim to reduce or eliminate competition. In contrast, oligopolistic interdependence stems from best response to market conditions, including other firms' behavior, that favor non-competitive performance. Thus, while the market outcome might appear to be “indistinguishable,” the specific strategies that lead to the outcome differ significantly. The gist of our approach to identify collusive behavior lies in an identification of patterns of behavior used by firms to bring about or facilitate tacit collusion.11 Antitrust law must not simply infer the existence of a punishable tacit agreement from the insight that a certain market outcome is the result of a collusive strategy. Rather, it is essential to distinguish the active promotion of a collusive strategy by one firm from the passive, best-response alignment of competing firms. Consequently, antitrust enforcement should not conceptualize such instances of collusive leader-follower behavior as an illegal coordination that would—with regard to the “followers”—result in punishing oligopolistic interdependence. Rather, antitrust law should capture such instances of “unilateral collusion” only through considering as illegal the unilateral conduct that actively promotes the implementation of a collusive strategy. To effectively fight tacit collusion, it is therefore necessary to strengthen legal instruments that target the unilateral conduct that firms strategically employ to promote collusion. To illustrate our behavioral approach to tackling tacit collusion and to demonstrate the capacity of econometric evidence, we refer to incidents on the Italian gasoline market. In a recent article we provide quantitative evidence of the means—that is, specific pricing strategies—and the effects—that is, higher prices—caused by a unilateral public announcement of ENI, the market leader.12 On October 6, 2004, ENI announced a new pricing policy that consisted of infrequent price variations (sticky pricing) and large price changes. Using daily firm level prices of gasoline in Italy and average weekly EU prices over the time period from January 2003 to May 2005, we show the effect of the new pricing policy. ENI increased the time lag between price changes from 6 to 20 days and increased the mean price change from 1 percent to 5.4 percent. After the policy change, ENI did not change its price for 57 days irrespective of cost changes. Initially ENI's competitors kept their short-run cost-based pricing and thus increased their prices following lagged cost increases.13 Once competitors started to align with ENI in mid-November 2004, a different pricing pattern emerged: sticky-leadership pricing. Competitors matched each large price variation and ENI endogenously emerged as the price leader in the market and coordinated price changes. While the first effect of the policy was to change the price interdependence in the Italian gasoline market, this newly emerged tacit coordination had the additional effect of a significant price increase. Using several estimation techniques, we show that Italian prices rose compared to EU prices after the new sticky leadership pricing emerged. Thus, the econometric analysis used to characterize pre-policy and post-policy pricing behavior and evaluate the effect of the new market conduct on the price level might provide solid “statistical” evidence that ENI's unilateral commitment to a policy of sticky pricing must be characterized as collusive. The structure of the article is as follows: Part II outlines the status quo of cartel enforcement that focuses on firms' communication and the law's difficulties with tackling tacit collusion. In Part III, we discuss incidents on the Italian gasoline market as an illustration for how our approach might be applied for purposes of antitrust enforcement. Part IV describes the way to integrate quantitative evidence of collusion with antitrust law. Part V concludes. II. ON COLLUSION AS A LEGAL CONCEPT, ITS LIMITS IN THE ABSENCE OF EVIDENCE OF COLLUSIVE COMMUNICATION, AND THE REASONS THEREFORE Collusion allows competing firms to charge supracompetitive prices and entails negative welfare effects. Meta-studies on cartel overcharges show that the median cartel-price increase ranges between 20 and 30 percent.14 This is why antitrust law aims at inhibiting collusion and why the horizontal coordination of prices and quantities is considered a per se violation of section 1 of the Sherman Act and a restriction of competition by object pursuant to Article 101 of the Treaty on the Functioning of the European Union (TFEU), respectively. Successful collusion requires inter alia an underlying—tacit or explicit—consensus on the terms of the cooperation. Thus, in order to counter collusion, it seems a logical step to regard such underlying understanding as illegal. However, the economic conception of a collusive agreement diverges significantly from the corresponding legal concepts of “conspiracy” according to section 1 of the Sherman Act or “agreement” and “concerted practice” according to Article 101(1) of the TFEU.15 While the former focuses on firms' incentives to engage in collusion and their strategies for sustaining a collusive equilibrium, the latter centers around the means to reach an understanding between firms. This divergent perspective on collusion becomes apparent with regard to instances of tacit collusion—that is, under circumstances where firms sustain a collusive market outcome without direct communication and where, therefore, no direct evidence (such as written records or insider testimony) of an exchange of assurances concerning a coordination of future market conduct is available. Though, as a matter of principle, both under the Sherman Act and the TFEU, circumstantial evidence may suffice to demonstrate the existence of a “conspiracy”16 or an “agreement,”17 respectively, there are doctrinal limits in this regard if it comes to supposedly tacit collusion between competitors. In the words of the U.S. Supreme Court, “conspiracy” requires “that [the defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective.”18 Reasonably, this may not be inferred from conscious parallelism alone.19 Rather, a plaintiff must produce additional evidence to prove that an observed parallel market conduct may not be considered the result of mere oligopolistic interdependence, but indeed forms part of a collusive strategy. Such so-called “plus factors” may encompass, first, elements of industry structure that indicate that an industry is conducive to collaboration, second, conduct that appears irrational or inefficient absent collusion, and, third, additional factors such as industry performance (for example, stabile market shares over time, supracompetitive pricing) or facilitating practices (for example, exchange of information).20 Although the U.S. Supreme Court has stated that plaintiffs can only survive summary judgment by presenting circumstantial evidence “that tends to exclude the possibility that the alleged conspirators acted independently,”21 the case law so far does not provide a taxonomy of plus factors that would allow us to determine which elements of evidence are required to infer an agreement. It has thus been concluded that “decisions analyzing plus factors generally have failed to establish a clear boundary between tacit agreements—to which section 1 applies—and parallel pricing stemming from oligopolistic interdependence … This condition makes judgments about future litigation outcomes unpredictable.”22 Observers have noted, however, that up until now “no case has held purely tacit behavior to be illegal.”23 Although the European Court of Justice (ECJ) considers it generally conceivable that consent to an agreement may be inferred from circumstantial evidence,24 the Court is reluctant to infer an “agreement” between competitors from their market conduct alone, notwithstanding the presence of certain “plus factors.” Given the current state of the jurisprudence, it appears that in the absence of direct evidence of collusion the Court does not presume the existence of an “agreement,” even if one has proved that observed parallel market conduct was an expression of tacit collusion rather than of oligopolistic interdependence. This has been reaffirmed by a decision on the doctrine of “collective dominance” under Article 2 EU Merger Regulation where the Court implicitly approved that tacit collusion per se may not fall under Article 101(1) of the TFEU: “Unless they can form a shared tacit understanding of the terms of the coordination, competitors might resort to practices that are prohibited by Article [101 TFEU] in order to be able to adopt a common policy on the market.”25 However, where tacit collusion has been induced by facilitating practices, such as an exchange of information, it may come under Article 101(1) of the TFEU as an illegal “concerted practice.” In this regard, the ECJ drew a line. On the one hand, by assigning market operators the legal leeway to “adapt themselves intelligently to the … conduct of their competitors,” the Court signaled that mere passive alignment would not be treated as an illegal form of coordination. On the other hand, the Court submitted that a strategy that actively aims at aligning competitors' market conduct may fall under Article 101(1) of the TFEU.26 Thus, to implement this standard, it is essential to identify elements of behavior that promote tacit collusion. This brief insight into legal concepts of coordination reveals ambiguities and restrictions with regard to tacit collusion. It invites one to question why the law finds it so difficult to cope with this phenomenon, given that it seems uncontroversial in terms of competition policy that tacit collusion on prices and quantities should be prevented as rigorously as collusion based on explicit consensus. The respective judicial definitions of “conspiracy” and “agreement” do not restrict these concepts in a way that would exclude collusion that has been sustained tacitly. Whatever the rhetoric of the courts might be when they characterize the requirements of an agreement—typically they refer to a need to show a “meeting of minds,”27 a “joint intention,”28 or a “concurrence of wills”29—the respective antitrust law concepts must be defined strictly instrumentally. Thus it is, first, the underlying policy to contain as far as possible any kind of welfare-reducing collusion and, second, the role a legal intervention and, in particular, a prohibition of agreements between competitors may feasibly play in this regard that determine which behavior should be regarded as illegal. Part of the law's problem in coping with tacit collusion lies with the difficulty to distinguish collusion from oligopolistic interdependence, as the latter may also result in suspiciously parallel market conduct and supracompetitive prices. This problem is addressed by the requirement of “plus factors” that—in addition to parallel pricing—are meant to indicate collusion, such as market conduct that may reasonably only be explained as part of a collusive strategy.30 From this perspective, the problem of distinguishing oligopolistic collusion from oligopolistic competition comes down to a question of error costs. By defining the “critical mass” of plus factors required to infer an illegal coordination, courts strike a balance between the ambition to contain tacit collusion and the risk of producing false positives.31 However, the courts' reluctance to infer an agreement in cases of mere tacit collusion suggests that there is more to the law's difficulties to cope with tacit collusion than the problem of multiple indistinguishable equilibria and the issue of reaching an acceptable degree of error costs in this regard. Legal standards and remedies that are supposed to influence market conduct in order to guarantee effective competition may not simply prohibit an undesired economic condition, such as a collusive equilibrium, and punish firms because they charge “collusive” prices. Such a policy effectively meant nothing other than price control. This unwelcome consequence is prevented when antitrust standards and remedies relate to individual behavior and define which acts or omissions are required or prohibited. When authorities or private plaintiffs order a firm to bring an infringement to an end or seek to obtain injunctions before a court, it is already the remedy's behavioral nature that requires a specification of elements of conduct that violate antitrust law. The intended deterrent effect of concurring remedies, such as imposing fines or damages, likewise depends on whether market operators are in a position to foresee what conduct they may be sanctioned for, and how they are expected to behave to avoid sanctions. This appears particularly challenging where an undesired economic effect or market condition is the consequence of the interdependent behavior of several market actors.32 But once again, if the elements of behavior that bring about a collusive equilibrium remain unclear, any legal intervention may ultimately amount to a price control by antitrust authorities or courts. Furthermore, with regard to criminal and quasi-criminal sanctions it is required by the principle of culpability33 and the need to prove intent34 or negligence,35 respectively, that antitrust enforcement ensures that market operators may anticipate their legal leeway and addresses certain modes of behavior rather than an economic effect or condition. Thus, the key to overcoming the law's difficulties to counter tacit collusion lies in an approach that identifies specific elements of behavior with the object or effect to bring about or facilitate collusion. Such an approach has a chance for success as market operators that seek to implement a collusive strategy need to adjust their market conduct to reach an optimal and stable collusive equilibrium. Even in oligopolistic markets that are characterized by features that facilitate tacit collusion, prices and other parameters have to be adjusted according to an underlying tacit agreement, and the need for such adjustments may lead firms to resort to a certain behavior that may be identified as serving a collusive strategy. Empirical and theoretical research36 on how cartels behave provides solid test hypotheses to identify such elements of collusive behavior. Precisely these elements of behavior are the focus of our approach to provide evidence of anticompetitive behavior. III. EMPIRICAL EVIDENCE Academic forensic economics and finance37 has long applied its tools in a number of areas to reveal conduct that agents strive to conceal. Some of the most prominent examples include teachers cheating in exams,38 violations of U.N. sanctions,39 and racial biases in employment decisions.40 This research is methodologically related to our topic of empirical cartel detection, as econometrics is employed to provide evidence of hidden wrongdoings. In academic forensic economics and finance, researchers use their knowledge of incentive schemes on observable variables, such as prices, in order to derive statistical tests to compare distinct hypotheses, such as collusion versus competition. While a test hypothesis for teachers to raise students' test scores or employment discrimination on the basis of race can be clearly defined, what should constitute an appropriate test for collusion? In line with the literature on economic screens41 we believe that the answer lies in economic theory and empirical evidence on cartel behavior.42 Since the foundational work by George Stigler,43 who highlighted firms' incentive to cheat as the preeminent challenge faced by cartels, much research has been carried out on “pricing structures” that can sustain a collusive outcome.44 The two key strategic aspects that are relevant for our analysis are the use of sticky and leadership pricing as a facilitating device to sustain collusion. The role of price leadership as a collusive strategy to coordinate on the focal price of the leader has been analyzed by Julio Rotemberg and Garth Saloner45 and Igor Mouraviev and Patrick Rey.46 The relation between price rigidity and collusion has been the focus of many theoretical articles using different settings.47 In addition, there is a growing empirical literature that uses data on detected cartels to test whether pricing behavior significantly changes during a cartel situation. For example, some scholars show that price stickiness is associated with collusive behavior.48 A. The Facts of the Case On October 6, 2004, ENI, the market leader in the Italian gasoline market, publicly announced the adoption of a new pricing policy. ENI declared that the purpose of this policy was to lower the short-term price-cost relation and to stabilize retail prices.49 While prices used to follow short-run cost changes, ENI's aim was to significantly lower the price responsiveness to persistent cost changes.50 As the volatility of crude oil was increasing, ENI claimed that the policy aimed at lowering the retail price variability would benefit customers. The new policy consisted in a reduction of the number of price changes relative to cost changes—that is, sticky pricing—and increased the magnitude of each variation. ENI increased the average time lag between price changes from 6 to 20 days and increased the mean price change from 1 percent to 5.4 percent. The result of this declaration can be seen in Figure 1, which shows the daily price per company over time before and after the introduction of the new pricing policy. Before the policy, firms' price changes were frequent. On average, firms changed their prices every five days. The average price change was 0.9 percent before the policy change. After the new pricing policy was introduced, price changes occurred infrequently, every 9 days on average, but their amount became larger, 2.4 percent on average. Figure 1. View largeDownload slide Cartel formation. Figure 1. View largeDownload slide Cartel formation. As a result, all but one competitor, ERG,51 followed ENI's new pricing strategy. About five months later in March 2005, the Italian Truckers' Association, FITA, complained to the Italian antitrust authority about high and aligned prices.52 This eventually led to an investigation by the antitrust authority for price fixing under Article 14 of Law 287/90 of October 10, 1990, the Italian legislation that restates Article 101 of the TFEU. The Italian antitrust authority claimed that the petrol firms' conduct of adapting their prices to the leader's price must be considered a collusion to stabilize prices and to coordinate price changes.53 The high transparency in the market facilitated an exchange of price information. Firms may easily observe their competitors' prices at each gas station and Italian law required weekly price communications to the Ministry of Industry that subsequently published the data. More importantly, companies communicated future price changes to a specialist Italian magazine, “Staffetta Quotidiana,” that published all price change announcements on its website. Cost transparency also facilitated coordination. The major source of cost is the Platts Cif Med,54 the wholesale price that refineries charge in the Mediterranean area for gasoline. This price can be thought of as the opportunity cost of companies to sell their gasoline on the Mediterranean wholesale market rather than to gas stations. It thus constitutes industry practice55 to compute firms' margins as the difference between their suggested consumer price and the Platts Cif Med. The Italian antitrust authority had no proof of direct communication between the firms, other than the price changes the firms communicated via the aforementioned online magazines. The authority claimed that ENI's policy created a focal price used to facilitate coordination. ENI's sticky pricing lowered competitors' uncertainty about the future pricing, while the large price variations helped to coordinate price changes.56 However, the Italian antitrust authority ended its investigation without issuing a formal decision—that is, it refrained from asserting infringements of Article 101(1) of the TFEU and Article 14 of the Law 287/90, respectively, and from imposing a fine on ENI or its competitors. Instead, the authority accepted a commitment by the firms to restrict pricing transparency on the market and, in particular, to abstain from the publication of future prices via the media.57 While there are good arguments supporting the opinion that regular and reciprocal announcements of future pricing, such as those that occurred on the Italian petrol market, should suffice to demonstrate a concerted practice,58 our analysis does not focus on this aspect of posted pricing,59 but concentrates rather on ENI's and its competitors' pricings.60 B. Sticky Pricing Sticky pricing constitutes an important element in a strategy to sustain collusion. An advantage of rigid pricing is that it is straightforward to implement and that deviations can be easily detected and punished. A series of studies61 analyze the profit maximizing scheme of cartels under different settings and find a direct relation between optimal collusive schemes and rigid pricing.62 For example, Athey and Bagwell show that when firms are moderately patient, the equilibrium that maximizes ex-ante profits is relatively simple: all firms adopt a sticky pricing scheme and charge the consumers' reserve.63 In this equilibrium, colluding firms adjust their prices infrequently and thus sacrifice productive efficiency to sustain a higher price level in the market.64 However, from a theoretical perspective, increasing the time lag between price changes—that is, the length of the commitment—also comes at a cost. The longer the time between price changes, the larger the gain from undercutting by rivals, as highlighted by Zhongmin Wang.65 Thus, from ENI's perspective, the large increase in the length of the price commitment might also imply a higher cost of being price leader, especially if ENI is the first firm to raise prices. While this argument certainly plays an important role, there is a fundamental difference between the standard exogenous commitment and ENI's endogenous commitment. Given the high market transparency, at any point in time ENI might detect and immediately punish deviation. ENI's commitment increases price stability, but this might be reverted instantaneously as market conditions change. Thus, this pricing scheme entails both the pro-collusive benefits of price stability and, at the same time, ENI does not have to lose market share for a long period of time before it can respond to deviation by rivals. In fact, most empirical studies conclude that prices are more rigid when the industry is in a collusive phase.66 A key example is the study by Abrantes-Metz, Froeb, Geweke and Taylor on the frozen perch market. Using ex-post evidence of collusion the authors find that the price variance during collusion was indeed distinctly lower than the price variance in the period after the end of the cartel.67 In a meta-study, Blanckenburg, Geist and Kholodilin compare the distribution of price changes between competition and collusion for 11 cartels. They find that the price variance decreased significantly in 8 out of 11 examined cartels.68 C. Leadership Pricing Price leadership has been considered “one of the most important institutions facilitating tacitly collusive pricing behavior.”69 Theoretical evidence has been presented by Rotemberg and Saloner, who demonstrate that price leadership facilitates collusion under asymmetric information and increases price rigidity.70 The authors conclude that such a pricing scheme has many positive attributes. It is easy to implement, it doesn't require communication, and it is very easy to detect and to punish deviations. In line with these findings, Mouraviev and Rey study the role of price or quantity leadership under circumstances where firms can act either simultaneously or sequentially in an infinitely repeated setting for both Bertrand and Cournot competition.71 They highlight that leadership facilitates collusion. Firms competing on prices a la Bertrand can use price leadership to sustain perfect collusion for any value of the discount factor while leadership is less effective with quantity competition a la Cournot. Both papers convey an important implication for antitrust policy: if firms are able to tacitly collude using price or quantity leadership, the negative effects on welfare are essentially the same compared with cases of explicit collusion. The way firms collude is not decisive for the negative effect collusion has on consumers' welfare. In addition, both papers show how leadership pricing can be used to implement an anticompetitive strategy in the market, as it facilitates coordination and makes deviation more visible. D. Key Empirical Findings Based on the previous finding of the role of sticky and leadership pricing to sustain collusion and on the effects of collusive agreements on prices, we show that ENI's pricing behavior facilitated price coordination and led to a price increase. Table 1 shows the different pricing conduct firms adopted after ENI's price commitment. In Panel A we summarize the frequency and magnitude of price changes. Columns 3 and 5 show the differences in the pre- and post-mean of these variables and test whether the pricing behavior significantly changed after ENI's policy. ENI significantly increased the time lag between price changes from one every 6 days to one every 20 days. This difference is significant at the 1-percent level and shows that ENI did hold its price commitment as publicly announced on October 6, 2004. In addition, the leader increased the absolute mean price change from 1 percent to 5.4 percent. This 4.4 percent increase is statistically significant at the 1-percent level. Similar results hold true for all firms. The average time lag between price changes increased from 5 to 9 days, while absolute price changes increased from 0.9 percent to 2.4 percent, both significant at the 1-percent level. Theoretical literature discussed above suggests that large price changes might have been used to coordinate price changes on the leader's focal price. Panel B tests this hypothesis and shows whether the average number of perfectly aligned competitors (that is, up to three digits) to the leader and the average price difference of competitors to ENI significantly changed after ENI's new pricing policy. In line with the collusive hypothesis, the number of aligned competitors significantly increased and the average price difference to the leader significantly decreased after the policy. Table 1. Pre and post policy pricing   (1)  (2)  (3)  (4)  (5)  (5)  Panel A: Frequency and Magnitude of Price Changes  Time period  Pre  Post    Pre  Post      Mean  Mean  Difference  Mean  Mean  Difference    (St. Dev.)  (St. Dev.)  t-stat  (St. Dev.)  (St. Dev.)  t-stat    Abs. % Price Change   Days between price changes   All Firms  0.0090  0.0238  0.0148  5.2  8.9  3.6    (0.0066)  (0.0299)  14.7  (5.2)  (8.0)  8.6  ENI  0.0104  0.0537  0.0433  6.1  20.1  13.9    (0.0072)  (0.0393)  10  (6.0)  (16.1)  5.9  Panel B: Average Alignment to the Market Leader    Sum of aligned firms   Price difference to ENI   ENI's Competitors  1.99  2.82  0.83  0.00175  0.00135  −0.0004    (0.093)  (0.169)  4.4  (.00006)  (.00047)  −1.34    (1)  (2)  (3)  (4)  (5)  (5)  Panel A: Frequency and Magnitude of Price Changes  Time period  Pre  Post    Pre  Post      Mean  Mean  Difference  Mean  Mean  Difference    (St. Dev.)  (St. Dev.)  t-stat  (St. Dev.)  (St. Dev.)  t-stat    Abs. % Price Change   Days between price changes   All Firms  0.0090  0.0238  0.0148  5.2  8.9  3.6    (0.0066)  (0.0299)  14.7  (5.2)  (8.0)  8.6  ENI  0.0104  0.0537  0.0433  6.1  20.1  13.9    (0.0072)  (0.0393)  10  (6.0)  (16.1)  5.9  Panel B: Average Alignment to the Market Leader    Sum of aligned firms   Price difference to ENI   ENI's Competitors  1.99  2.82  0.83  0.00175  0.00135  −0.0004    (0.093)  (0.169)  4.4  (.00006)  (.00047)  −1.34  Notes: Table 1 summarizes the pre and post policy pricing behavior of the nine firms acting in the Italian wholesale gasoline market. Panel A shows the frequency and magnitude of price changes. ENI increased the mean price change from 1% to 5.4%, while the average price change of all firms increased from 0.9% to 2.4%. Similarly, ENI increased the average time lag between price changes from one every 6 days to one every 20 days. Panel B shows the sum of aligned firms to ENI (specification 1 and 2) and the average price difference to the leader (specification 4 and 5). The number of aligned competitors significantly increased after the policy, while the average price difference to the leader decreased after the policy. View Large In addition to the price coordination adopted by firms, we report the key coefficients on the causal effect of the policy on prices and margins in Table 2.72 Specification 1 shows the result of the dif-in-dif model with standard errors clustered at the country level. In this regression, weekly prices of eight EU countries73 were used as a control group.74 The estimate on the dif-in-dif effect of the policy on Italian prices is positive and highly significant. As one might question the subjective selection and the sufficient similarity of the control group, in specification 2, we first construct an “optimal” data-driven benchmark—that is, a synthetic control group—and then take the weekly difference between the Italian price and the “optimal benchmark” as the stationary, dependent variable. The synthetic control group estimation was developed by Abadie and Gardeazabal75 and Abadie, Diamond and Hainmueller,76 respectively, and is constructed using a data-driven weight of European prices that minimizes the pre-treatment differences between the Italian price and the resulting synthetic control group. Consistent with specification 1, we find a positive and significant effect of the policy on prices. Finally, specification 3 shows the within market regression of firm-level margins—that is, without benchmark—that also points to a positive and significant effect of the new policy on firms' profits.77 Table 2. Effect of the policy on prices   (1)  (2)  (3)  Dependent Variable  Price EU Country j at week t  Price Difference Italy-Synthetic Control week t  Margin firm i day t  Type of Data  Panel Data  Time Series  Panel Data  Regression Model  Dif-in-Dif  OLS  Firm Fixed Effect  Policy*Italy  10.02***        (2.219)      Policy    11.19***  22.88***      (3.879)  (1.871)  Controls  Crude oil (4 Lags), Year and Month FE  Crude oil (4 Lags), Time trend  Time trend  Observations  882  113  7,794  R-squared  0.665  0.493  0.121    (1)  (2)  (3)  Dependent Variable  Price EU Country j at week t  Price Difference Italy-Synthetic Control week t  Margin firm i day t  Type of Data  Panel Data  Time Series  Panel Data  Regression Model  Dif-in-Dif  OLS  Firm Fixed Effect  Policy*Italy  10.02***        (2.219)      Policy    11.19***  22.88***      (3.879)  (1.871)  Controls  Crude oil (4 Lags), Year and Month FE  Crude oil (4 Lags), Time trend  Time trend  Observations  882  113  7,794  R-squared  0.665  0.493  0.121  Notes: Table 2 reports the coefficients on the full specification regression models that capture the effect of the new pricing policy. Policy*Italy is the intersection between two dummies (Italian price after the policy), while Policy is a dummy being one after October 6, 2004. FE stands for fixed effects. Prices and margins are expressed in € per 1000 liters. In specification (1) standard errors are clustered at country level, while in specification (2) and (3) robust standard errors are reported. In all specifications prices/margins significantly increased after the competitors adopted the same pricing behavior as the market leader. View Large Figure 2: View largeDownload slide Italian Price, EU Price and Brent. Figure 2: View largeDownload slide Italian Price, EU Price and Brent. The results of the econometric analysis show that ENI's policy had two effects. It facilitated price coordination and increased average prices. E. Discussion and Robustness of the Empirical Results In oligopolistic markets, the way firms interact with their competitors determines their profits. Our empirical analysis shows that the ex-post effect of the leader's credible commitment to sticky pricing was an equilibrium with higher prices. ENI's success in the implementation of a collusive scheme depended on the individual incentives for its competitors to adhere. The first issue that arises, therefore, is whether it is reasonable to think that the leader could expect ex ante that its competitors would adopt its pricing and that this would cause an increase in prices. Firms' behavior is a key element of managerial choice. Spagnolo shows that typical compensation schemes for CEOs are designed to incentivize tacit collusion at the cost of “income smoothing.”78 In addition, managers are aware of or, at least, well-advised of strategic behavior that favors collusion.79 Since the seminal work by Thomas Schelling,80 it is common knowledge that commitment lies at the heart of strategic behavior.81 If competing firms could write enforceable contracts on prices, most industries would collude. However, as explicit collusion is illegal and the decision to communicate is endogenous, firms may opt for tacit collusion instead. Yet any collusive strategy must be incentive compatible, irrespective of whether it is implemented explicitly or tacitly. After its announcement on October 6, 2004, ENI kept prices fixed for 57 days, until December 3, 2004.82 This means that ENI kept sticky prices for almost 10 times the usual price-change interval of 6 days, irrespective of cost changes. Just after ENI's announcement costs increased and its competitors kept cost-based pricing. As costs fell again competitors started to align to ENI at the beginning of November, that is, about a month after ENI's change in pricing policy. We can only speculate about what would have happened if costs had risen after ENI's announcement. However, it clearly emerges, both from Figure 1 and from the price-interdependence analysis, that ENI strongly committed itself to sticky pricing. As can be inferred from Table 2, specification (3), ENI's competitors behaved in their best interest as industry margins increased. ENI emerged endogenously as the price leader through its use of market power and then used its position to coordinate the price changes of its competitors, which ultimately caused a price increase. While each market has its traits, and results from an individual market cannot be easily generalized, leadership pricing has been consistently associated with collusion. Our empirical results provide substantial evidence that ENI's strategy aimed at coordinating and increasing prices came at the expense of consumers. A second concern that arises is where to set the boundaries between a firm's freedom to set its profit-maximizing price on the one hand, and antitrust authorities' power to prevent certain behavior that results in supracompetitive pricing on the other. To address this issue, we need to distinguish the “source” of market power that made that market outcome possible. In this respect, it is helpful to compare our empirical results with those of Borenstein, Bushnell and Wolak, who analyze inefficiencies in the restructured Californian electricity market.83 They find that wholesale electricity expenditures increased in the summer of 2000 with respect to the summer of 1999 from $2.04 billion to $8.98 billion and that about 59 percent of this increase was caused by the exercise of unilateral market power. Both the Italian gasoline market and the Californian electricity market suffered from higher prices. However, there is a key difference—in California, market power stemmed from exogenous shocks. Electricity prices were relatively low compared to a benchmark in 1998 and 1999, but dramatically increased in the summer of 2000. While there are many structural factors that make it easy for electricity firms to exercise market power, such as binding constraints at peak times or difficulties to forecast demand and high storage costs, firms did not actively implement a new strategy to coordinate and increase their prices, but rather individually best-responded to shocks that favored the exercise of market power. Three of the many factors that Borenstein, Bushnell, and Wolak identify are that hydroelectric production decreased in 2000 because of dryness; economic growth in the Western United States increased demand for energy; and the cost of nitrogen pollution permits increased from about $1 per pound to over $30 per pound, resulting in an increase in the price of gas. In the Californian electricity market, regulation should address the structural problems that have been revealed by the incidents of the summer of 2000. However, insofar as the firms only best-responded to exogenous shocks, their conduct should not be addressed by cartel enforcement. In contrast, our analysis reveals that the active implementation of a collusive strategy by one firm resulted in an anticompetitive market outcome and should therefore be targeted by antitrust enforcement. Finally, one may wonder whether the pricing policy implemented by ENI in fact had pro-competitive effects. Being the market leader, ENI possibly had better information about demand or supply trends. Thus, competitors might have followed ENI's pricing because it was based on such information, and as a result firms' prices were more aligned. Although, in theory, each strategy might entail some pro-competitive elements, our approach makes use of the ex-post observed effect of a specific conduct on the market outcome. Sticky-leadership pricing causally led to a higher price level compared to a European benchmark and thus had a welfare decreasing effect on consumers. None of ENI's competitors adopted a myopic best response and undercut the other firms when costs were decreasing, but rather waited for the leader to move first. This is consistent with a dynamic best-response behavior, in which firms may anticipate the pro-collusive effect of keeping their prices aligned, and reinforces the conclusion that the observable pricing pattern must be characterized as collusive. Moreover, as we already discussed in Parts III.B and III.C, there is sound theoretical and empirical evidence that firms use leadership pricing and sticky pricing to facilitate coordination and raise prices to a supracompetitive level. ENI's strategic behavior addresses two main problems of a cartel–coordination and stability. In a setting in which prices are transparent, ENI's stable price served as a focal price and facilitated collusion. In addition, the initial commitment not to increase prices when costs were increasing shows that ENI was trading off short term losses against future gains. In doing so, ENI signaled that it was prepared to pay the costs in terms of lower margins to maintain its pricing strategy and to punish competitors should they deviate from the pricing pattern as set by ENI. IV. INTEGRATING ECONOMIC INSIGHTS ON COLLUSIVE STRATEGIES INTO THE LEGAL FRAMEWORK As any collusion between competitors may result in welfare losses, it is essential to strive to contain collusive behavior irrespective of direct evidence of a “meeting of minds” or explicit communication between firms. There remains, however, an outstanding question of how economics may be integrated with the legal framework and how antitrust law should be developed to counter tacit collusion. There are several reasons to believe that this challenge deserves more attention than ever. First of all, prevalence of tacit collusion may increase in times of globalization. Information on competitors' actions as capacity choices, prices, and transactions are widely reported by international media and transparency increases. Firms interact in many markets, increasing their scope to collude. Second, market players must not be regarded as naïve, but as professionally advised and capable of employing economic know-how strategically to avoid price wars and to reach collusive equilibria, instead. Third, the introduction of leniency or other types of immunity programs increased the capability of antitrust authorities to produce direct evidence of collusion, such as documents or insider testimony, and has thus significantly strengthened the effectiveness of the law to counter collusive behavior that occurs via explicit communication.84 As the decision to communicate is endogenous to market players, leniency programs have increased firms' cost of following such a strategy. This is likely to cause or to have already caused a shift from explicit to tacit collusion. These are grounds to expect that social welfare damage caused by tacit collusion will increase. Certainly, merger control may work preventively against coordinated effects. However, it will hardly suffice to counter increasing anticompetitive effects caused by tacit collusion. This is mainly the result of inherent limits of scope. Market concentration may increase as a result of a firm's internal growth and the exit from the market of firms that operate less efficiently. Thus, tacit collusion remains unaddressed by merger control in cases of stable oligopolies and especially in duopolistic markets—that is, exactly in those industries where the market structure entails high risks of tacit collusion.85 It appears to be crucial, therefore, for antitrust law to find a way to target those elements of behavior that are employed by firms to implement a collusive strategy and whose collusive character may be demonstrated by the kind of analysis as suggested in this article. Inasmuch as it appears inadequate to regard such behavior as illegal coordination, this calls for a development of the law against unilateral anticompetitive conduct. A. “Unilateral Collusion” and Unlawful Coordination Price leadership may serve as a mechanism to find a consensus about the collusive price, a challenge any cartel faces. From this perspective, one might be tempted to conclude that price leadership should be prohibited. In this case, if ENI was not allowed to be a price leader, ENI would not be able to coordinate price changes, and the anticompetitive effects—that is, the significant price increase that resulted from such behavior—would not have taken place. However, price leadership and any other “leader-follower” behavior may equally be the result of effective oligopolistic competition. For example, the market leader often has an informational advantage over smaller competitors in respect of relevant changes in market condition. A price change by the leader might entail useful information for market participants and their individual best response might be to align to the market leader. If firms with superior information are not allowed to adjust prices and competitors are not allowed to respond best to price changes by competitors, or both, then the price setting freedom of firms will be significantly limited, a situation that might ultimately lead to regulated prices. This raises the question of whether under circumstances such as those in the present case—that is, where it may be demonstrated that “leader-follower” behavior sustained a collusive equilibrium—such conduct should be considered illegal. In other words, should the kind of evidence presented herein be regarded a “super plus factor”86 that allows courts to infer an illegal tacit agreement? If certain conduct of two or more firms is conceptualized as an unlawful coordination—that is, a violation—of, for example, section 1 of the Sherman Act or Article 101(1) of the TFEU, this implies that the law regards the behavior of these firms as a wrongdoing that may be punished. In other words, where a certain collusive equilibrium has been brought about by the unilateral collusive conduct of one firm, one should only infer a punishable agreement if one also considers the competitors' reactions as inappropriate behavior. Turning again to the general regulatory and legal requirements we formulated with regard to antitrust enforcement,87 we recall that antitrust standards and remedies should address specific elements of behavior, and that market operators should be provided with an idea of which conduct may be regarded as acceptable or not acceptable under defined market conditions. Such standards of conduct must be in line with the general purpose of antitrust law to foster effective competition. Thus, if tacit price alignment in response to unilateral collusive conduct ought to be prohibited, the law has to define how firms should behave once a competitor's conduct may be interpreted as a tacit invitation to engage in tacit collusion. When ENI held prices constant despite cost increases and thereby signaled its commitment to a policy of sticky pricing, its actions might be viewed as a “suggestion” to its competitors to align their pricing policy and as an “offer” to take on the role as price leader. Should ENI's competitors have been legally obligated to refrain from any market conduct that ultimately could have been regarded as having brought about a collusive equilibrium and thus proof of an underlying illegal agreement? We believe the answer is no. First of all, it does not seem feasible to define any meaningful and administrable legal standard of conduct in this respect. Should it have been forbidden for ENI's competitors to tacitly align their prices to ENI's policy of sticky pricing? Should they have been obligated to stick to their higher prices and with open eyes to put up with losing market share? And even if an alignment of pricing to the strategy of a price leader such as ENI was prohibited, the question would remain: How closely and how quickly a competitor would be allowed to adjust its market parameters? In the absence of any clear standard of behavior, a legal intervention in situations of supposedly collusive pricing may ultimately amount to judicial price regulation. In addition, under such a legal regime a market player could strategically restrict the competitive room for maneuver of its competitors. If it was prohibited for ENI's competitors to align their pricing to ENI's strategy because such an alignment would be conceptualized as an acceptance of an offer to engage in illegal coordination, ENI could have restricted the price-setting freedom of its competitors by implementing its strategy of sticky pricing. These considerations point to the heart of the regulatory problem with regard to “unilateral collusion.” The reaction of ENI's competitors to ENI's pricing policy must be regarded as mere best response. Their behavior is an expression of mere oligopolistic interdependence, even though they benefitted from the higher price level in the market. Consequently, a passive adaptation to collusive market conduct should not be considered illegal, but part of functioning oligopolistic competition. Thus, collusive “leader-follower” behavior must not be conceptualized as a form of unlawful coordination, and thus illegal according to section 1 of the Sherman Act or Article 101(1) of the TFEU. Antitrust law should instead target unilateral collusive behavior that facilitates “best response” that ultimately leads to collusion. This appraisal of collusive “leader-follower” behavior appears to be in line with the treatment of non-conspiring firms that, for example, adjust their prices in reaction to a price increase by cartelizing competitors. Such a constellation is generally referred to as “umbrella pricing,” as the nonparticipant benefits from the “price umbrella” are spread by its cartelizing rivals.88 This metaphor somewhat obscures the interdependence between the optimal cartel price and the behavior of the firms outside the circle of cartel participants. Firms that are aware of illegal price coordination by their competitors and engage in “umbrella pricing” consciously contribute towards reaching a certain collusive equilibrium. Furthermore, these firms will also consider the risk of retaliatory measures by the cartelists in the case that they ignore the cartel policy. Thus, their reaction should be characterized as a dynamic best response similar to the alignment of pricing (for example, alignment of pricing by ENI's competitors). Nevertheless, even if a non-cartelist acts consciously in accordance with the collusive strategy of a cartel, “umbrella pricing” is generally regarded as being innocent, and the legal discussion centers only around the question of whether customers of nonparticipants may recover damages from the cartelists.89 Thus, notwithstanding that “umbrella pricing” contributes toward sustaining collusive equilibria, antitrust law does not require market operators to abstain from a best-response strategy in reaction to their competitors' pricing. The law refrains from imposing on non-cartelists a duty to keep prices constant (or at least at a lower level than the cartel price), which would effectively amount to a duty to increase output to offset the cutback of conspiring competitors. If we accept therefore that there are valid economic and regulatory reasons why collusive “leader-follower” behavior, such as the pricing alignment by ENI's competitors, should not be considered as participation in an illegal coordination, it seems consequent that the law should instead target ENI's decision to implement a collusive strategy. B. Developing the Legal Framework: Targeting Unilateral Conduct with Collusive Impetus Unilateral conduct that has as its object or effect to promote tacit collusion ought to be prevented. Based on findings of the collusive potential of sticky pricing, we have proved empirically that ENI employed such a pricing policy successfully to bring about a collusive equilibrium in the Italian gasoline market. But is there a feasible way of legal intervention? Should we ban a firm from implementing a policy of sticky pricing because it may facilitate collusion, and punish the firm in case of an infringement? A firm's unilateral behavior may only be regarded as having as its object or effect the promotion of tacit collusion if that is the only reasonable inference—that is, in the absence of any competitive rationale that could convincingly explain the behavior in question, in particular in view of the economic conditions of the market and the effect of the behavior on prices or other market parameters. We submit that under such a doctrine, ENI's pricing strategy should be considered a breach of antitrust laws. The Italian gasoline market90 was characterized by features that indicate its conduciveness to tacit collusion, such as its concentrated oligopolistic market structure, a high price transparency, and entry barriers. More specifically, due to its market share of about 35 percent and the asymmetric distribution of market shares in the Italian petrol industry, ENI clearly held the position as market leader. Thus, its commitment to a strategy of sticky pricing91 resulted in a credible signal to its competitors and entailed a strong potential to encourage them to align their pricing in order to bring about a collusive equilibrium. Given the circumstances of the case and in view of our empirical findings, there is no convincing explanation to defend ENI's pricing strategy as being competitive. It is true that due to better information resources, for example, a market leader may be in a superior position to set the market price and its price leadership may have pro-competitive effects as a result. However, in the case at hand, the only reasonable explanation for ENI's implementation of sticky pricing along with large price changes is that ENI intended to take the position as price leader in order to coordinate prices.92 This is in particular reinforced by our finding that ENI's pricing strategy as a matter of fact resulted in higher and aligned prices. Turning to antitrust provisions that address firms' unilateral behavior, we need to recognize, however, that the law appears to be fragmented—to say the least—when it comes to conduct whose object or effect it is to promote collusion. Neither section 1 of the Sherman Act nor Article 101 of the TFEU embodies an offense of attempted coordination. Section 2 of the Sherman Act and Article 102 of the TFEU, the essential provisions on unilateral conduct, apply first of all to firms with monopoly power or to firms that dominate a market, respectively, and thus based on criteria that typically exclude single oligopolists. It seems, for instance, rather uncertain whether ENI's position may be characterized as “dominant” pursuant to Article 102 of the TFEU. The ECJ presumes predominance where a firm has a market share of 50 percent or more.93 Below this threshold, additional factors must be put forward to show market dominance. Given certain market characteristics, such as the asymmetrically distributed market shares and high barriers to market entry, ENI's position might be considered as being just above the lower bound that defines the area of market dominance.94 Yet, in cases where single firm dominance may be established, conduct such as ENI's, which aims at promoting or sustaining tacit collusion, should be considered abusive. Beyond single market dominance, firms' conduct that aims at promoting or sustaining tacit collusion could be considered as an abuse of collective dominance under Article 102 of the TFEU.95 While the concept of collective dominance, as developed under the EU Merger Regulation, may be transposed to Article 102 of the TFEU, the latter requires an existing dominant position. Consequently, practices that firms strategically employ to reach a collusive equilibrium in the first place may not be covered, and thus the potential of an approach that precisely identifies the shift toward a tacitly collusive equilibrium could not be fully exploited. Practices that facilitate tacit collusion may at most be tackled under Article 102 of the TFEU insofar as they safeguard an already-established collectively dominant position.96 It appears, however, that the European Commission has never initiated any TFEU Article 102 proceedings on grounds of collective dominance. Therefore, as the application of Article 102 of the TFEU to tacit collusion hitherto is only a theoretical option, it remains all the more unclear what conduct could be prohibited under such a doctrine. Under section 2 of the Sherman Act, the prohibition of any “attempt to monopolize” broadens the scope and may allow the inclusion of the unilateral conduct of firms that individually do not hold a monopoly position. Thus, in United States v. American Airlines, Inc.,97 an explicit invitation to collude was considered an infringement of section 2 of the Sherman Act, as the court considered the aggregate market share of offerer and offeree. However, because there is no established case law on unilateral attempts to bring about tacit collusion, and thus a shared monopoly, it remains doubtful whether section 2 of the Sherman Act may apply to such practices. In line with the approach suggested in this article, the Federal Trade Commission (FTC) strove to tackle unilaterally adopted, supposedly facilitating practices under section 5 of the FTC Act. While the U.S. Supreme Court has recognized that this provision may in principle comprise anticompetitive conduct beyond the Sherman Act,98 the FTC's ambition received a decisive99 blow from the decision of the Court of Appeals for the Second Circuit in the Du Pont (Ethyl) case.100 In Ethyl, the FTC blamed four producers of gasoline antiknock compounds of having unilaterally adopted practices that were aimed at facilitating parallel pricing at a supracompetitive level. These practices included 30-day advance announcements of price changes, “most favored nations” clauses in sales contracts, and uniform delivered prices.101 The Court, however, held that the evidence presented by the FTC did not sufficiently support the view that these practices did indeed have an anticompetitive purpose or effect.102 Econometric evidence as suggested in this article could fill such gaps by relating a specific practice with a certain market outcome. With adequate firm level data and a benchmark, an antitrust authority or a court may test whether or not supposedly facilitative practices contributed to a supracompetitive price level. This shows on the one hand that economic methods may support an effective use of available legal instruments to counter unilateral behavior that has as its object or effect to promote collusion. On the other hand, the analysis reveals a significant gap in the arsenal of antitrust enforcement when it comes to targeting unilateral conduct that serves a collusive strategy. Thus, under the current legal framework, the potential of economics to identify the collusive character of specific elements of behavior may not be fully realized. It seems therefore essential to strengthen legal instruments that frustrate unilateral conduct through which firms strive to promote or sustain collusion. V. CONCLUSION Collusion in oligopolistic markets has been a perennial topic both for economics and antitrust law. Antitrust law rests on economic welfare analysis that shows that collusion inflicts substantial negative welfare effects. However, antitrust authorities and private plaintiffs are substantially restricted in their fight against collusion, as they strongly depend on evidence of explicit communication between competitors. The mild reaction of the Italian antitrust authority to the incidents on the Italian gasoline market illustrates the limits of antitrust enforcement in the absence of such evidence. The crucial role attributed to explicit communication in the practice of antitrust enforcement hinders the detection and punishment of cartels precisely in those industries where the collusion rate is expected to be relatively high and communication appears to be less needed. Theoretical and empirical findings on cartel behavior provide a basis to derive clear test hypotheses to distinguish lawful oligopolistic interdependence from tacit collusion. On that basis, econometric evidence may step in and reveal collusive strategies behind firms' actions. Thus, it entails the potential to decisively increase the effectiveness of cartel enforcement in oligopolistic markets. Analyzing the incidents on the Italian gasoline market where the market leader announced it was changing its pricing strategy reveals how firms might use their market power to facilitate price alignment and coordinate price changes. The econometric analysis reveals how the leader's sticky pricing policy coordinated prices, and what effect this had on the price levels with respect to a benchmark: After the new policy was implemented, all competitors adjusted their prices following the leader's price changes. In addition, the new pricing behavior resulted in a significant price increase. Combined, this price coordination mechanism and its effect show that it was the object and effect of the introduced pricing policy to collude through facilitating price coordination and to raise prices. While antitrust enforcement may certainly benefit from an enhanced economic methodology to identify tacit collusion, antitrust law cannot straightforwardly prohibit the participation in tacit collusion as a form of illegal coordination. The active promotion of collusive pricing by ENI and the passive best-response alignment of its competitors must not be normatively equated. Thus, antitrust law should not infer a punishable tacit agreement between ENI and its competitors from the collusive market outcome, but should instead consider conduct, such as ENI's pricing strategy, a unilateral anticompetitive practice. To effectively fight tacit collusion it appears therefore to be necessary to strengthen legal instruments that target unilateral conduct that firms strategically employ to promote or sustain collusion. 1 Throughout this article, we use the term “cartel” to describe any kind of welfare-decreasing form of collusion, be it an explicit or a tacit one, and irrespective of whether or not we consider it an infringement of antitrust law. 2 See Drew Fudenberg & David Tirole, Game Theory 150 (MIT Press 1991). Much of the theoretical discussion on collusion is based on the supergame approach. The best known result describing firms' incentives to collude is the “Folk Theorem,” which states that, for sufficiently low discount rates, almost any price may be sustained as the equilibrium outcome of a repeated game. While the “Folk Theorem” provides fairly general conditions under which tacit collusion may be sustained as an equilibrium, it says nothing about how firms behave in reality. The strategies used in the “Folk Theorem” are chosen because of their analytical ease and not because they describe firms' collusive behavior. 3 See Margaret C. Levenstein & Valerie Y. Suslow, What Determines Cartel Success, 44 J. Econ. Literature 43 (2006). While economic theory shows that concentration facilitates collusion, and thus predicts a positive relation between cartelization rate and market concentration, empirical evidence seems to contradict this result. This gap between the number of cartels predicted from a theoretical perspective and the number of cartels that appear in the empirical analysis may plausibly be explained by a sample-selection bias. Only cartels that, first, have been detected and that, second, were regarded as illegal by antitrust authorities or courts are contained in the sample. 4 Miguel A. Fonseca & Hans-Theo Normann, Explicit vs. Tacit Collusion—The Impact of Communication in Oligopoly Experiments, 56 Eur. Econ. Rev. 1759 (2012). Fonseca and Normann use a laboratory experiment to investigate the role of communication in sustaining collusion. They show that highly concentrated industries collude irrespective of communication. 5 See Gregory J. Werden, Economic Evidence on the Existence of Collusion: Reconciling Antitrust Law with Oligopoly Theory, 71 Antitrust L.J. 719 (2004); Louis Kaplow, An Economic Approach to Price Fixing, 77 Antitrust L.J. 343 (2011) (comprehensively analyzing the use of economics to support cartel enforcement). 6 See William G. Christie & Paul H. Schultz, Why Do Nasdaq Market Makers Avoid Odd-Eighth Quotes?, 49 J. Fin. 1813 (1994). One of the best known examples of economic detection of collusion is provided by the work of Christie and Schultz. They detected collusion between Nasdaq market makers by comparing their bid-ask spread to the equivalent spread on the New York Stock Exchange. Christie and Schultz's work had an impressive impact, as it led to regulatory investigations by the Securities and Exchange Commission (SEC) and class action lawsuits that were settled for over $1 billion. 7 See David Genovese & Wallace P. Mullin, Rules, Communication, and Collusion: Narrative Evidence from the Sugar Institute Case, 91 Am. Econ. Rev. 379 (2001). Genovese and Mullin provide narrative evidence of the role of communication for collusion in the Sugar Institute Case. They find that one key missing aspect in formal theories of collusion is the role for rich communication within the collusive agreement. 8 To use the words of Thomas Hobbes, “the bonds of words are too weak to bridle men's ambition, avarice, anger, and other Passions, without the fear of some coercive Power … ” Thomas Hobbes, The Leviathan 71 (J.M. Dent & Sons 1959) (1651). 9 See, e.g., Joseph E. Harrington, Jr. & Andrzey Skrzypacz, Private Monitoring and Communication in Cartels: Explaining Recent Collusive Practices, 101 Am. Econ. Rev. 2425 (2011) (characterizing an equilibrium in which colluding firms truthfully self-report their sales and then make transfers within the cartel). 10 Louis Phlips, On the Detection of Collusion and Predation, 40 Eur. Econ. Rev. 495 (1996). 11 In this respect, our approach is conceptually in line with Hay, who argues that “if there is to be a category of unlawful tacit collusion which is to be distinguished from classic oligopoly, the difference must lie … on the specific elements of behavior that brought about that state of mind.” See George A. Hay, The Meaning of “Agreement” Under the Sherman Act: Thoughts from the “Facilitating Practices” Experience, 16 Rev. Indus. Org. 113, 128 (2000). 12 Patrick Andreoli-Versbach & Jens-Uwe Franck, Endogenous Price Commitment, Sticky and Leadership Pricing: Evidence from the Italian Petrol Market, 40 Int'l J. Indus. Org. 32 (2015). 13 See infra Figure 1 (plotting daily prices and the major source of cost, that is, the Platts Cif Med, around ENI's new price policy announcement (first vertical line)). Firms respond to cost shocks with some lags. While current costs decreased immediately after ENI's policy, lagged costs increased, and thus competitors increased their prices. 14 Yulia V. Bolotova, Cartel Overcharges: An Empirical Analysis, 70 J. Econ. Behav. Org. 321 (2009); John M. Connor, Collusion and Price Dispersion, 12 Applied Econ. Letters 335 (2005). 15 This conceptual divergence may also give rise to terminological misunderstandings between economists and lawyers. Throughout this article, we will indicate when we use terms such as “collusion” or “agreement” in their technical economic or legal meaning. 16 American Tobacco Co. v. United States, 328 U.S. 781, 809 (1946) (“No formal agreement is necessary to constitute an unlawful conspiracy”); Norfolk Monument Co. v. Woodlawn Memorial Gardens, Inc., 394 U.S. 700, 704 (1969) (“business behavior is admissible circumstantial evidence from which the fact finder may infer agreement”). 17 Case T-41/96, Bayer AG v. Comm'n, 2000 E.C.R. II-3383 ¶ 69, aff'd Case C-2/01 P, Bundesverband der Arzneimittel-Importeure eV & Comm'n v. Bayer AG, 2004 E.C.R. I-23 ¶ 97. 18 Monsanto Co. v. Spray-Rite Servs. Corp., 465 U.S. 752, 768 (1984); see also Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588 (1986). 19 See, e.g., Theatre Enters., Inc. v. Paramount Film Distrib. Corp., 346 U.S. 537, 541 (1954). 20 See Andrew I. Gavil, William E. Kovacic & Jonathan B. Baker, Antitrust Law In Perspective 310 (Thomson West 2d ed. 2008). 21 Matsushita, 475 U.S. at 588 (citing Monsanto Co. v. Spray-Rite Servs. Corp., 465 U.S. 752, 764 (1984)). In this regard, the Supreme Court's decision in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007) has arguably neither amended nor clarified the pleading standard. See William H. Page, Twombly and Communication: The Emerging Definition of Concerted Action Under the New Pleading Standards, 5 J. Competition L. & Econ. 439, 447 (2009) (“To meet the evidentiary sufficiency standard, the plaintiff must produce evidence that tends to exclude the possibility of conscious parallelism.”). 22 Gavil, Kovacic & Baker, supra note 20, at 311. See also Louis Kaplow, On the Meaning of Horizontal Agreements in Competition Law, 99 Cal. L. Rev. 683, 816 (2011) (“[T]his Article does not come close to demonstrating that it would be good policy to proscribe and highly penalize all instances in which interdependent oligopolistic behavior appears to occur. The design of optimal policy is not dictated by definitions but rather by direct assessment of the consequences of different regulatory approaches.”). 23 Roger D. Blair & David L. Kasermann, Antitrust Economics 251 (Oxford Univ. Press 2d ed. 2009) (“The law fails to deal with tacit collusion very effectively. In examining the case law on conscious parallelism, basing-point pricing, and price leadership, we found that no case has held purely tacit behavior to be illegal. The basic problem seems to be that the law on collusion has developed around the fact of agreement rather than the economic effects of collusion.”). 24 See Joined Cases C-189/02 P, C-202/02 P, C-205–208/02 P & C-213/02, P Dansk Rørindustrie A/S & Others v. Comm'n, 2005 E.C.R. I-5425 ¶ 143. Accordingly, the Court infers a tacit approval of a collusive initiative from the mere attendance of a meeting where an anticompetitive agreement was concluded. “That complicity constitutes a passive mode of participation in the infringement which is therefore capable of rendering the undertaking liable in the context of a single agreement … .” Id. 25 Case C-413/06, P Bertelsmann & Sony Corp. of America v. Impala, 2008 E.C.R. I-4951 ¶ 123. 26 Joined Cases 40–48, 50, 54–56, 111, 113 & 114/73, Suiker Unie & Others v. Comm'n, 1975 E.C.R. 1663 ¶ 173; Case 172/80, Züchner v. Bayerische Vereinsbank, 1981 E.C.R. 2021 ¶ 12–14; Case C-8/08, T-Mobile Netherlands & Others v. Comm'n, 2009 E.C.R. I-4529 ¶ 33–35. 27 American Tobacco Co. v. United States, 328 U.S. 781, 810 (1946); Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752, 771 (1984). 28 Case 41/69, ACF Chemiefarma NV v. Comm'n, 1970 E.C.R. 661 ¶ 112. 29 Case T-99/04, AC Treuhand v. Comm'n, 2008 E.C.R. II-1501 ¶ 118. 30 6 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law § 1415c, at 107 (Wolters Kluwer 3d ed. 2010) (citing Blomkest Fertilizer v. Potash Corp., 203 F.3d 1028, 1037 (8th Cir. 2000)). This is presumed if, for example, a certain conduct “is so perilous when not imitated and imitation so uncertain that no reasonable actor would so act, then parallel action does imply some exchange of commitments or at least some comforting assurances connoting a traditional conspiracy.” Id. 31 See Richard A. Posner, Antitrust Law 99 (Univ. of Chi. Press 2d ed. 2001) (“[A] damages judgment in a tacit collusion case would promote competition at a tolerable cost in legal uncertainty and judicial supervision.”). 32 Cf. E.I. Du Pont De Nemours & Co. v. Fed. Trade Comm'n, 729 F.2d 128, 139 (2d Cir. 1984) (“In view of this patent uncertainty the [Federal Trade] Commission owes a duty to define the conditions under which conduct claimed to facilitate price uniformity would be unfair so that businesses will have an inkling as to what they can lawfully do. … The Commission's decision in the present case does not provide any guidelines; it would require each producer not only to assess the general conduct of the antiknock business but also that of each of its competitors and the reaction of each to the other, which would be virtually impossible.”). 33 See Joined Cases C-189/02 P, C-202/02 P, C-205–208/02 & C-213/02, Dansk Rørindustrie A/S & Others v. Comm'n, 2005 E.C.R. I-5425 ¶ 202. Under European law, Article 7(1) ECHR enshrines the principle that offenses and punishments are to be strictly defined by law. 34 Cf. United States v. United States Gypsum Co., 438 U.S. 422, 435 (1978) (“We agree with the Court of Appeals that an effect on prices, without more, will not support a criminal conviction under the Sherman Act. … [A] defendant's state of mind or intent is an element of a criminal antitrust offense which must be established by evidence and inferences drawn therefrom, and cannot be taken from the trier of fact through reliance on a legal presumption of wrongful intent from proof of an effect on prices.”). As to the required standard of intent the Court concluded “that action undertaken with knowledge of its probable consequences and having the requisite anticompetitive effects can be a sufficient predicate for a finding of criminal liability under the antitrust laws” (Id. at 444). 35 See Council Regulation (EC) No. 1/2003 of 16 December 2002 on the Implementation of the Rules on Competition Laid Down in Articles 81 and 82 of the Treaty, 2003 O.J. (L 1) 1, http://eur-lex.europa.eu/legal-content/EN/ALL/?uri=celex:32003R0001. 36 See Joseph E. Harrington, Jr., How do Cartels Operate, 2 Found. & Trends Microeconomics 1 (2006) (analyzing case studies on cartels); see also Levenstein & Suslow, supra note 3, (conducting a meta-study on cartels' features); George A. Hay and Daniel Kelley, An Empirical Survey of Price-Fixing Conspiracies, 17 J.L. & Econ. 13 (1974); Margaret C. Levenstein & Valerie Y. Suslow, Breaking Up Is Hard to Do: Determinants of Cartel Duration, 54 J.L. & Econ. 455 (2011). 37 See Eric Zitzewitz, Forensic Economics, 50 J. Econ. Literature 731 (2012); Jay R. Ritter, Forensic Finance, 22 J. Econ. Persp., no. 3, Summer 2008, at 127. 38 Brian A. Jacob & Steven D. Levitt, Rotten Apples: An Investigation of the Prevalence and Predictors of Teacher Cheating, 118 Q.J. Econ. 843 (2003). 39 Stefano Dellavigna & Eliana La Ferrara, Detecting Illegal Arms Trade, 2 Am. Econ. J. 26 (2010). 40 Marianne Bertrand, Dolly Chugh & Sendhill Mullainathan, Implicit Discrimination, 95 Am. Econ. Rev. 94 (2005). 41 Rosa Abrantes-Metz & Patrick Bajari, Screens for Conspiracies and Their Multiple Applications, 24 Antitrust 66 (2009). 42 See, e.g., Patrick Bajari & Lixin Ye, Deciding Between Competition and Collision, 85 Rev. Econ. & Stat. 971 (2003) (developing an approach to identify and test for bid rigging in procurement auctions); Robert H. Porter, Detecting Collusion, 26 Rev. Indus. Org. 147 (2005); Joseph E. Harrington, Jr., Detecting Cartels, in Handbook in Antitrust Economics (Paolo Buccirossi ed., MIT Press 2008); Patrick Rey, On the Use of Economic Analysis in Cartel Detection, in European Competition Law Annual 2006, Enforcement of Prohibition of Cartels (Calus-Dieter Ehlermann & Isabela Atanasiu eds., Hart Publishing 2007) (generally discussing methods to detect collusion); Hans W. Friederiszick & Frank P. Maier-Rigaud, Triggering Inspections Ex Officio: Moving Beyond a Passive EU Cartel Policy, 4 J. Competition L. & Econ. 89 (2008) (arguing for a greater role of ex officio investigations based, for example, on economic screens). 43 George J. Stigler, A Theory of Oligopoly, 72 J. Pol. Econ. 44 (1964). 44 See, e.g., Edward J. Green & Robert H. Porter, Noncooperative Collusion Under Imperfect Price Information, 52 Econometrica 87 (1984); Julio J. Rotemberg & Garth Saloner, A Supergame - Theoretic Model of Price Wars During Booms, 76 Am. Econ. Rev. 390 (1986); Eric Maskin & Jean Tirole, A Theory of Dynamic Oligopoly, II: Price Competition, Kinked Demand Curves, and Edgeworth Cycles, 56 Econometrica 571 (1988). 45 Julio J. Rotemberg & Garth Saloner, Collusive Price Leadership, 39 J. Indus. Econ. 93 (1990). 46 Igor Mouraviev & Patrick Rey, Collusion and Leadership, 29 Int'l J. Indus. Org. 705 (2011). 47 See Susan Athey & Kyle Bagwell, Optimal Collusion with Private Information, 32 RAND J. Econ. 428 (2001); see also Susan Athey & Kyle Bagwell, Collusion with Persistent Cost Shocks, 76 Econometrica 493 (2008); Susan Athey, Kyle Bagwell & Chris Sanchirico, Collusion and Price Rigidity, 71 Rev. Econ. Stud. 317 (2004); Makoto Hanazono & Huanxing Yang, Collusion, Fluctuating Demand, and Price Rigidity, 48 Int'l Econ. Rev. 483 (2007); Luke Garrod, Collusive Price Rigidity Under Price-Matching Punishments, 30 Int'l J. Indus. Org. 471 (2012). 48 See Rosa M. Abrantes-Metz, Luke M. Froeb, John Geweke & Christopher T. Taylor, A Variance Screen for Collusion, 24 Int'l J. Indus. Org. 467 (2006); see also Korbinian Blanckenburg, Alexander Geist & Konstantin A. Kholodilin, The Influence of Collusion on Price Changes: New Evidence from Major Cartel Cases, 13 German Econ. Rev. 245 (2012); Connor, supra note 14. 49 Press Release, AGCM (Italian Competition Authority), Petrol Prices: Antitrust Authority Launches Investigation into Nine Oil Companies over Possible Anti-competitive Pricing Agreement (Jan. 18, 2007), available athttp://www.agcm.it [hereinafter Petrol Pricing Competition Investigation]. 50 In our setting, cost changes are persistent and, thus, the strategic decision of firms concerns when to change prices and adapt to the new cost level emerging from past cost changes. 51 See Petrol Pricing Competition Investigation, supra note 49. ERG publicly declared that it would not follow ENI's new pricing and stick to its own method, which it did not further specify. 52 Id. 53 Id. 54 The Platts Company is a leading global provider of energy information that collects and publishes details on the prices of bids on a daily basis and offers for specialized oil products and regions from traders and exchange platforms. 55 See Italian Petrol Union, http://www.unionepetrolifera.it/it/show/34/La%20struttura%20del%20prezzo (defining gross margin as the difference between the retail price net of taxes and the Platts Cif Med). 56 Petrol Pricing Competition Investigation, supra note 49. 57 Id. 58 To conceptualize the reciprocal pricing announcements as an act of “concertation” would in turn trigger a presumption according to which such an indirect information exchange did indeed affect the subsequent pricing behavior of the participating firms, see Case C-199/92, P Hüls v. Comm'n, 1999 E.C.R. I-4287 ¶ 162; Case C-8/08, T-Mobile Netherlands & Others v. Comm'n, 2009 E.C.R. I-4529 ¶ 52. Cf. Joined Cases C-89/85, C-104/85, C-116/85, C-117/85 & C-125–129/85, Ahlström Osakeyhtiö & Others v. Comm'n, 1993 E.C.R. 1307 ¶ 75–79 & 126 (The case could arguably be distinguished from the ECJ's judgment in “Wood Pulp” as it seems rather doubtful that ENI and its competitors may invoke legitimate business interests for making their future pricing public); European Commission, Guidelines on the Application of Article 101 of the Treaty on the Functioning of the European Union to Horizontal Co-operation Agreements, 2011 O.J. (C 11) 1, http://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:52011XC0114%2804%29 (“However, depending on the facts underlying the case at hand, the possibility of finding a concerted practice cannot be excluded, for example in a situation where … [a public] announcement was followed by public announcements by other competitors, not least because strategic responses of competitors to each other's public announcements (which, to take one instance, might involve readjustments of their own earlier announcements to announcements made by competitors) could prove to be a strategy for reaching a common understanding about the terms of coordination.”). On November 22, 2013, the European Commission announced that it had opened proceedings against container liner shipping companies on the grounds that “the companies have been making regular public announcements of price increase intentions through press releases on their websites and in the specialised trade press.” See Press Release, European Commission, Antitrust: Commission Opens Proceedings Against Container Liner Shipping Companies (Nov. 22, 2013), http://europa.eu/rapid/press-release_IP-13-1144_en.htm. 59 See Joseph E. Harrington, Jr., Posted Pricing as a Plus Factor, 7 J. Competition L. & Econ. 1 (2011) (identifying market conditions under which the use of posted pricing is inconsistent with competition). 60 Given market conditions such as those on the Italian petrol market that are characterized by relatively high price transparency, an implementation of a collusive pricing strategy is also conceivable without the use of facilitative devices, such as announcements of future price steps, despite the fact that under such circumstances an implementation could not occur that smoothly. The approach suggested in this paper including an econometric analysis could also then demonstrate that an observable pricing policy would have to be characterized as collusive. 61 Athey & Bagwell, supra note 47; Athey, Bagwell & Sanchirico, supra note 47; Hanazono & Yang, supra note 47; Garrod, supra note 47. 62 The concepts of “rigid” and “sticky” pricing differ. Rigid pricing refers to a market environment with independent and identically distributed costs where firms' prices are unresponsive to current cost changes. In contrast, sticky pricing refers to settings with permanent cost changes, such as the gasoline market. In those markets the issue is when to change the price and to respond to permanent cost changes. 63 Athey & Bagwell, supra note 47. It should be noted, however, that with other parameter configuration other, more complex type of equilibria are possible. 64 While these are general oligopolistic models and thus none of them perfectly matches the Italian gasoline industry, they provide a sound theoretical justification of observed rigid pricing patterns in cartels. 65 Zhongmin Wang, (Mixed) Strategy in Oligopoly Pricing: Evidence from Gasoline Price Cycles Before and Under a Timing Regulation, 117 J. Pol. Econ. 987 (2009). 66 Abrantes-Metz, Froeb, Geweke & Taylor, supra note 48; Blanckenburg, Geist & Kholodilin, supra note 48; Connor, supra note 14. 67 Abrantes-Metz, Froeb, Geweke & Taylor, supra note 48. 68 Blanckenburg, Geist & Kholodilin, supra note 48. 69 Frederic M. Scherer & David Ross, Industrial Market Structure and Economic Performance 346 (Houghton Mifflin Co. 3d ed. 1990). 70 Rotemberg & Saloner, supra note 45. 71 Mouraviev & Rey, supra note 46. 72 Andreoli-Versbach & Franck, supra note 12. 73 See Paul R. Zimmerman, The Competitive Impact of Hypermarket Retailers on Gasoline Prices, 55 J.L. & Econ. 27 (2012). EU countries differ with respect to Italy, for example, in the number of gas stations owned by hypermarkets that compete aggressively to attract customers to their stores. Using state-level data of U.S. gasoline prices, Zimmerman shows the positive competitive impact of hypermarket retailers. The dif-in-dif analysis as carried out in Andreoli-Versbach & Franck, supra note 12, assumes that “market trends” would be the same in the treatment and control group, while structural country specific market differences are captured by the fixed effects. 74 See infra Figure II (plotting average weekly prices in Italy, the EU, and the Brent). 75 Alberto Abadie & Javier Gardeazabal, The Economic Costs of Conflict: A Case Study of the Basque Country, 93 Am. Econ. Rev. 112 (2003). 76 Alberto Abadie, Alexis Diamond & Jens Hainmueller, Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California's Tobacco Control Program, 105 J. Am. Stat. Ass'n 493 (2010). 77 Both specification (2) and specification (3) were performed using robust standard errors. 78 See Giancarlo Spagnolo, Managerial Incentives and Collusive Behavior, 49 Eur. Econ. Rev. 1501, 1502 (2005). Spagnolo focuses on the role of observable CEO compensation schemes with regard to tacit collusion. He concludes that “a strong pro-collusive effect may well outweigh agency costs and transform apparently puzzling compensation practices into profitable ‘governance’ instruments.” 79 See David Besanko, David Dranove, Mark Shanley & Scott Schaefer, Economics of Strategy (Wiley 5th ed. 2010). One of the standard textbooks used in MBA courses that deals with competitive strategy is “Economics of Strategy.” Chapters 9 and 10 extensively deal with the issues of “Strategic Commitment” and “The Dynamics of Pricing Rivalry,” respectively, which are key elements to sustain collusion. Under the heading “The golden age of micro,” the journal “The Economist” discussed in its issue of October 19, 2012 why leading academic microeconomists are top advisers at firms such as Microsoft and Amazon. 80 Thomas C. Schelling, The Strategy of Conflict (Harvard Univ. Press 1960). 81 See Maskin & Tirole, supra note 44 (building commitment in a dynamic Bertrand model through exogenous costs such as menu costs and showing that sticky prices can serve as a commitment device to sustain higher prices than under static Bertrand). See also Wang, supra note 65 (studying firms' pricing strategies in a gasoline market before and after the introduction of a law which regulated firms' timing of price changes and highlighting the importance of short-run price commitment in tacit collusion). 82 See Figure 1. 83 Severin Borenstein, James B. Bushnell & Frank A. Wolak, Measuring Market Inefficiencies in California's Restructured Wholesale Electricity Market, 92 Am. Econ. Rev. 1376 (2002). 84 See Massimo Motta & Michele Polo, Leniency Programs and Cartel Prosecution, 29 Int'l J. Indus. Org. 705 (2003) (discussing leniency theoretically); see also Joseph E. Harrington, Jr., Optimal Corporate Leniency Program, 56 J. Indus. Econ. 215 (2008); Nathan H. Miller, Strategic Leniency and Cartel Enforcement, 99 Am. Econ. Rev. 750 (2009) (providing empirical evidence on the effects of leniency). 85 See Nicholas Petit, The Oligopoly Problem in EU Competition Law, in Handbook on European Competition Law 259, 314 (Ioannis Lianos & Damien Geradin eds., Edward Elgar 2013) (identifying, with a view to the EU Merger Regulation and, in particular, considering its application by the European Commission, substantive, procedural, and remedial issues that reveal significant shortcomings of merger control as a tool to address tacit collusion). 86 See William E. Kovacic, Robert C. Marshall, Leslie M. Marx & Halbert L. White, Plus Factors and Agreement in Antitrust Law, 110 Mich. L. Rev. 393, 435 (2011) (offering a list of “super plus factors,” including inter alia, “[a] reliable predictive econometric model that accounts for all material noncollusive effects on price, estimated using benchmark data where conduct was presumed noncollusive, produces predictions of prices that do not explain the path of actual prices in the period or region of potential collusion, at a specified high confidence level”). 87 See supra Part II. 88 6 Phillip E. Areeda, Roger D. Blair, Herbert Hovenkamp & Christine Piette Durrance, Antitrust Law § 1347, at 198 (Wolters Kluwer 3d ed. 2010). 89 In the U.S., several courts have recognized such claims for “umbrella damages.” See, e.g., Loeb Indus., Inc. v. Sumitomo Corp., 306 F.3d 469 (7th Cir. 2002); In re Beef Indus. Antitrust Litigation, 600 F.2d 1148 (5th Cir. 1979). The European Court of Justice held that Article 101 TFEU precludes the interpretation and application of domestic legislation enacted by a Member State which categorically excludes, for legal reasons, any civil liability of undertakings belonging to a cartel for loss resulting from umbrella pricing. See Case C-557/12, Kone AG and Others v. ÖBB-Infrastruktur AG, 2014. 90 See supra Parts III.A and III.B. 91 There would be nothing inherently new in prohibiting a certain pricing behavior. Market dominant firms are not allowed to engage in predatory pricing. And just as it has to be defined with regard to a specific industry whether a certain pricing policy has to be considered “predatory,” courts would also have to define “sticky pricing” industry-specifically as infrequent price changes in response to changes of input costs or demand patterns. 92 See supra Part III.E. 93 Case C-62/86, AKZO v. Commission 1991 E.C.R. I-3359 ¶ 60. 94 In British Airways v. Commission, the General Court (then known as the Court of First Instance) confirmed the Commission's view that British Airways was dominant in the UK market for the procurement of air travel agency services with a market share of 39.7 percent, Case T-219/99, British Airways v. Commission 2003 E.C.R. II-5917 ¶ 189-225. Market dominance was not contested at appeal, Case C-95/04 P British Airways v. Commission 2007 E.C.R. I-2331 ¶ 14-15. 95 See, e.g., Sigrid Stroux, US and EC Oligopoly Control 168 (Kluwer Law International 2004); Valentine Korah, An Introductory Guide to EC Competition Law and Practice 126 (Hart Publishing 9th ed. 2007); Giorgio Monti, EC Competition Law 335 (Cambridge Univ. Press 2007); Petit, supra note 85, at 334. 96 Monti,supra note 95, at 341; Petit, supra note 85, at 336. 97 743 F.2d 1114 (5th Cir. 1984). See, e.g., United States v. Ames Sintering Co., 927 F.2d 232 (6th Cir. 1990) (charging explicit attempts to initiate collusion as violations of the wire fraud or mail fraud statutes). 98 See, e.g., FTC v. Indiana Federation of Dentists, 476 U.S. 447, 454 (1986); FTC v. Sperry & Hutchinson, 405 U.S. 233, 244 (1972); FTC v. Brown Shoe Co., 384 U.S. 316, 320-321 (1966). 99 Two more recent enforcement actions by the FTC which are commonly referred to as the Valassis case and the U-Haul case (and both of which were resolved without litigation through a negotiated consent decree) do not indicate otherwise as both cases involved express unilateral efforts to engage in collusion, see Administrative Complaint, Valassis Comm'ns, Inc., FTC Dkt. No. C-4160 (Mar. 14, 2006), available at http://www.ftc.gov/os/caselist/0510008/060314cmps051008.pdf and In the Matter of U-Haul Int'l and AMERCO, 2010 WL 2966779 (F.T.C.). 100   E.I. Du Pont De Nemours & Co. v. Fed. Trade Comm'n, 729 F.2d 128 (2d Cir. 1984). 101 Id. at 133. 102 Id. at 139–140. © The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Competition Law & Economics Oxford University Press

ECONOMETRIC EVIDENCE TO TARGET TACIT COLLUSION IN OLIGOPOLISTIC MARKETS

Preview Only
30 pages

Loading next page...
 
/lp/ou_press/econometric-evidence-to-target-tacit-collusion-in-oligopolistic-QI013DeD3r
Publisher
Oxford University Press
Copyright
© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
1744-6414
eISSN
1744-6422
D.O.I.
10.1093/joclec/nhv011
Publisher site
See Article on Publisher Site

Abstract

Abstract Tacit collusion may reduce welfare comparably to explicit collusion, but it remains mostly unaddressed by antitrust enforcement that greatly depends on evidence of explicit communication. We propose to target specific elements of firms' behavior that facilitate tacit collusion by providing quantitative evidence that links these actions to an anticompetitive market outcome. We apply our approach to incidents on the Italian gasoline market, where the market leader unilaterally announced its commitment to a policy of sticky pricing and large price changes that facilitated price alignment and coordination of price changes. Antitrust policy must distinguish such active promotion of a collusive strategy from passive, best-response, alignment. Our results imply the necessity of stronger legal instruments that target unilateral conduct that aims at bringing about collusion. I. INTRODUCTION In most markets, firms quickly realize that they can earn supracompetitive profits by coordinating their market conduct. In response, antitrust policy seeks to foster “effective competition” by targeting collusive activities. The current legal framework to accomplish this goal has mainly evolved around communication as a means to reach a collusive agreement. In contrast, purely tacit collusion remains largely unaddressed by antitrust law, though it may bring about the same negative welfare effects. We argue that a crucial step forward in targeting tacit collusion could be taken through the forensic use of econometric evidence that may reveal collusive strategies. Theoretical and empirical findings on collusive behavior provide a basis for deriving clear test hypotheses to distinguish lawful oligopolistic interdependence from tacit collusion. Thus, econometric analyses may provide quantitative evidence that firms strategically use specific elements of market conduct to tacitly collude. Antitrust remedies should in turn take up such instances of market behavior to tackle tacit collusion. The paramount significance of evidence of explicit communication entails fundamental problems for the fight against cartels.1 Communication is not a necessary condition to collude. At the heart of collusion lies the incentive of firms to cooperate rather than to compete.2 In oligopolies, firms can exercise their unilateral market power to facilitate anticompetitive coordination without engaging in communication. As firms weigh the costs and benefits of explicit collusion, antitrust law's focus on communication incentivizes them to concentrate on tacit means of collusion. Legal instruments to counter collusion, the effectiveness of which depends on evidence of explicit communication, are least effective in concentrated industries3—that is, precisely in those industries where the cartelization rate is presumably the highest and communication is least needed to sustain collusion.4 Any economic approach to support the enforcement of antitrust law5 is challenged by a legal significance of evidence of communication. Economists can use observable variables, such as prices and their knowledge of the strategies employed by firms to infer collusion,6 but have no instruments to prove whether firms collude with or without communication. From an incentive-based perspective, illegal communication appears to be of relative unimportance; while non-enforceable communication might facilitate coordination on a particular collusive equilibrium,7 “talk is cheap” in the absence of effective enforcement mechanisms.8 This does not, however, mean that economic theory regards communication as uninformative. On the contrary, there is a large amount of literature on the conditions under which costless communication is informative.9 The point is that firms collude tacitly or explicitly because it is in their best interest to do so. If antitrust law is confined to addressing communication that facilitates collusion, or through which firms collude, firms will shift to other means of coordination that potentially have similar effects on welfare. It is, however, not out of economic naivety that antitrust law concentrates so much on evidence of communication in its struggle against collusion. First, this reflects skepticism about whether instances of tacit collusion may be distinguished from oligopolistic competition with a degree of precision that suffices for forensic purposes. This concern may be associated with the so-called “indistinguishability problem,” as put forward by Louis Phlips.10 He suggested that game theoretic arguments, combined with the unavailability of some key data, can make an economic based proof of collusion very difficult, as something that looks like collusion might stem from a multiplicity of indistinguishable equilibria. Thus, the application of any legal instrument that addresses tacit collusion faces the challenge to prevent an unacceptable high number of false positives. Secondly, for purposes of antitrust enforcement it does not suffice to show that an observable market outcome emerged as the result of a collusive strategy. Antitrust remedies may not straightforwardly tackle firms because they charge “collusive”—that is, supracompetitive—prices but must address specific elements of firms' market conduct that may be characterized as collusive. Without taking into account these issues, antitrust enforcement that tackles tacit collusion risks either unduly restricting market operators' leeway to compete or ultimately amounting to an instrument of price control. In the following, we outline an approach that addresses both these concerns, and thus provides the basis for an expansion of the law's ambition to tackle tacit collusion. Oligopolistic interdependence as such and oligopolistic collusion are conceptually distinct. Tacit collusion arises from decisions endogenous to the market by one or several firms that aim to reduce or eliminate competition. In contrast, oligopolistic interdependence stems from best response to market conditions, including other firms' behavior, that favor non-competitive performance. Thus, while the market outcome might appear to be “indistinguishable,” the specific strategies that lead to the outcome differ significantly. The gist of our approach to identify collusive behavior lies in an identification of patterns of behavior used by firms to bring about or facilitate tacit collusion.11 Antitrust law must not simply infer the existence of a punishable tacit agreement from the insight that a certain market outcome is the result of a collusive strategy. Rather, it is essential to distinguish the active promotion of a collusive strategy by one firm from the passive, best-response alignment of competing firms. Consequently, antitrust enforcement should not conceptualize such instances of collusive leader-follower behavior as an illegal coordination that would—with regard to the “followers”—result in punishing oligopolistic interdependence. Rather, antitrust law should capture such instances of “unilateral collusion” only through considering as illegal the unilateral conduct that actively promotes the implementation of a collusive strategy. To effectively fight tacit collusion, it is therefore necessary to strengthen legal instruments that target the unilateral conduct that firms strategically employ to promote collusion. To illustrate our behavioral approach to tackling tacit collusion and to demonstrate the capacity of econometric evidence, we refer to incidents on the Italian gasoline market. In a recent article we provide quantitative evidence of the means—that is, specific pricing strategies—and the effects—that is, higher prices—caused by a unilateral public announcement of ENI, the market leader.12 On October 6, 2004, ENI announced a new pricing policy that consisted of infrequent price variations (sticky pricing) and large price changes. Using daily firm level prices of gasoline in Italy and average weekly EU prices over the time period from January 2003 to May 2005, we show the effect of the new pricing policy. ENI increased the time lag between price changes from 6 to 20 days and increased the mean price change from 1 percent to 5.4 percent. After the policy change, ENI did not change its price for 57 days irrespective of cost changes. Initially ENI's competitors kept their short-run cost-based pricing and thus increased their prices following lagged cost increases.13 Once competitors started to align with ENI in mid-November 2004, a different pricing pattern emerged: sticky-leadership pricing. Competitors matched each large price variation and ENI endogenously emerged as the price leader in the market and coordinated price changes. While the first effect of the policy was to change the price interdependence in the Italian gasoline market, this newly emerged tacit coordination had the additional effect of a significant price increase. Using several estimation techniques, we show that Italian prices rose compared to EU prices after the new sticky leadership pricing emerged. Thus, the econometric analysis used to characterize pre-policy and post-policy pricing behavior and evaluate the effect of the new market conduct on the price level might provide solid “statistical” evidence that ENI's unilateral commitment to a policy of sticky pricing must be characterized as collusive. The structure of the article is as follows: Part II outlines the status quo of cartel enforcement that focuses on firms' communication and the law's difficulties with tackling tacit collusion. In Part III, we discuss incidents on the Italian gasoline market as an illustration for how our approach might be applied for purposes of antitrust enforcement. Part IV describes the way to integrate quantitative evidence of collusion with antitrust law. Part V concludes. II. ON COLLUSION AS A LEGAL CONCEPT, ITS LIMITS IN THE ABSENCE OF EVIDENCE OF COLLUSIVE COMMUNICATION, AND THE REASONS THEREFORE Collusion allows competing firms to charge supracompetitive prices and entails negative welfare effects. Meta-studies on cartel overcharges show that the median cartel-price increase ranges between 20 and 30 percent.14 This is why antitrust law aims at inhibiting collusion and why the horizontal coordination of prices and quantities is considered a per se violation of section 1 of the Sherman Act and a restriction of competition by object pursuant to Article 101 of the Treaty on the Functioning of the European Union (TFEU), respectively. Successful collusion requires inter alia an underlying—tacit or explicit—consensus on the terms of the cooperation. Thus, in order to counter collusion, it seems a logical step to regard such underlying understanding as illegal. However, the economic conception of a collusive agreement diverges significantly from the corresponding legal concepts of “conspiracy” according to section 1 of the Sherman Act or “agreement” and “concerted practice” according to Article 101(1) of the TFEU.15 While the former focuses on firms' incentives to engage in collusion and their strategies for sustaining a collusive equilibrium, the latter centers around the means to reach an understanding between firms. This divergent perspective on collusion becomes apparent with regard to instances of tacit collusion—that is, under circumstances where firms sustain a collusive market outcome without direct communication and where, therefore, no direct evidence (such as written records or insider testimony) of an exchange of assurances concerning a coordination of future market conduct is available. Though, as a matter of principle, both under the Sherman Act and the TFEU, circumstantial evidence may suffice to demonstrate the existence of a “conspiracy”16 or an “agreement,”17 respectively, there are doctrinal limits in this regard if it comes to supposedly tacit collusion between competitors. In the words of the U.S. Supreme Court, “conspiracy” requires “that [the defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective.”18 Reasonably, this may not be inferred from conscious parallelism alone.19 Rather, a plaintiff must produce additional evidence to prove that an observed parallel market conduct may not be considered the result of mere oligopolistic interdependence, but indeed forms part of a collusive strategy. Such so-called “plus factors” may encompass, first, elements of industry structure that indicate that an industry is conducive to collaboration, second, conduct that appears irrational or inefficient absent collusion, and, third, additional factors such as industry performance (for example, stabile market shares over time, supracompetitive pricing) or facilitating practices (for example, exchange of information).20 Although the U.S. Supreme Court has stated that plaintiffs can only survive summary judgment by presenting circumstantial evidence “that tends to exclude the possibility that the alleged conspirators acted independently,”21 the case law so far does not provide a taxonomy of plus factors that would allow us to determine which elements of evidence are required to infer an agreement. It has thus been concluded that “decisions analyzing plus factors generally have failed to establish a clear boundary between tacit agreements—to which section 1 applies—and parallel pricing stemming from oligopolistic interdependence … This condition makes judgments about future litigation outcomes unpredictable.”22 Observers have noted, however, that up until now “no case has held purely tacit behavior to be illegal.”23 Although the European Court of Justice (ECJ) considers it generally conceivable that consent to an agreement may be inferred from circumstantial evidence,24 the Court is reluctant to infer an “agreement” between competitors from their market conduct alone, notwithstanding the presence of certain “plus factors.” Given the current state of the jurisprudence, it appears that in the absence of direct evidence of collusion the Court does not presume the existence of an “agreement,” even if one has proved that observed parallel market conduct was an expression of tacit collusion rather than of oligopolistic interdependence. This has been reaffirmed by a decision on the doctrine of “collective dominance” under Article 2 EU Merger Regulation where the Court implicitly approved that tacit collusion per se may not fall under Article 101(1) of the TFEU: “Unless they can form a shared tacit understanding of the terms of the coordination, competitors might resort to practices that are prohibited by Article [101 TFEU] in order to be able to adopt a common policy on the market.”25 However, where tacit collusion has been induced by facilitating practices, such as an exchange of information, it may come under Article 101(1) of the TFEU as an illegal “concerted practice.” In this regard, the ECJ drew a line. On the one hand, by assigning market operators the legal leeway to “adapt themselves intelligently to the … conduct of their competitors,” the Court signaled that mere passive alignment would not be treated as an illegal form of coordination. On the other hand, the Court submitted that a strategy that actively aims at aligning competitors' market conduct may fall under Article 101(1) of the TFEU.26 Thus, to implement this standard, it is essential to identify elements of behavior that promote tacit collusion. This brief insight into legal concepts of coordination reveals ambiguities and restrictions with regard to tacit collusion. It invites one to question why the law finds it so difficult to cope with this phenomenon, given that it seems uncontroversial in terms of competition policy that tacit collusion on prices and quantities should be prevented as rigorously as collusion based on explicit consensus. The respective judicial definitions of “conspiracy” and “agreement” do not restrict these concepts in a way that would exclude collusion that has been sustained tacitly. Whatever the rhetoric of the courts might be when they characterize the requirements of an agreement—typically they refer to a need to show a “meeting of minds,”27 a “joint intention,”28 or a “concurrence of wills”29—the respective antitrust law concepts must be defined strictly instrumentally. Thus it is, first, the underlying policy to contain as far as possible any kind of welfare-reducing collusion and, second, the role a legal intervention and, in particular, a prohibition of agreements between competitors may feasibly play in this regard that determine which behavior should be regarded as illegal. Part of the law's problem in coping with tacit collusion lies with the difficulty to distinguish collusion from oligopolistic interdependence, as the latter may also result in suspiciously parallel market conduct and supracompetitive prices. This problem is addressed by the requirement of “plus factors” that—in addition to parallel pricing—are meant to indicate collusion, such as market conduct that may reasonably only be explained as part of a collusive strategy.30 From this perspective, the problem of distinguishing oligopolistic collusion from oligopolistic competition comes down to a question of error costs. By defining the “critical mass” of plus factors required to infer an illegal coordination, courts strike a balance between the ambition to contain tacit collusion and the risk of producing false positives.31 However, the courts' reluctance to infer an agreement in cases of mere tacit collusion suggests that there is more to the law's difficulties to cope with tacit collusion than the problem of multiple indistinguishable equilibria and the issue of reaching an acceptable degree of error costs in this regard. Legal standards and remedies that are supposed to influence market conduct in order to guarantee effective competition may not simply prohibit an undesired economic condition, such as a collusive equilibrium, and punish firms because they charge “collusive” prices. Such a policy effectively meant nothing other than price control. This unwelcome consequence is prevented when antitrust standards and remedies relate to individual behavior and define which acts or omissions are required or prohibited. When authorities or private plaintiffs order a firm to bring an infringement to an end or seek to obtain injunctions before a court, it is already the remedy's behavioral nature that requires a specification of elements of conduct that violate antitrust law. The intended deterrent effect of concurring remedies, such as imposing fines or damages, likewise depends on whether market operators are in a position to foresee what conduct they may be sanctioned for, and how they are expected to behave to avoid sanctions. This appears particularly challenging where an undesired economic effect or market condition is the consequence of the interdependent behavior of several market actors.32 But once again, if the elements of behavior that bring about a collusive equilibrium remain unclear, any legal intervention may ultimately amount to a price control by antitrust authorities or courts. Furthermore, with regard to criminal and quasi-criminal sanctions it is required by the principle of culpability33 and the need to prove intent34 or negligence,35 respectively, that antitrust enforcement ensures that market operators may anticipate their legal leeway and addresses certain modes of behavior rather than an economic effect or condition. Thus, the key to overcoming the law's difficulties to counter tacit collusion lies in an approach that identifies specific elements of behavior with the object or effect to bring about or facilitate collusion. Such an approach has a chance for success as market operators that seek to implement a collusive strategy need to adjust their market conduct to reach an optimal and stable collusive equilibrium. Even in oligopolistic markets that are characterized by features that facilitate tacit collusion, prices and other parameters have to be adjusted according to an underlying tacit agreement, and the need for such adjustments may lead firms to resort to a certain behavior that may be identified as serving a collusive strategy. Empirical and theoretical research36 on how cartels behave provides solid test hypotheses to identify such elements of collusive behavior. Precisely these elements of behavior are the focus of our approach to provide evidence of anticompetitive behavior. III. EMPIRICAL EVIDENCE Academic forensic economics and finance37 has long applied its tools in a number of areas to reveal conduct that agents strive to conceal. Some of the most prominent examples include teachers cheating in exams,38 violations of U.N. sanctions,39 and racial biases in employment decisions.40 This research is methodologically related to our topic of empirical cartel detection, as econometrics is employed to provide evidence of hidden wrongdoings. In academic forensic economics and finance, researchers use their knowledge of incentive schemes on observable variables, such as prices, in order to derive statistical tests to compare distinct hypotheses, such as collusion versus competition. While a test hypothesis for teachers to raise students' test scores or employment discrimination on the basis of race can be clearly defined, what should constitute an appropriate test for collusion? In line with the literature on economic screens41 we believe that the answer lies in economic theory and empirical evidence on cartel behavior.42 Since the foundational work by George Stigler,43 who highlighted firms' incentive to cheat as the preeminent challenge faced by cartels, much research has been carried out on “pricing structures” that can sustain a collusive outcome.44 The two key strategic aspects that are relevant for our analysis are the use of sticky and leadership pricing as a facilitating device to sustain collusion. The role of price leadership as a collusive strategy to coordinate on the focal price of the leader has been analyzed by Julio Rotemberg and Garth Saloner45 and Igor Mouraviev and Patrick Rey.46 The relation between price rigidity and collusion has been the focus of many theoretical articles using different settings.47 In addition, there is a growing empirical literature that uses data on detected cartels to test whether pricing behavior significantly changes during a cartel situation. For example, some scholars show that price stickiness is associated with collusive behavior.48 A. The Facts of the Case On October 6, 2004, ENI, the market leader in the Italian gasoline market, publicly announced the adoption of a new pricing policy. ENI declared that the purpose of this policy was to lower the short-term price-cost relation and to stabilize retail prices.49 While prices used to follow short-run cost changes, ENI's aim was to significantly lower the price responsiveness to persistent cost changes.50 As the volatility of crude oil was increasing, ENI claimed that the policy aimed at lowering the retail price variability would benefit customers. The new policy consisted in a reduction of the number of price changes relative to cost changes—that is, sticky pricing—and increased the magnitude of each variation. ENI increased the average time lag between price changes from 6 to 20 days and increased the mean price change from 1 percent to 5.4 percent. The result of this declaration can be seen in Figure 1, which shows the daily price per company over time before and after the introduction of the new pricing policy. Before the policy, firms' price changes were frequent. On average, firms changed their prices every five days. The average price change was 0.9 percent before the policy change. After the new pricing policy was introduced, price changes occurred infrequently, every 9 days on average, but their amount became larger, 2.4 percent on average. Figure 1. View largeDownload slide Cartel formation. Figure 1. View largeDownload slide Cartel formation. As a result, all but one competitor, ERG,51 followed ENI's new pricing strategy. About five months later in March 2005, the Italian Truckers' Association, FITA, complained to the Italian antitrust authority about high and aligned prices.52 This eventually led to an investigation by the antitrust authority for price fixing under Article 14 of Law 287/90 of October 10, 1990, the Italian legislation that restates Article 101 of the TFEU. The Italian antitrust authority claimed that the petrol firms' conduct of adapting their prices to the leader's price must be considered a collusion to stabilize prices and to coordinate price changes.53 The high transparency in the market facilitated an exchange of price information. Firms may easily observe their competitors' prices at each gas station and Italian law required weekly price communications to the Ministry of Industry that subsequently published the data. More importantly, companies communicated future price changes to a specialist Italian magazine, “Staffetta Quotidiana,” that published all price change announcements on its website. Cost transparency also facilitated coordination. The major source of cost is the Platts Cif Med,54 the wholesale price that refineries charge in the Mediterranean area for gasoline. This price can be thought of as the opportunity cost of companies to sell their gasoline on the Mediterranean wholesale market rather than to gas stations. It thus constitutes industry practice55 to compute firms' margins as the difference between their suggested consumer price and the Platts Cif Med. The Italian antitrust authority had no proof of direct communication between the firms, other than the price changes the firms communicated via the aforementioned online magazines. The authority claimed that ENI's policy created a focal price used to facilitate coordination. ENI's sticky pricing lowered competitors' uncertainty about the future pricing, while the large price variations helped to coordinate price changes.56 However, the Italian antitrust authority ended its investigation without issuing a formal decision—that is, it refrained from asserting infringements of Article 101(1) of the TFEU and Article 14 of the Law 287/90, respectively, and from imposing a fine on ENI or its competitors. Instead, the authority accepted a commitment by the firms to restrict pricing transparency on the market and, in particular, to abstain from the publication of future prices via the media.57 While there are good arguments supporting the opinion that regular and reciprocal announcements of future pricing, such as those that occurred on the Italian petrol market, should suffice to demonstrate a concerted practice,58 our analysis does not focus on this aspect of posted pricing,59 but concentrates rather on ENI's and its competitors' pricings.60 B. Sticky Pricing Sticky pricing constitutes an important element in a strategy to sustain collusion. An advantage of rigid pricing is that it is straightforward to implement and that deviations can be easily detected and punished. A series of studies61 analyze the profit maximizing scheme of cartels under different settings and find a direct relation between optimal collusive schemes and rigid pricing.62 For example, Athey and Bagwell show that when firms are moderately patient, the equilibrium that maximizes ex-ante profits is relatively simple: all firms adopt a sticky pricing scheme and charge the consumers' reserve.63 In this equilibrium, colluding firms adjust their prices infrequently and thus sacrifice productive efficiency to sustain a higher price level in the market.64 However, from a theoretical perspective, increasing the time lag between price changes—that is, the length of the commitment—also comes at a cost. The longer the time between price changes, the larger the gain from undercutting by rivals, as highlighted by Zhongmin Wang.65 Thus, from ENI's perspective, the large increase in the length of the price commitment might also imply a higher cost of being price leader, especially if ENI is the first firm to raise prices. While this argument certainly plays an important role, there is a fundamental difference between the standard exogenous commitment and ENI's endogenous commitment. Given the high market transparency, at any point in time ENI might detect and immediately punish deviation. ENI's commitment increases price stability, but this might be reverted instantaneously as market conditions change. Thus, this pricing scheme entails both the pro-collusive benefits of price stability and, at the same time, ENI does not have to lose market share for a long period of time before it can respond to deviation by rivals. In fact, most empirical studies conclude that prices are more rigid when the industry is in a collusive phase.66 A key example is the study by Abrantes-Metz, Froeb, Geweke and Taylor on the frozen perch market. Using ex-post evidence of collusion the authors find that the price variance during collusion was indeed distinctly lower than the price variance in the period after the end of the cartel.67 In a meta-study, Blanckenburg, Geist and Kholodilin compare the distribution of price changes between competition and collusion for 11 cartels. They find that the price variance decreased significantly in 8 out of 11 examined cartels.68 C. Leadership Pricing Price leadership has been considered “one of the most important institutions facilitating tacitly collusive pricing behavior.”69 Theoretical evidence has been presented by Rotemberg and Saloner, who demonstrate that price leadership facilitates collusion under asymmetric information and increases price rigidity.70 The authors conclude that such a pricing scheme has many positive attributes. It is easy to implement, it doesn't require communication, and it is very easy to detect and to punish deviations. In line with these findings, Mouraviev and Rey study the role of price or quantity leadership under circumstances where firms can act either simultaneously or sequentially in an infinitely repeated setting for both Bertrand and Cournot competition.71 They highlight that leadership facilitates collusion. Firms competing on prices a la Bertrand can use price leadership to sustain perfect collusion for any value of the discount factor while leadership is less effective with quantity competition a la Cournot. Both papers convey an important implication for antitrust policy: if firms are able to tacitly collude using price or quantity leadership, the negative effects on welfare are essentially the same compared with cases of explicit collusion. The way firms collude is not decisive for the negative effect collusion has on consumers' welfare. In addition, both papers show how leadership pricing can be used to implement an anticompetitive strategy in the market, as it facilitates coordination and makes deviation more visible. D. Key Empirical Findings Based on the previous finding of the role of sticky and leadership pricing to sustain collusion and on the effects of collusive agreements on prices, we show that ENI's pricing behavior facilitated price coordination and led to a price increase. Table 1 shows the different pricing conduct firms adopted after ENI's price commitment. In Panel A we summarize the frequency and magnitude of price changes. Columns 3 and 5 show the differences in the pre- and post-mean of these variables and test whether the pricing behavior significantly changed after ENI's policy. ENI significantly increased the time lag between price changes from one every 6 days to one every 20 days. This difference is significant at the 1-percent level and shows that ENI did hold its price commitment as publicly announced on October 6, 2004. In addition, the leader increased the absolute mean price change from 1 percent to 5.4 percent. This 4.4 percent increase is statistically significant at the 1-percent level. Similar results hold true for all firms. The average time lag between price changes increased from 5 to 9 days, while absolute price changes increased from 0.9 percent to 2.4 percent, both significant at the 1-percent level. Theoretical literature discussed above suggests that large price changes might have been used to coordinate price changes on the leader's focal price. Panel B tests this hypothesis and shows whether the average number of perfectly aligned competitors (that is, up to three digits) to the leader and the average price difference of competitors to ENI significantly changed after ENI's new pricing policy. In line with the collusive hypothesis, the number of aligned competitors significantly increased and the average price difference to the leader significantly decreased after the policy. Table 1. Pre and post policy pricing   (1)  (2)  (3)  (4)  (5)  (5)  Panel A: Frequency and Magnitude of Price Changes  Time period  Pre  Post    Pre  Post      Mean  Mean  Difference  Mean  Mean  Difference    (St. Dev.)  (St. Dev.)  t-stat  (St. Dev.)  (St. Dev.)  t-stat    Abs. % Price Change   Days between price changes   All Firms  0.0090  0.0238  0.0148  5.2  8.9  3.6    (0.0066)  (0.0299)  14.7  (5.2)  (8.0)  8.6  ENI  0.0104  0.0537  0.0433  6.1  20.1  13.9    (0.0072)  (0.0393)  10  (6.0)  (16.1)  5.9  Panel B: Average Alignment to the Market Leader    Sum of aligned firms   Price difference to ENI   ENI's Competitors  1.99  2.82  0.83  0.00175  0.00135  −0.0004    (0.093)  (0.169)  4.4  (.00006)  (.00047)  −1.34    (1)  (2)  (3)  (4)  (5)  (5)  Panel A: Frequency and Magnitude of Price Changes  Time period  Pre  Post    Pre  Post      Mean  Mean  Difference  Mean  Mean  Difference    (St. Dev.)  (St. Dev.)  t-stat  (St. Dev.)  (St. Dev.)  t-stat    Abs. % Price Change   Days between price changes   All Firms  0.0090  0.0238  0.0148  5.2  8.9  3.6    (0.0066)  (0.0299)  14.7  (5.2)  (8.0)  8.6  ENI  0.0104  0.0537  0.0433  6.1  20.1  13.9    (0.0072)  (0.0393)  10  (6.0)  (16.1)  5.9  Panel B: Average Alignment to the Market Leader    Sum of aligned firms   Price difference to ENI   ENI's Competitors  1.99  2.82  0.83  0.00175  0.00135  −0.0004    (0.093)  (0.169)  4.4  (.00006)  (.00047)  −1.34  Notes: Table 1 summarizes the pre and post policy pricing behavior of the nine firms acting in the Italian wholesale gasoline market. Panel A shows the frequency and magnitude of price changes. ENI increased the mean price change from 1% to 5.4%, while the average price change of all firms increased from 0.9% to 2.4%. Similarly, ENI increased the average time lag between price changes from one every 6 days to one every 20 days. Panel B shows the sum of aligned firms to ENI (specification 1 and 2) and the average price difference to the leader (specification 4 and 5). The number of aligned competitors significantly increased after the policy, while the average price difference to the leader decreased after the policy. View Large In addition to the price coordination adopted by firms, we report the key coefficients on the causal effect of the policy on prices and margins in Table 2.72 Specification 1 shows the result of the dif-in-dif model with standard errors clustered at the country level. In this regression, weekly prices of eight EU countries73 were used as a control group.74 The estimate on the dif-in-dif effect of the policy on Italian prices is positive and highly significant. As one might question the subjective selection and the sufficient similarity of the control group, in specification 2, we first construct an “optimal” data-driven benchmark—that is, a synthetic control group—and then take the weekly difference between the Italian price and the “optimal benchmark” as the stationary, dependent variable. The synthetic control group estimation was developed by Abadie and Gardeazabal75 and Abadie, Diamond and Hainmueller,76 respectively, and is constructed using a data-driven weight of European prices that minimizes the pre-treatment differences between the Italian price and the resulting synthetic control group. Consistent with specification 1, we find a positive and significant effect of the policy on prices. Finally, specification 3 shows the within market regression of firm-level margins—that is, without benchmark—that also points to a positive and significant effect of the new policy on firms' profits.77 Table 2. Effect of the policy on prices   (1)  (2)  (3)  Dependent Variable  Price EU Country j at week t  Price Difference Italy-Synthetic Control week t  Margin firm i day t  Type of Data  Panel Data  Time Series  Panel Data  Regression Model  Dif-in-Dif  OLS  Firm Fixed Effect  Policy*Italy  10.02***        (2.219)      Policy    11.19***  22.88***      (3.879)  (1.871)  Controls  Crude oil (4 Lags), Year and Month FE  Crude oil (4 Lags), Time trend  Time trend  Observations  882  113  7,794  R-squared  0.665  0.493  0.121    (1)  (2)  (3)  Dependent Variable  Price EU Country j at week t  Price Difference Italy-Synthetic Control week t  Margin firm i day t  Type of Data  Panel Data  Time Series  Panel Data  Regression Model  Dif-in-Dif  OLS  Firm Fixed Effect  Policy*Italy  10.02***        (2.219)      Policy    11.19***  22.88***      (3.879)  (1.871)  Controls  Crude oil (4 Lags), Year and Month FE  Crude oil (4 Lags), Time trend  Time trend  Observations  882  113  7,794  R-squared  0.665  0.493  0.121  Notes: Table 2 reports the coefficients on the full specification regression models that capture the effect of the new pricing policy. Policy*Italy is the intersection between two dummies (Italian price after the policy), while Policy is a dummy being one after October 6, 2004. FE stands for fixed effects. Prices and margins are expressed in € per 1000 liters. In specification (1) standard errors are clustered at country level, while in specification (2) and (3) robust standard errors are reported. In all specifications prices/margins significantly increased after the competitors adopted the same pricing behavior as the market leader. View Large Figure 2: View largeDownload slide Italian Price, EU Price and Brent. Figure 2: View largeDownload slide Italian Price, EU Price and Brent. The results of the econometric analysis show that ENI's policy had two effects. It facilitated price coordination and increased average prices. E. Discussion and Robustness of the Empirical Results In oligopolistic markets, the way firms interact with their competitors determines their profits. Our empirical analysis shows that the ex-post effect of the leader's credible commitment to sticky pricing was an equilibrium with higher prices. ENI's success in the implementation of a collusive scheme depended on the individual incentives for its competitors to adhere. The first issue that arises, therefore, is whether it is reasonable to think that the leader could expect ex ante that its competitors would adopt its pricing and that this would cause an increase in prices. Firms' behavior is a key element of managerial choice. Spagnolo shows that typical compensation schemes for CEOs are designed to incentivize tacit collusion at the cost of “income smoothing.”78 In addition, managers are aware of or, at least, well-advised of strategic behavior that favors collusion.79 Since the seminal work by Thomas Schelling,80 it is common knowledge that commitment lies at the heart of strategic behavior.81 If competing firms could write enforceable contracts on prices, most industries would collude. However, as explicit collusion is illegal and the decision to communicate is endogenous, firms may opt for tacit collusion instead. Yet any collusive strategy must be incentive compatible, irrespective of whether it is implemented explicitly or tacitly. After its announcement on October 6, 2004, ENI kept prices fixed for 57 days, until December 3, 2004.82 This means that ENI kept sticky prices for almost 10 times the usual price-change interval of 6 days, irrespective of cost changes. Just after ENI's announcement costs increased and its competitors kept cost-based pricing. As costs fell again competitors started to align to ENI at the beginning of November, that is, about a month after ENI's change in pricing policy. We can only speculate about what would have happened if costs had risen after ENI's announcement. However, it clearly emerges, both from Figure 1 and from the price-interdependence analysis, that ENI strongly committed itself to sticky pricing. As can be inferred from Table 2, specification (3), ENI's competitors behaved in their best interest as industry margins increased. ENI emerged endogenously as the price leader through its use of market power and then used its position to coordinate the price changes of its competitors, which ultimately caused a price increase. While each market has its traits, and results from an individual market cannot be easily generalized, leadership pricing has been consistently associated with collusion. Our empirical results provide substantial evidence that ENI's strategy aimed at coordinating and increasing prices came at the expense of consumers. A second concern that arises is where to set the boundaries between a firm's freedom to set its profit-maximizing price on the one hand, and antitrust authorities' power to prevent certain behavior that results in supracompetitive pricing on the other. To address this issue, we need to distinguish the “source” of market power that made that market outcome possible. In this respect, it is helpful to compare our empirical results with those of Borenstein, Bushnell and Wolak, who analyze inefficiencies in the restructured Californian electricity market.83 They find that wholesale electricity expenditures increased in the summer of 2000 with respect to the summer of 1999 from $2.04 billion to $8.98 billion and that about 59 percent of this increase was caused by the exercise of unilateral market power. Both the Italian gasoline market and the Californian electricity market suffered from higher prices. However, there is a key difference—in California, market power stemmed from exogenous shocks. Electricity prices were relatively low compared to a benchmark in 1998 and 1999, but dramatically increased in the summer of 2000. While there are many structural factors that make it easy for electricity firms to exercise market power, such as binding constraints at peak times or difficulties to forecast demand and high storage costs, firms did not actively implement a new strategy to coordinate and increase their prices, but rather individually best-responded to shocks that favored the exercise of market power. Three of the many factors that Borenstein, Bushnell, and Wolak identify are that hydroelectric production decreased in 2000 because of dryness; economic growth in the Western United States increased demand for energy; and the cost of nitrogen pollution permits increased from about $1 per pound to over $30 per pound, resulting in an increase in the price of gas. In the Californian electricity market, regulation should address the structural problems that have been revealed by the incidents of the summer of 2000. However, insofar as the firms only best-responded to exogenous shocks, their conduct should not be addressed by cartel enforcement. In contrast, our analysis reveals that the active implementation of a collusive strategy by one firm resulted in an anticompetitive market outcome and should therefore be targeted by antitrust enforcement. Finally, one may wonder whether the pricing policy implemented by ENI in fact had pro-competitive effects. Being the market leader, ENI possibly had better information about demand or supply trends. Thus, competitors might have followed ENI's pricing because it was based on such information, and as a result firms' prices were more aligned. Although, in theory, each strategy might entail some pro-competitive elements, our approach makes use of the ex-post observed effect of a specific conduct on the market outcome. Sticky-leadership pricing causally led to a higher price level compared to a European benchmark and thus had a welfare decreasing effect on consumers. None of ENI's competitors adopted a myopic best response and undercut the other firms when costs were decreasing, but rather waited for the leader to move first. This is consistent with a dynamic best-response behavior, in which firms may anticipate the pro-collusive effect of keeping their prices aligned, and reinforces the conclusion that the observable pricing pattern must be characterized as collusive. Moreover, as we already discussed in Parts III.B and III.C, there is sound theoretical and empirical evidence that firms use leadership pricing and sticky pricing to facilitate coordination and raise prices to a supracompetitive level. ENI's strategic behavior addresses two main problems of a cartel–coordination and stability. In a setting in which prices are transparent, ENI's stable price served as a focal price and facilitated collusion. In addition, the initial commitment not to increase prices when costs were increasing shows that ENI was trading off short term losses against future gains. In doing so, ENI signaled that it was prepared to pay the costs in terms of lower margins to maintain its pricing strategy and to punish competitors should they deviate from the pricing pattern as set by ENI. IV. INTEGRATING ECONOMIC INSIGHTS ON COLLUSIVE STRATEGIES INTO THE LEGAL FRAMEWORK As any collusion between competitors may result in welfare losses, it is essential to strive to contain collusive behavior irrespective of direct evidence of a “meeting of minds” or explicit communication between firms. There remains, however, an outstanding question of how economics may be integrated with the legal framework and how antitrust law should be developed to counter tacit collusion. There are several reasons to believe that this challenge deserves more attention than ever. First of all, prevalence of tacit collusion may increase in times of globalization. Information on competitors' actions as capacity choices, prices, and transactions are widely reported by international media and transparency increases. Firms interact in many markets, increasing their scope to collude. Second, market players must not be regarded as naïve, but as professionally advised and capable of employing economic know-how strategically to avoid price wars and to reach collusive equilibria, instead. Third, the introduction of leniency or other types of immunity programs increased the capability of antitrust authorities to produce direct evidence of collusion, such as documents or insider testimony, and has thus significantly strengthened the effectiveness of the law to counter collusive behavior that occurs via explicit communication.84 As the decision to communicate is endogenous to market players, leniency programs have increased firms' cost of following such a strategy. This is likely to cause or to have already caused a shift from explicit to tacit collusion. These are grounds to expect that social welfare damage caused by tacit collusion will increase. Certainly, merger control may work preventively against coordinated effects. However, it will hardly suffice to counter increasing anticompetitive effects caused by tacit collusion. This is mainly the result of inherent limits of scope. Market concentration may increase as a result of a firm's internal growth and the exit from the market of firms that operate less efficiently. Thus, tacit collusion remains unaddressed by merger control in cases of stable oligopolies and especially in duopolistic markets—that is, exactly in those industries where the market structure entails high risks of tacit collusion.85 It appears to be crucial, therefore, for antitrust law to find a way to target those elements of behavior that are employed by firms to implement a collusive strategy and whose collusive character may be demonstrated by the kind of analysis as suggested in this article. Inasmuch as it appears inadequate to regard such behavior as illegal coordination, this calls for a development of the law against unilateral anticompetitive conduct. A. “Unilateral Collusion” and Unlawful Coordination Price leadership may serve as a mechanism to find a consensus about the collusive price, a challenge any cartel faces. From this perspective, one might be tempted to conclude that price leadership should be prohibited. In this case, if ENI was not allowed to be a price leader, ENI would not be able to coordinate price changes, and the anticompetitive effects—that is, the significant price increase that resulted from such behavior—would not have taken place. However, price leadership and any other “leader-follower” behavior may equally be the result of effective oligopolistic competition. For example, the market leader often has an informational advantage over smaller competitors in respect of relevant changes in market condition. A price change by the leader might entail useful information for market participants and their individual best response might be to align to the market leader. If firms with superior information are not allowed to adjust prices and competitors are not allowed to respond best to price changes by competitors, or both, then the price setting freedom of firms will be significantly limited, a situation that might ultimately lead to regulated prices. This raises the question of whether under circumstances such as those in the present case—that is, where it may be demonstrated that “leader-follower” behavior sustained a collusive equilibrium—such conduct should be considered illegal. In other words, should the kind of evidence presented herein be regarded a “super plus factor”86 that allows courts to infer an illegal tacit agreement? If certain conduct of two or more firms is conceptualized as an unlawful coordination—that is, a violation—of, for example, section 1 of the Sherman Act or Article 101(1) of the TFEU, this implies that the law regards the behavior of these firms as a wrongdoing that may be punished. In other words, where a certain collusive equilibrium has been brought about by the unilateral collusive conduct of one firm, one should only infer a punishable agreement if one also considers the competitors' reactions as inappropriate behavior. Turning again to the general regulatory and legal requirements we formulated with regard to antitrust enforcement,87 we recall that antitrust standards and remedies should address specific elements of behavior, and that market operators should be provided with an idea of which conduct may be regarded as acceptable or not acceptable under defined market conditions. Such standards of conduct must be in line with the general purpose of antitrust law to foster effective competition. Thus, if tacit price alignment in response to unilateral collusive conduct ought to be prohibited, the law has to define how firms should behave once a competitor's conduct may be interpreted as a tacit invitation to engage in tacit collusion. When ENI held prices constant despite cost increases and thereby signaled its commitment to a policy of sticky pricing, its actions might be viewed as a “suggestion” to its competitors to align their pricing policy and as an “offer” to take on the role as price leader. Should ENI's competitors have been legally obligated to refrain from any market conduct that ultimately could have been regarded as having brought about a collusive equilibrium and thus proof of an underlying illegal agreement? We believe the answer is no. First of all, it does not seem feasible to define any meaningful and administrable legal standard of conduct in this respect. Should it have been forbidden for ENI's competitors to tacitly align their prices to ENI's policy of sticky pricing? Should they have been obligated to stick to their higher prices and with open eyes to put up with losing market share? And even if an alignment of pricing to the strategy of a price leader such as ENI was prohibited, the question would remain: How closely and how quickly a competitor would be allowed to adjust its market parameters? In the absence of any clear standard of behavior, a legal intervention in situations of supposedly collusive pricing may ultimately amount to judicial price regulation. In addition, under such a legal regime a market player could strategically restrict the competitive room for maneuver of its competitors. If it was prohibited for ENI's competitors to align their pricing to ENI's strategy because such an alignment would be conceptualized as an acceptance of an offer to engage in illegal coordination, ENI could have restricted the price-setting freedom of its competitors by implementing its strategy of sticky pricing. These considerations point to the heart of the regulatory problem with regard to “unilateral collusion.” The reaction of ENI's competitors to ENI's pricing policy must be regarded as mere best response. Their behavior is an expression of mere oligopolistic interdependence, even though they benefitted from the higher price level in the market. Consequently, a passive adaptation to collusive market conduct should not be considered illegal, but part of functioning oligopolistic competition. Thus, collusive “leader-follower” behavior must not be conceptualized as a form of unlawful coordination, and thus illegal according to section 1 of the Sherman Act or Article 101(1) of the TFEU. Antitrust law should instead target unilateral collusive behavior that facilitates “best response” that ultimately leads to collusion. This appraisal of collusive “leader-follower” behavior appears to be in line with the treatment of non-conspiring firms that, for example, adjust their prices in reaction to a price increase by cartelizing competitors. Such a constellation is generally referred to as “umbrella pricing,” as the nonparticipant benefits from the “price umbrella” are spread by its cartelizing rivals.88 This metaphor somewhat obscures the interdependence between the optimal cartel price and the behavior of the firms outside the circle of cartel participants. Firms that are aware of illegal price coordination by their competitors and engage in “umbrella pricing” consciously contribute towards reaching a certain collusive equilibrium. Furthermore, these firms will also consider the risk of retaliatory measures by the cartelists in the case that they ignore the cartel policy. Thus, their reaction should be characterized as a dynamic best response similar to the alignment of pricing (for example, alignment of pricing by ENI's competitors). Nevertheless, even if a non-cartelist acts consciously in accordance with the collusive strategy of a cartel, “umbrella pricing” is generally regarded as being innocent, and the legal discussion centers only around the question of whether customers of nonparticipants may recover damages from the cartelists.89 Thus, notwithstanding that “umbrella pricing” contributes toward sustaining collusive equilibria, antitrust law does not require market operators to abstain from a best-response strategy in reaction to their competitors' pricing. The law refrains from imposing on non-cartelists a duty to keep prices constant (or at least at a lower level than the cartel price), which would effectively amount to a duty to increase output to offset the cutback of conspiring competitors. If we accept therefore that there are valid economic and regulatory reasons why collusive “leader-follower” behavior, such as the pricing alignment by ENI's competitors, should not be considered as participation in an illegal coordination, it seems consequent that the law should instead target ENI's decision to implement a collusive strategy. B. Developing the Legal Framework: Targeting Unilateral Conduct with Collusive Impetus Unilateral conduct that has as its object or effect to promote tacit collusion ought to be prevented. Based on findings of the collusive potential of sticky pricing, we have proved empirically that ENI employed such a pricing policy successfully to bring about a collusive equilibrium in the Italian gasoline market. But is there a feasible way of legal intervention? Should we ban a firm from implementing a policy of sticky pricing because it may facilitate collusion, and punish the firm in case of an infringement? A firm's unilateral behavior may only be regarded as having as its object or effect the promotion of tacit collusion if that is the only reasonable inference—that is, in the absence of any competitive rationale that could convincingly explain the behavior in question, in particular in view of the economic conditions of the market and the effect of the behavior on prices or other market parameters. We submit that under such a doctrine, ENI's pricing strategy should be considered a breach of antitrust laws. The Italian gasoline market90 was characterized by features that indicate its conduciveness to tacit collusion, such as its concentrated oligopolistic market structure, a high price transparency, and entry barriers. More specifically, due to its market share of about 35 percent and the asymmetric distribution of market shares in the Italian petrol industry, ENI clearly held the position as market leader. Thus, its commitment to a strategy of sticky pricing91 resulted in a credible signal to its competitors and entailed a strong potential to encourage them to align their pricing in order to bring about a collusive equilibrium. Given the circumstances of the case and in view of our empirical findings, there is no convincing explanation to defend ENI's pricing strategy as being competitive. It is true that due to better information resources, for example, a market leader may be in a superior position to set the market price and its price leadership may have pro-competitive effects as a result. However, in the case at hand, the only reasonable explanation for ENI's implementation of sticky pricing along with large price changes is that ENI intended to take the position as price leader in order to coordinate prices.92 This is in particular reinforced by our finding that ENI's pricing strategy as a matter of fact resulted in higher and aligned prices. Turning to antitrust provisions that address firms' unilateral behavior, we need to recognize, however, that the law appears to be fragmented—to say the least—when it comes to conduct whose object or effect it is to promote collusion. Neither section 1 of the Sherman Act nor Article 101 of the TFEU embodies an offense of attempted coordination. Section 2 of the Sherman Act and Article 102 of the TFEU, the essential provisions on unilateral conduct, apply first of all to firms with monopoly power or to firms that dominate a market, respectively, and thus based on criteria that typically exclude single oligopolists. It seems, for instance, rather uncertain whether ENI's position may be characterized as “dominant” pursuant to Article 102 of the TFEU. The ECJ presumes predominance where a firm has a market share of 50 percent or more.93 Below this threshold, additional factors must be put forward to show market dominance. Given certain market characteristics, such as the asymmetrically distributed market shares and high barriers to market entry, ENI's position might be considered as being just above the lower bound that defines the area of market dominance.94 Yet, in cases where single firm dominance may be established, conduct such as ENI's, which aims at promoting or sustaining tacit collusion, should be considered abusive. Beyond single market dominance, firms' conduct that aims at promoting or sustaining tacit collusion could be considered as an abuse of collective dominance under Article 102 of the TFEU.95 While the concept of collective dominance, as developed under the EU Merger Regulation, may be transposed to Article 102 of the TFEU, the latter requires an existing dominant position. Consequently, practices that firms strategically employ to reach a collusive equilibrium in the first place may not be covered, and thus the potential of an approach that precisely identifies the shift toward a tacitly collusive equilibrium could not be fully exploited. Practices that facilitate tacit collusion may at most be tackled under Article 102 of the TFEU insofar as they safeguard an already-established collectively dominant position.96 It appears, however, that the European Commission has never initiated any TFEU Article 102 proceedings on grounds of collective dominance. Therefore, as the application of Article 102 of the TFEU to tacit collusion hitherto is only a theoretical option, it remains all the more unclear what conduct could be prohibited under such a doctrine. Under section 2 of the Sherman Act, the prohibition of any “attempt to monopolize” broadens the scope and may allow the inclusion of the unilateral conduct of firms that individually do not hold a monopoly position. Thus, in United States v. American Airlines, Inc.,97 an explicit invitation to collude was considered an infringement of section 2 of the Sherman Act, as the court considered the aggregate market share of offerer and offeree. However, because there is no established case law on unilateral attempts to bring about tacit collusion, and thus a shared monopoly, it remains doubtful whether section 2 of the Sherman Act may apply to such practices. In line with the approach suggested in this article, the Federal Trade Commission (FTC) strove to tackle unilaterally adopted, supposedly facilitating practices under section 5 of the FTC Act. While the U.S. Supreme Court has recognized that this provision may in principle comprise anticompetitive conduct beyond the Sherman Act,98 the FTC's ambition received a decisive99 blow from the decision of the Court of Appeals for the Second Circuit in the Du Pont (Ethyl) case.100 In Ethyl, the FTC blamed four producers of gasoline antiknock compounds of having unilaterally adopted practices that were aimed at facilitating parallel pricing at a supracompetitive level. These practices included 30-day advance announcements of price changes, “most favored nations” clauses in sales contracts, and uniform delivered prices.101 The Court, however, held that the evidence presented by the FTC did not sufficiently support the view that these practices did indeed have an anticompetitive purpose or effect.102 Econometric evidence as suggested in this article could fill such gaps by relating a specific practice with a certain market outcome. With adequate firm level data and a benchmark, an antitrust authority or a court may test whether or not supposedly facilitative practices contributed to a supracompetitive price level. This shows on the one hand that economic methods may support an effective use of available legal instruments to counter unilateral behavior that has as its object or effect to promote collusion. On the other hand, the analysis reveals a significant gap in the arsenal of antitrust enforcement when it comes to targeting unilateral conduct that serves a collusive strategy. Thus, under the current legal framework, the potential of economics to identify the collusive character of specific elements of behavior may not be fully realized. It seems therefore essential to strengthen legal instruments that frustrate unilateral conduct through which firms strive to promote or sustain collusion. V. CONCLUSION Collusion in oligopolistic markets has been a perennial topic both for economics and antitrust law. Antitrust law rests on economic welfare analysis that shows that collusion inflicts substantial negative welfare effects. However, antitrust authorities and private plaintiffs are substantially restricted in their fight against collusion, as they strongly depend on evidence of explicit communication between competitors. The mild reaction of the Italian antitrust authority to the incidents on the Italian gasoline market illustrates the limits of antitrust enforcement in the absence of such evidence. The crucial role attributed to explicit communication in the practice of antitrust enforcement hinders the detection and punishment of cartels precisely in those industries where the collusion rate is expected to be relatively high and communication appears to be less needed. Theoretical and empirical findings on cartel behavior provide a basis to derive clear test hypotheses to distinguish lawful oligopolistic interdependence from tacit collusion. On that basis, econometric evidence may step in and reveal collusive strategies behind firms' actions. Thus, it entails the potential to decisively increase the effectiveness of cartel enforcement in oligopolistic markets. Analyzing the incidents on the Italian gasoline market where the market leader announced it was changing its pricing strategy reveals how firms might use their market power to facilitate price alignment and coordinate price changes. The econometric analysis reveals how the leader's sticky pricing policy coordinated prices, and what effect this had on the price levels with respect to a benchmark: After the new policy was implemented, all competitors adjusted their prices following the leader's price changes. In addition, the new pricing behavior resulted in a significant price increase. Combined, this price coordination mechanism and its effect show that it was the object and effect of the introduced pricing policy to collude through facilitating price coordination and to raise prices. While antitrust enforcement may certainly benefit from an enhanced economic methodology to identify tacit collusion, antitrust law cannot straightforwardly prohibit the participation in tacit collusion as a form of illegal coordination. The active promotion of collusive pricing by ENI and the passive best-response alignment of its competitors must not be normatively equated. Thus, antitrust law should not infer a punishable tacit agreement between ENI and its competitors from the collusive market outcome, but should instead consider conduct, such as ENI's pricing strategy, a unilateral anticompetitive practice. To effectively fight tacit collusion it appears therefore to be necessary to strengthen legal instruments that target unilateral conduct that firms strategically employ to promote or sustain collusion. 1 Throughout this article, we use the term “cartel” to describe any kind of welfare-decreasing form of collusion, be it an explicit or a tacit one, and irrespective of whether or not we consider it an infringement of antitrust law. 2 See Drew Fudenberg & David Tirole, Game Theory 150 (MIT Press 1991). Much of the theoretical discussion on collusion is based on the supergame approach. The best known result describing firms' incentives to collude is the “Folk Theorem,” which states that, for sufficiently low discount rates, almost any price may be sustained as the equilibrium outcome of a repeated game. While the “Folk Theorem” provides fairly general conditions under which tacit collusion may be sustained as an equilibrium, it says nothing about how firms behave in reality. The strategies used in the “Folk Theorem” are chosen because of their analytical ease and not because they describe firms' collusive behavior. 3 See Margaret C. Levenstein & Valerie Y. Suslow, What Determines Cartel Success, 44 J. Econ. Literature 43 (2006). While economic theory shows that concentration facilitates collusion, and thus predicts a positive relation between cartelization rate and market concentration, empirical evidence seems to contradict this result. This gap between the number of cartels predicted from a theoretical perspective and the number of cartels that appear in the empirical analysis may plausibly be explained by a sample-selection bias. Only cartels that, first, have been detected and that, second, were regarded as illegal by antitrust authorities or courts are contained in the sample. 4 Miguel A. Fonseca & Hans-Theo Normann, Explicit vs. Tacit Collusion—The Impact of Communication in Oligopoly Experiments, 56 Eur. Econ. Rev. 1759 (2012). Fonseca and Normann use a laboratory experiment to investigate the role of communication in sustaining collusion. They show that highly concentrated industries collude irrespective of communication. 5 See Gregory J. Werden, Economic Evidence on the Existence of Collusion: Reconciling Antitrust Law with Oligopoly Theory, 71 Antitrust L.J. 719 (2004); Louis Kaplow, An Economic Approach to Price Fixing, 77 Antitrust L.J. 343 (2011) (comprehensively analyzing the use of economics to support cartel enforcement). 6 See William G. Christie & Paul H. Schultz, Why Do Nasdaq Market Makers Avoid Odd-Eighth Quotes?, 49 J. Fin. 1813 (1994). One of the best known examples of economic detection of collusion is provided by the work of Christie and Schultz. They detected collusion between Nasdaq market makers by comparing their bid-ask spread to the equivalent spread on the New York Stock Exchange. Christie and Schultz's work had an impressive impact, as it led to regulatory investigations by the Securities and Exchange Commission (SEC) and class action lawsuits that were settled for over $1 billion. 7 See David Genovese & Wallace P. Mullin, Rules, Communication, and Collusion: Narrative Evidence from the Sugar Institute Case, 91 Am. Econ. Rev. 379 (2001). Genovese and Mullin provide narrative evidence of the role of communication for collusion in the Sugar Institute Case. They find that one key missing aspect in formal theories of collusion is the role for rich communication within the collusive agreement. 8 To use the words of Thomas Hobbes, “the bonds of words are too weak to bridle men's ambition, avarice, anger, and other Passions, without the fear of some coercive Power … ” Thomas Hobbes, The Leviathan 71 (J.M. Dent & Sons 1959) (1651). 9 See, e.g., Joseph E. Harrington, Jr. & Andrzey Skrzypacz, Private Monitoring and Communication in Cartels: Explaining Recent Collusive Practices, 101 Am. Econ. Rev. 2425 (2011) (characterizing an equilibrium in which colluding firms truthfully self-report their sales and then make transfers within the cartel). 10 Louis Phlips, On the Detection of Collusion and Predation, 40 Eur. Econ. Rev. 495 (1996). 11 In this respect, our approach is conceptually in line with Hay, who argues that “if there is to be a category of unlawful tacit collusion which is to be distinguished from classic oligopoly, the difference must lie … on the specific elements of behavior that brought about that state of mind.” See George A. Hay, The Meaning of “Agreement” Under the Sherman Act: Thoughts from the “Facilitating Practices” Experience, 16 Rev. Indus. Org. 113, 128 (2000). 12 Patrick Andreoli-Versbach & Jens-Uwe Franck, Endogenous Price Commitment, Sticky and Leadership Pricing: Evidence from the Italian Petrol Market, 40 Int'l J. Indus. Org. 32 (2015). 13 See infra Figure 1 (plotting daily prices and the major source of cost, that is, the Platts Cif Med, around ENI's new price policy announcement (first vertical line)). Firms respond to cost shocks with some lags. While current costs decreased immediately after ENI's policy, lagged costs increased, and thus competitors increased their prices. 14 Yulia V. Bolotova, Cartel Overcharges: An Empirical Analysis, 70 J. Econ. Behav. Org. 321 (2009); John M. Connor, Collusion and Price Dispersion, 12 Applied Econ. Letters 335 (2005). 15 This conceptual divergence may also give rise to terminological misunderstandings between economists and lawyers. Throughout this article, we will indicate when we use terms such as “collusion” or “agreement” in their technical economic or legal meaning. 16 American Tobacco Co. v. United States, 328 U.S. 781, 809 (1946) (“No formal agreement is necessary to constitute an unlawful conspiracy”); Norfolk Monument Co. v. Woodlawn Memorial Gardens, Inc., 394 U.S. 700, 704 (1969) (“business behavior is admissible circumstantial evidence from which the fact finder may infer agreement”). 17 Case T-41/96, Bayer AG v. Comm'n, 2000 E.C.R. II-3383 ¶ 69, aff'd Case C-2/01 P, Bundesverband der Arzneimittel-Importeure eV & Comm'n v. Bayer AG, 2004 E.C.R. I-23 ¶ 97. 18 Monsanto Co. v. Spray-Rite Servs. Corp., 465 U.S. 752, 768 (1984); see also Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588 (1986). 19 See, e.g., Theatre Enters., Inc. v. Paramount Film Distrib. Corp., 346 U.S. 537, 541 (1954). 20 See Andrew I. Gavil, William E. Kovacic & Jonathan B. Baker, Antitrust Law In Perspective 310 (Thomson West 2d ed. 2008). 21 Matsushita, 475 U.S. at 588 (citing Monsanto Co. v. Spray-Rite Servs. Corp., 465 U.S. 752, 764 (1984)). In this regard, the Supreme Court's decision in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007) has arguably neither amended nor clarified the pleading standard. See William H. Page, Twombly and Communication: The Emerging Definition of Concerted Action Under the New Pleading Standards, 5 J. Competition L. & Econ. 439, 447 (2009) (“To meet the evidentiary sufficiency standard, the plaintiff must produce evidence that tends to exclude the possibility of conscious parallelism.”). 22 Gavil, Kovacic & Baker, supra note 20, at 311. See also Louis Kaplow, On the Meaning of Horizontal Agreements in Competition Law, 99 Cal. L. Rev. 683, 816 (2011) (“[T]his Article does not come close to demonstrating that it would be good policy to proscribe and highly penalize all instances in which interdependent oligopolistic behavior appears to occur. The design of optimal policy is not dictated by definitions but rather by direct assessment of the consequences of different regulatory approaches.”). 23 Roger D. Blair & David L. Kasermann, Antitrust Economics 251 (Oxford Univ. Press 2d ed. 2009) (“The law fails to deal with tacit collusion very effectively. In examining the case law on conscious parallelism, basing-point pricing, and price leadership, we found that no case has held purely tacit behavior to be illegal. The basic problem seems to be that the law on collusion has developed around the fact of agreement rather than the economic effects of collusion.”). 24 See Joined Cases C-189/02 P, C-202/02 P, C-205–208/02 P & C-213/02, P Dansk Rørindustrie A/S & Others v. Comm'n, 2005 E.C.R. I-5425 ¶ 143. Accordingly, the Court infers a tacit approval of a collusive initiative from the mere attendance of a meeting where an anticompetitive agreement was concluded. “That complicity constitutes a passive mode of participation in the infringement which is therefore capable of rendering the undertaking liable in the context of a single agreement … .” Id. 25 Case C-413/06, P Bertelsmann & Sony Corp. of America v. Impala, 2008 E.C.R. I-4951 ¶ 123. 26 Joined Cases 40–48, 50, 54–56, 111, 113 & 114/73, Suiker Unie & Others v. Comm'n, 1975 E.C.R. 1663 ¶ 173; Case 172/80, Züchner v. Bayerische Vereinsbank, 1981 E.C.R. 2021 ¶ 12–14; Case C-8/08, T-Mobile Netherlands & Others v. Comm'n, 2009 E.C.R. I-4529 ¶ 33–35. 27 American Tobacco Co. v. United States, 328 U.S. 781, 810 (1946); Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752, 771 (1984). 28 Case 41/69, ACF Chemiefarma NV v. Comm'n, 1970 E.C.R. 661 ¶ 112. 29 Case T-99/04, AC Treuhand v. Comm'n, 2008 E.C.R. II-1501 ¶ 118. 30 6 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law § 1415c, at 107 (Wolters Kluwer 3d ed. 2010) (citing Blomkest Fertilizer v. Potash Corp., 203 F.3d 1028, 1037 (8th Cir. 2000)). This is presumed if, for example, a certain conduct “is so perilous when not imitated and imitation so uncertain that no reasonable actor would so act, then parallel action does imply some exchange of commitments or at least some comforting assurances connoting a traditional conspiracy.” Id. 31 See Richard A. Posner, Antitrust Law 99 (Univ. of Chi. Press 2d ed. 2001) (“[A] damages judgment in a tacit collusion case would promote competition at a tolerable cost in legal uncertainty and judicial supervision.”). 32 Cf. E.I. Du Pont De Nemours & Co. v. Fed. Trade Comm'n, 729 F.2d 128, 139 (2d Cir. 1984) (“In view of this patent uncertainty the [Federal Trade] Commission owes a duty to define the conditions under which conduct claimed to facilitate price uniformity would be unfair so that businesses will have an inkling as to what they can lawfully do. … The Commission's decision in the present case does not provide any guidelines; it would require each producer not only to assess the general conduct of the antiknock business but also that of each of its competitors and the reaction of each to the other, which would be virtually impossible.”). 33 See Joined Cases C-189/02 P, C-202/02 P, C-205–208/02 & C-213/02, Dansk Rørindustrie A/S & Others v. Comm'n, 2005 E.C.R. I-5425 ¶ 202. Under European law, Article 7(1) ECHR enshrines the principle that offenses and punishments are to be strictly defined by law. 34 Cf. United States v. United States Gypsum Co., 438 U.S. 422, 435 (1978) (“We agree with the Court of Appeals that an effect on prices, without more, will not support a criminal conviction under the Sherman Act. … [A] defendant's state of mind or intent is an element of a criminal antitrust offense which must be established by evidence and inferences drawn therefrom, and cannot be taken from the trier of fact through reliance on a legal presumption of wrongful intent from proof of an effect on prices.”). As to the required standard of intent the Court concluded “that action undertaken with knowledge of its probable consequences and having the requisite anticompetitive effects can be a sufficient predicate for a finding of criminal liability under the antitrust laws” (Id. at 444). 35 See Council Regulation (EC) No. 1/2003 of 16 December 2002 on the Implementation of the Rules on Competition Laid Down in Articles 81 and 82 of the Treaty, 2003 O.J. (L 1) 1, http://eur-lex.europa.eu/legal-content/EN/ALL/?uri=celex:32003R0001. 36 See Joseph E. Harrington, Jr., How do Cartels Operate, 2 Found. & Trends Microeconomics 1 (2006) (analyzing case studies on cartels); see also Levenstein & Suslow, supra note 3, (conducting a meta-study on cartels' features); George A. Hay and Daniel Kelley, An Empirical Survey of Price-Fixing Conspiracies, 17 J.L. & Econ. 13 (1974); Margaret C. Levenstein & Valerie Y. Suslow, Breaking Up Is Hard to Do: Determinants of Cartel Duration, 54 J.L. & Econ. 455 (2011). 37 See Eric Zitzewitz, Forensic Economics, 50 J. Econ. Literature 731 (2012); Jay R. Ritter, Forensic Finance, 22 J. Econ. Persp., no. 3, Summer 2008, at 127. 38 Brian A. Jacob & Steven D. Levitt, Rotten Apples: An Investigation of the Prevalence and Predictors of Teacher Cheating, 118 Q.J. Econ. 843 (2003). 39 Stefano Dellavigna & Eliana La Ferrara, Detecting Illegal Arms Trade, 2 Am. Econ. J. 26 (2010). 40 Marianne Bertrand, Dolly Chugh & Sendhill Mullainathan, Implicit Discrimination, 95 Am. Econ. Rev. 94 (2005). 41 Rosa Abrantes-Metz & Patrick Bajari, Screens for Conspiracies and Their Multiple Applications, 24 Antitrust 66 (2009). 42 See, e.g., Patrick Bajari & Lixin Ye, Deciding Between Competition and Collision, 85 Rev. Econ. & Stat. 971 (2003) (developing an approach to identify and test for bid rigging in procurement auctions); Robert H. Porter, Detecting Collusion, 26 Rev. Indus. Org. 147 (2005); Joseph E. Harrington, Jr., Detecting Cartels, in Handbook in Antitrust Economics (Paolo Buccirossi ed., MIT Press 2008); Patrick Rey, On the Use of Economic Analysis in Cartel Detection, in European Competition Law Annual 2006, Enforcement of Prohibition of Cartels (Calus-Dieter Ehlermann & Isabela Atanasiu eds., Hart Publishing 2007) (generally discussing methods to detect collusion); Hans W. Friederiszick & Frank P. Maier-Rigaud, Triggering Inspections Ex Officio: Moving Beyond a Passive EU Cartel Policy, 4 J. Competition L. & Econ. 89 (2008) (arguing for a greater role of ex officio investigations based, for example, on economic screens). 43 George J. Stigler, A Theory of Oligopoly, 72 J. Pol. Econ. 44 (1964). 44 See, e.g., Edward J. Green & Robert H. Porter, Noncooperative Collusion Under Imperfect Price Information, 52 Econometrica 87 (1984); Julio J. Rotemberg & Garth Saloner, A Supergame - Theoretic Model of Price Wars During Booms, 76 Am. Econ. Rev. 390 (1986); Eric Maskin & Jean Tirole, A Theory of Dynamic Oligopoly, II: Price Competition, Kinked Demand Curves, and Edgeworth Cycles, 56 Econometrica 571 (1988). 45 Julio J. Rotemberg & Garth Saloner, Collusive Price Leadership, 39 J. Indus. Econ. 93 (1990). 46 Igor Mouraviev & Patrick Rey, Collusion and Leadership, 29 Int'l J. Indus. Org. 705 (2011). 47 See Susan Athey & Kyle Bagwell, Optimal Collusion with Private Information, 32 RAND J. Econ. 428 (2001); see also Susan Athey & Kyle Bagwell, Collusion with Persistent Cost Shocks, 76 Econometrica 493 (2008); Susan Athey, Kyle Bagwell & Chris Sanchirico, Collusion and Price Rigidity, 71 Rev. Econ. Stud. 317 (2004); Makoto Hanazono & Huanxing Yang, Collusion, Fluctuating Demand, and Price Rigidity, 48 Int'l Econ. Rev. 483 (2007); Luke Garrod, Collusive Price Rigidity Under Price-Matching Punishments, 30 Int'l J. Indus. Org. 471 (2012). 48 See Rosa M. Abrantes-Metz, Luke M. Froeb, John Geweke & Christopher T. Taylor, A Variance Screen for Collusion, 24 Int'l J. Indus. Org. 467 (2006); see also Korbinian Blanckenburg, Alexander Geist & Konstantin A. Kholodilin, The Influence of Collusion on Price Changes: New Evidence from Major Cartel Cases, 13 German Econ. Rev. 245 (2012); Connor, supra note 14. 49 Press Release, AGCM (Italian Competition Authority), Petrol Prices: Antitrust Authority Launches Investigation into Nine Oil Companies over Possible Anti-competitive Pricing Agreement (Jan. 18, 2007), available athttp://www.agcm.it [hereinafter Petrol Pricing Competition Investigation]. 50 In our setting, cost changes are persistent and, thus, the strategic decision of firms concerns when to change prices and adapt to the new cost level emerging from past cost changes. 51 See Petrol Pricing Competition Investigation, supra note 49. ERG publicly declared that it would not follow ENI's new pricing and stick to its own method, which it did not further specify. 52 Id. 53 Id. 54 The Platts Company is a leading global provider of energy information that collects and publishes details on the prices of bids on a daily basis and offers for specialized oil products and regions from traders and exchange platforms. 55 See Italian Petrol Union, http://www.unionepetrolifera.it/it/show/34/La%20struttura%20del%20prezzo (defining gross margin as the difference between the retail price net of taxes and the Platts Cif Med). 56 Petrol Pricing Competition Investigation, supra note 49. 57 Id. 58 To conceptualize the reciprocal pricing announcements as an act of “concertation” would in turn trigger a presumption according to which such an indirect information exchange did indeed affect the subsequent pricing behavior of the participating firms, see Case C-199/92, P Hüls v. Comm'n, 1999 E.C.R. I-4287 ¶ 162; Case C-8/08, T-Mobile Netherlands & Others v. Comm'n, 2009 E.C.R. I-4529 ¶ 52. Cf. Joined Cases C-89/85, C-104/85, C-116/85, C-117/85 & C-125–129/85, Ahlström Osakeyhtiö & Others v. Comm'n, 1993 E.C.R. 1307 ¶ 75–79 & 126 (The case could arguably be distinguished from the ECJ's judgment in “Wood Pulp” as it seems rather doubtful that ENI and its competitors may invoke legitimate business interests for making their future pricing public); European Commission, Guidelines on the Application of Article 101 of the Treaty on the Functioning of the European Union to Horizontal Co-operation Agreements, 2011 O.J. (C 11) 1, http://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:52011XC0114%2804%29 (“However, depending on the facts underlying the case at hand, the possibility of finding a concerted practice cannot be excluded, for example in a situation where … [a public] announcement was followed by public announcements by other competitors, not least because strategic responses of competitors to each other's public announcements (which, to take one instance, might involve readjustments of their own earlier announcements to announcements made by competitors) could prove to be a strategy for reaching a common understanding about the terms of coordination.”). On November 22, 2013, the European Commission announced that it had opened proceedings against container liner shipping companies on the grounds that “the companies have been making regular public announcements of price increase intentions through press releases on their websites and in the specialised trade press.” See Press Release, European Commission, Antitrust: Commission Opens Proceedings Against Container Liner Shipping Companies (Nov. 22, 2013), http://europa.eu/rapid/press-release_IP-13-1144_en.htm. 59 See Joseph E. Harrington, Jr., Posted Pricing as a Plus Factor, 7 J. Competition L. & Econ. 1 (2011) (identifying market conditions under which the use of posted pricing is inconsistent with competition). 60 Given market conditions such as those on the Italian petrol market that are characterized by relatively high price transparency, an implementation of a collusive pricing strategy is also conceivable without the use of facilitative devices, such as announcements of future price steps, despite the fact that under such circumstances an implementation could not occur that smoothly. The approach suggested in this paper including an econometric analysis could also then demonstrate that an observable pricing policy would have to be characterized as collusive. 61 Athey & Bagwell, supra note 47; Athey, Bagwell & Sanchirico, supra note 47; Hanazono & Yang, supra note 47; Garrod, supra note 47. 62 The concepts of “rigid” and “sticky” pricing differ. Rigid pricing refers to a market environment with independent and identically distributed costs where firms' prices are unresponsive to current cost changes. In contrast, sticky pricing refers to settings with permanent cost changes, such as the gasoline market. In those markets the issue is when to change the price and to respond to permanent cost changes. 63 Athey & Bagwell, supra note 47. It should be noted, however, that with other parameter configuration other, more complex type of equilibria are possible. 64 While these are general oligopolistic models and thus none of them perfectly matches the Italian gasoline industry, they provide a sound theoretical justification of observed rigid pricing patterns in cartels. 65 Zhongmin Wang, (Mixed) Strategy in Oligopoly Pricing: Evidence from Gasoline Price Cycles Before and Under a Timing Regulation, 117 J. Pol. Econ. 987 (2009). 66 Abrantes-Metz, Froeb, Geweke & Taylor, supra note 48; Blanckenburg, Geist & Kholodilin, supra note 48; Connor, supra note 14. 67 Abrantes-Metz, Froeb, Geweke & Taylor, supra note 48. 68 Blanckenburg, Geist & Kholodilin, supra note 48. 69 Frederic M. Scherer & David Ross, Industrial Market Structure and Economic Performance 346 (Houghton Mifflin Co. 3d ed. 1990). 70 Rotemberg & Saloner, supra note 45. 71 Mouraviev & Rey, supra note 46. 72 Andreoli-Versbach & Franck, supra note 12. 73 See Paul R. Zimmerman, The Competitive Impact of Hypermarket Retailers on Gasoline Prices, 55 J.L. & Econ. 27 (2012). EU countries differ with respect to Italy, for example, in the number of gas stations owned by hypermarkets that compete aggressively to attract customers to their stores. Using state-level data of U.S. gasoline prices, Zimmerman shows the positive competitive impact of hypermarket retailers. The dif-in-dif analysis as carried out in Andreoli-Versbach & Franck, supra note 12, assumes that “market trends” would be the same in the treatment and control group, while structural country specific market differences are captured by the fixed effects. 74 See infra Figure II (plotting average weekly prices in Italy, the EU, and the Brent). 75 Alberto Abadie & Javier Gardeazabal, The Economic Costs of Conflict: A Case Study of the Basque Country, 93 Am. Econ. Rev. 112 (2003). 76 Alberto Abadie, Alexis Diamond & Jens Hainmueller, Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California's Tobacco Control Program, 105 J. Am. Stat. Ass'n 493 (2010). 77 Both specification (2) and specification (3) were performed using robust standard errors. 78 See Giancarlo Spagnolo, Managerial Incentives and Collusive Behavior, 49 Eur. Econ. Rev. 1501, 1502 (2005). Spagnolo focuses on the role of observable CEO compensation schemes with regard to tacit collusion. He concludes that “a strong pro-collusive effect may well outweigh agency costs and transform apparently puzzling compensation practices into profitable ‘governance’ instruments.” 79 See David Besanko, David Dranove, Mark Shanley & Scott Schaefer, Economics of Strategy (Wiley 5th ed. 2010). One of the standard textbooks used in MBA courses that deals with competitive strategy is “Economics of Strategy.” Chapters 9 and 10 extensively deal with the issues of “Strategic Commitment” and “The Dynamics of Pricing Rivalry,” respectively, which are key elements to sustain collusion. Under the heading “The golden age of micro,” the journal “The Economist” discussed in its issue of October 19, 2012 why leading academic microeconomists are top advisers at firms such as Microsoft and Amazon. 80 Thomas C. Schelling, The Strategy of Conflict (Harvard Univ. Press 1960). 81 See Maskin & Tirole, supra note 44 (building commitment in a dynamic Bertrand model through exogenous costs such as menu costs and showing that sticky prices can serve as a commitment device to sustain higher prices than under static Bertrand). See also Wang, supra note 65 (studying firms' pricing strategies in a gasoline market before and after the introduction of a law which regulated firms' timing of price changes and highlighting the importance of short-run price commitment in tacit collusion). 82 See Figure 1. 83 Severin Borenstein, James B. Bushnell & Frank A. Wolak, Measuring Market Inefficiencies in California's Restructured Wholesale Electricity Market, 92 Am. Econ. Rev. 1376 (2002). 84 See Massimo Motta & Michele Polo, Leniency Programs and Cartel Prosecution, 29 Int'l J. Indus. Org. 705 (2003) (discussing leniency theoretically); see also Joseph E. Harrington, Jr., Optimal Corporate Leniency Program, 56 J. Indus. Econ. 215 (2008); Nathan H. Miller, Strategic Leniency and Cartel Enforcement, 99 Am. Econ. Rev. 750 (2009) (providing empirical evidence on the effects of leniency). 85 See Nicholas Petit, The Oligopoly Problem in EU Competition Law, in Handbook on European Competition Law 259, 314 (Ioannis Lianos & Damien Geradin eds., Edward Elgar 2013) (identifying, with a view to the EU Merger Regulation and, in particular, considering its application by the European Commission, substantive, procedural, and remedial issues that reveal significant shortcomings of merger control as a tool to address tacit collusion). 86 See William E. Kovacic, Robert C. Marshall, Leslie M. Marx & Halbert L. White, Plus Factors and Agreement in Antitrust Law, 110 Mich. L. Rev. 393, 435 (2011) (offering a list of “super plus factors,” including inter alia, “[a] reliable predictive econometric model that accounts for all material noncollusive effects on price, estimated using benchmark data where conduct was presumed noncollusive, produces predictions of prices that do not explain the path of actual prices in the period or region of potential collusion, at a specified high confidence level”). 87 See supra Part II. 88 6 Phillip E. Areeda, Roger D. Blair, Herbert Hovenkamp & Christine Piette Durrance, Antitrust Law § 1347, at 198 (Wolters Kluwer 3d ed. 2010). 89 In the U.S., several courts have recognized such claims for “umbrella damages.” See, e.g., Loeb Indus., Inc. v. Sumitomo Corp., 306 F.3d 469 (7th Cir. 2002); In re Beef Indus. Antitrust Litigation, 600 F.2d 1148 (5th Cir. 1979). The European Court of Justice held that Article 101 TFEU precludes the interpretation and application of domestic legislation enacted by a Member State which categorically excludes, for legal reasons, any civil liability of undertakings belonging to a cartel for loss resulting from umbrella pricing. See Case C-557/12, Kone AG and Others v. ÖBB-Infrastruktur AG, 2014. 90 See supra Parts III.A and III.B. 91 There would be nothing inherently new in prohibiting a certain pricing behavior. Market dominant firms are not allowed to engage in predatory pricing. And just as it has to be defined with regard to a specific industry whether a certain pricing policy has to be considered “predatory,” courts would also have to define “sticky pricing” industry-specifically as infrequent price changes in response to changes of input costs or demand patterns. 92 See supra Part III.E. 93 Case C-62/86, AKZO v. Commission 1991 E.C.R. I-3359 ¶ 60. 94 In British Airways v. Commission, the General Court (then known as the Court of First Instance) confirmed the Commission's view that British Airways was dominant in the UK market for the procurement of air travel agency services with a market share of 39.7 percent, Case T-219/99, British Airways v. Commission 2003 E.C.R. II-5917 ¶ 189-225. Market dominance was not contested at appeal, Case C-95/04 P British Airways v. Commission 2007 E.C.R. I-2331 ¶ 14-15. 95 See, e.g., Sigrid Stroux, US and EC Oligopoly Control 168 (Kluwer Law International 2004); Valentine Korah, An Introductory Guide to EC Competition Law and Practice 126 (Hart Publishing 9th ed. 2007); Giorgio Monti, EC Competition Law 335 (Cambridge Univ. Press 2007); Petit, supra note 85, at 334. 96 Monti,supra note 95, at 341; Petit, supra note 85, at 336. 97 743 F.2d 1114 (5th Cir. 1984). See, e.g., United States v. Ames Sintering Co., 927 F.2d 232 (6th Cir. 1990) (charging explicit attempts to initiate collusion as violations of the wire fraud or mail fraud statutes). 98 See, e.g., FTC v. Indiana Federation of Dentists, 476 U.S. 447, 454 (1986); FTC v. Sperry & Hutchinson, 405 U.S. 233, 244 (1972); FTC v. Brown Shoe Co., 384 U.S. 316, 320-321 (1966). 99 Two more recent enforcement actions by the FTC which are commonly referred to as the Valassis case and the U-Haul case (and both of which were resolved without litigation through a negotiated consent decree) do not indicate otherwise as both cases involved express unilateral efforts to engage in collusion, see Administrative Complaint, Valassis Comm'ns, Inc., FTC Dkt. No. C-4160 (Mar. 14, 2006), available at http://www.ftc.gov/os/caselist/0510008/060314cmps051008.pdf and In the Matter of U-Haul Int'l and AMERCO, 2010 WL 2966779 (F.T.C.). 100   E.I. Du Pont De Nemours & Co. v. Fed. Trade Comm'n, 729 F.2d 128 (2d Cir. 1984). 101 Id. at 133. 102 Id. at 139–140. © The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

Journal of Competition Law & EconomicsOxford University Press

Published: May 18, 2015

There are no references for this article.

Sorry, we don’t have permission to share this article on DeepDyve,
but here are related articles that you can start reading right now:

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from
Google Scholar,
PubMed
Create lists to
organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off