This paper, available here, is long and so, I am afraid, is the review. In short, the authors of this paper take issue with the assumption that each cartel in which a given firm participates is a single instance of conduct that is independent of other cartel conduct by the firm. Evidence of serial collusion by major multi-product firms is readily observable from the public record in a number of sectors, such as chemicals, electronics, car-parts, financial products or graphite. Further, collusion persists in at least three of these industries, with new investigations having recently been opened into collusion in the chemical, auto parts, and financial products markets.
The paper provides empirical evidence that many multi-product firms have each participated in several cartels over the past 50 years. It argues that traditional assumptions regarding how cartelists operate, and consequent enforcement strategies, are deficient in many aspects. Reflecting this, the authors make policy recommendations to reign in serial collusion. The article is structured as follows:
Section II provides empirical evidence of serial collusion.
This article does not deal with instances where a company is simultaneously engaged in multiple collusive practices. Instead, this article focuses on firms that participate in multiple cartels over time – i.e. firms that enter into a cartel when they have the opportunity, and move out of them when cartels collapse because of cheating or detection. Serial cartelisation is particularly worrying when it encompasses a broad range of the firms active in a sector, since this implies that all participating firms see value in colluding across many product lines. A further implication is that, in such a scenario, cartel issues could be resolved across products lines, rather than solely with respect to a given product.
The authors identify instances of serial collusion in chemicals, electronics, car-parts, financial products and graphite. The paper looks at the average duration of each cartel in these sectors, and at how many cartels each colluding firm participated in and for how long – while limiting their analysis to firms that were found to have participated in three or more cartels.
There a number of interesting takeaways from this analysis. For example, in the chemical industry – where some cartels have lasted 30 years (e.g. oxygen peroxide) – a common player to most cartels was not a chemical company. Instead, AC Treuhand is a consulting company that has played the role of cartel facilitator in at least nine uncovered cartels for prolonged periods and across a range of products – not only in the chemical, but also in the wood products’ sector. Another finding is that a number of firms participated in numerous cartels in both the auto-parts and electronic sectors: Hitachi participated in 16 cartels in both the electronics and auto parts markets, while both Mitsubishi and Panasonic participated in eight different cartels in these two sectors.
Taken together, the frequency of collusion by the firms reviewed across these sectors is striking. Despite aggravated penalties for recidivism imposed by some competition law systems, many of these firms chose to engage in a series of collusive schemes over a significant period. Furthermore, if one looks further back: (i) a number of these companies have been involved in cartels for well over a century, and (ii) collusion seems to have been endemic in a number of these and other sectors – such as chemicals or electronics, to which one can add steel.
Section III reviews the standard economic and legal analysis of single-product collusion.
In the past 20 years, anti-cartel enforcement has been modernised in important ways, especially through the adoption of vigorous enforcement policies and leniency programs in many countries. The total number of cartel cases sanctioned, the amount of civil and criminal fines imposed, and the length of prison sentences have all increased substantially during this period. Traditional activity-based measures of enforcement performance accurately suggest that the hazards for cartel participants are greater today than they were a few decades ago. However, doubts about the effectiveness of anti-cartel enforcement subsist. After all, the amount of fines and the length of prison sentences are imperfect proxies for gains in social welfare, and provide no indication about whether antitrust enforcement deters firms from engaging in illegal collusion in the first place.
To understand whether antitrust enforcement is indeed effective in deterring cartel, it is valuable to refer to studies on the law and economics of cartels – which are predominantly focused on single-product collusion. A theme that emerges from a review of the economics and law related to single-product collusion is how difficult it is to form and implement a successful cartel. These difficulties arise from challenges with reaching an agreement, detecting cheating and punishing deviations to it, as well as from external threats to collusion such as buyer resistance, new entry, and market entry of substitute products.
Overcoming these difficulties requires considerable technical and managerial skill—for example, in devising and adjusting the assignment of market shares, in tracking sales’ volumes by individual cartel members, or in constructing methods for executing the side payments necessary to ensure adherence to the common scheme. It follows that a cartel is by necessity highly relational. Cartel members cannot successfully perform complex, difficult cartelisation tasks without continual engagement, which is necessary to adapt to changing circumstances and to respond to external threats to the operation of the collusive scheme. With time, communications can be improved and the performance of the cartel’s key operational tasks can be enhanced. As with any other dimension of organisation and management, there are benefits from learning and experience with repeated performance of the same tasks over time. In effect, cartel members could realise these benefits either through repeated participation in collusive schemes (among themselves or with other firms) or by engaging third parties (e.g. an accounting firm such as AC Treuhand) which have broad experience in assisting cartels.
An inference that from the complexity, durability, and scope of many cartels is that they require commitment at the highest level of the company and the participation of key senior officials, as well as middle managers. Low level subordinates devising and executing collusion schemes without the knowledge or acquiescence of top management is unlikely in the context of massive, intricate, long-lived arrangements that figure so often in modern cartel prosecutions. Optimists believe that antitrust enforcement against cartels functions effectively, and that detected cartels are attributable to rogue managers. They believe that good corporate governance implies that top managers act on the preferences of owners and are adequately deterred from collusion given current enforcement practices. They typically endorse compliance training and criminal punishment of rogue managers as enforcement tools that increase the deterrence of rogues and mitigate agency problems within firms that sometimes allow rogues to flourish.
The authors of this paper are, to put it mildly, not particularly sympathetic to this view. While they see the value of compliance training, the authors are also concerned that such training is sometimes used to teach managers how to collude. In addition, penalties against rogue managers may be offset by indemnification or some form of compensation to managers who suffer personal penalties after a collusive agreement is uncovered. Furthermore, widespread multi-product price fixing by the same firms is inconsistent with the idea that rogue managers are alone responsible for collusion or for coincidentally hiring the same consulting firm that helped organise earlier cartels for the firm. The authors fear that inadequate deterrence means that top management, and perhaps even shareholders, sometimes welcome price fixing agreements as a tool for increasing expected profits—even after factoring in the expected costs from detection and punishment.
Section IV extends this analysis to multi-product firms engaged in collusion across a broad range of products.
Serial multi-product cartels are likely to be more resilient to economic shocks, defection and cheating. Multi-product cartel members that defect face stronger punishment than single product cartelists do. Multi-product collusion can also more easily reward cartel members by enabling side-payments in one product to offset losses that cartelists might suffer in another – while simultaneously making it easier to hide these side payments from antitrust enforcers. In effect, cartel portfolios allow for coordinated product exit/contraction and entry/expansion – a practice observed in a number of chemicals’ cartels –, which is useful to reallocate cartel profit across cartel members but is difficult for external agents to observe. Other tools available for multiproduct cartelists to allocate profits internally include joint ventures – which are common in electronics’ cartels –, subcontracts and patent licences.
Importantly, leniency can be used in multi-product cartels to punish defectors. While sacrificing collusive profit in one product, a well-targeted leniency application can strengthen the punishment threat and the stability of the cartel in other products. This is particularly the case if the defector is a single-product firm that is unaware of cartelisation by the leniency applicant in other products. It is reasonable to presume that any small firm that is engaged in a conspiracy with a large multi-product firm will become aware if their large multi-product co-conspirator has sought amnesty in a different cartelised product. In light of this information, the small single-product firm will likely inquire why their large multi-product co-conspirator took such an action. The large multi-product firm will explain that they are not willing to tolerate deviations by small firms in their cartels, and that when these deviations occur they will seek amnesty. A reputation for toughness by the large multi-product serial colluders can have two benefits for cartel payoffs. First, cheating by small co-conspirators will be substantially deterred across their entire portfolio of cartels. Second, it can promote cartel creation when fear of defection is a reason preventing firms from colluding.
Multi-product cartels also benefit from efficiencies associated with monitoring the actions of cartel members across multiple related product lines. Monitoring cartel conduct is likely to cost less for each additional cartel. Colluders and their allies (like consulting firms) are also likely to get better at managing collusion with experience. For example, some aspects of corporate culture, corporate organization, and corporate incentives that work to stabilise collusion in one product may be transferred to another product. Finally, developing and using outside cartel consultants, like AC Treuhand, should be easier and less costly on a per cartel basis.
Section V makes a number of policy recommendations.
Although the formation and implementation of a cartel is no easy task, the rewards for successful cartelisation are substantial. The history of antitrust policy since the late 19th century demonstrates that firms have responded to enhancements in competition enforcement with a host of countermeasures to avoid detection or blunt the impact of more severe punishments. The enhancement of sanctions and means of detection has led firms to take various steps to mask their coordination. There is thus an arms race between cartelists and enforcers, which places a premium on establishing a framework of legal rules and enforcement mechanisms that accounts for the dynamism of cartel behaviour and makes the fullest possible use of knowledge gained about how cartels operate and respond to new advances in enforcement methods.
Some common features of cartel structures provide a basis for meaningful policy reform. In particular, effective competition enforcement should substantially encumber (i) the ability of cartel firms to monitor one another, (ii) the ability of cartel firms to rectify unintentional deviations, and (iii) communication between cartel firms. Competition policy should also reward cartel members that make a limited range of products, have smaller market shares, and have not been detected in previous cartels for revealing cartel conduct. In the light of this, the authors propose a number of reforms to improve anti-cartel enforcement.
The first step is cartel reconstruction. The authors think that enforcers should work with colluding firms to learn how each cartel worked, who was responsible and what other markets might be affected. A better understanding of the nature of serial collusion is needed to identify indicia of serial collusion. Such work would serve three purposes: (i) to identify promising focal points for investigations; (ii) to increase our understanding of which types of observed conduct should be treated as plus factors to sustain an inference of anticompetitive agreement; and (iii) to improve the operation of compliance programs. Ways to achieve this could be to require, as a condition of leniency or for entering a plea agreement, or as an element of sentencing in a contested case, the culpable firms to engage in an externally audited reconstruction of the cartel and publicise the results.
The second step would be to pursue more extensive monitoring of serial colluders. If a multi-product firm participated in multiple cartels, it should be subject to additional requirements beyond standard penalties and antitrust compliance training. Such companies should also be required to inform the competition authority of any and all communications and transactions with competitors. Another step would be for enforcers to complement leniency programs with bounty programs aimed at peeling small firms away from cartels. A successful bounty program would make small firms unappealing targets for inclusion in a cartel, and could destabilise a cartel or at least limit its ability to raise prices. The authors also suggest that cartel facilitators should become focal points for cartel investigations.
Thirdly, the authors argue that, whenever a serial colluder submits a merger for review, this merger should be subject to a coordinated effects analysis. Horizontal mergers may be triggered by firms attempting to regain the pricing discipline they enjoyed prior to a cartel in which they participated being uncovered. Mergers may be prohibited on this basis; another reason to block a merger is that it will stabilise or encourage a cartels, i.e. the merger will produce coordinated effects. If two companies have no horizontal overlaps, but one of the firms is a serial colluder active in markets where competitors of the other firm are present, such a merger can raise concerns about the potential post-merger expansion of the serial colluder’s cartel portfolio. Enforcement agencies should make use of cartel reconstructions and the results of expanded monitoring when reviewing mergers involving a serial cartelist. The authors also suggest that government agencies should treat past collusive behaviour as a serious indication of likely future anticompetitive effects.
I think this is a very interesting piece. As you can see from the discussion above, the article contains many ideas on a variety of topics. I did not address this, but the paper also contains thorough discussions of various other relevant matters for cartel enforcement – e.g. what market structures are conducive to cartelisation, what penalty levels will be deterrent, whether and how to deter individuals working for cartelists, what are the typical mechanisms of collusion, etc..
Without going into the policy recommendations, which are worthwhile but raise a number of questions which are better discussed elsewhere – e.g. on the additional demands they would imposes on competition agencies, on the challenges faced by bounty programs or on the difficulties of coordinated effects’ reviews –, I think the paper would have benefited from a more systematic evaluation of past cartel enforcement. This is not the authors’ fault, as they may not have the necessary information at their disposal, but the paper’s analysis is based on a (large) number of individual cases and, as a result, impressionistic.
For example, I would have liked to see more evidence that collusion often requires senior management involvement. The article – which, to be fair, is already quite long – presents a number of reasons for why such involvement is likely. However, these reasons seem to be inferences derived from the complexity of forming and implementing a cartel, and the fact that many cartelists engage in serial collusion across various products. Instead, I expected the paper to provide evidence of senior-management involvement in the cartel cases reviewed or in the business literature. I think it is to the paper’s credit, however, that it opens many new avenues of research along these lines.