‘Countervailing Buyer Power in EU Competition Law’

Our second guest speaker today is Ignacio Herrera-Anchustegui (BECCLE and University of Bergen), a PhD researcher and expert in public procurement law. Ignacio will discussing ”Countervailing Buyer Power in EU Competition Law”. An abstract for his presentation can be found below.


Countervailing buyer power (“CBP”) is arguably the sub-topic of buyer power that has received the most attention by the literature and the case law. Contrastingly, in the EU case law and Commission’s practice its importance when deciding cases has been rather limited. CBP acts as a competitive constraint mitigating or nullifying the market power effects by a supplying undertaking. It acts as a defence mechanism precluding an undertaking from significantly impeding competition or from it being dominant through the exercise of market power as it is not able to behave independently of its buyers. Because of its nature, it is assessed as part of the relevant buyer market power analysis.

This paper analyses CBP from a general perspective in the light of the case law, Commission practice and other authoritative sources. I aim at contributing to the current literature by defining what ought to be understood by CBP, analysing how the EU judiciary and Commission treat this competitive constrain as seller market power neutralizer, identifying its sources, clarifying whether EU competition law employs a generalized and coherent legal treatment to CBP or whether the approach varies depending on the case type, and discussing substantive aspects of it, such as its scope of application and the required extent of its effect. My approach, unlike most previous literature, is comprehensive in the sense that I do not limit my analysis to specific competition law areas nor to just case commentary, which is also a contribution to the current state of the law.

Ignacio’s presentation takes place from 13:45-14:30 in the Thomas Paine Study Centre, Room 0.1.

‘The relationship between the Commission’s leniency programme and the Directive on damages regarding breach of the competition rules (Directive 2014/104/EU)’

The Autumn edition of the CCP Seminar Series is well under way – check out our programme here. On Friday 30th October, we have the double pleasure of welcoming two outstanding PhD researchers from the University of Bergen, Ingrid Halvorsen Barlund and Ignacio Herrera-Anchustegui. Ingrid is our first presenter in this special double-session, and will be exploring ”The relationship between the Commission’s leniency programme and the Directive on damages regarding breach of the competition rules (Directive 2014/104/EU)”. An abstract for her paper can be found below.


This abstract is based on a chapter of my PhD-project on ‘the Commission’s leniency programme within EU competition law – With emphasis on the regulatory framework, scope and effects of the Commission’s leniency programme as a public enforcement instrument of secret cartels, and the interplay with private enforcement through damages claims following infringements of the competition rules’ (working title). The chapter examines whether the Directive on Damages succeeds in balancing the public leniency enforcement of secret cartels with subsequent private damages claims arising from that same cartel activity from an optimal enforcement perspective. In this regard, the chapter looks into the situation of private enforcement of cartel activity through follow-on damages claims in the EU starting with Courage until now, going through the relevant case law and the Commission’s documentation leading up to the Directive. Concerning the Directive, the chapter focuses on the rules on disclosure and liability. Based on these analyses, the chapter questions whether there are other solutions to this interaction which possibly promotes a more efficient enforcement of Article 101 of the Treaty on the Functioning of the European Union (TFEU).

By leniency in EU competition law, this thesis refers to an enforcement tool that provides a cartel participant, who confesses and delivers evidence against their co-conspirators to the Commission, an immunity from or reduction of administrative fines, depending on the time and value of the evidence provided. The most recent leniency programme of the Commission is from 2006 and is called the ‘Commission Notice on Immunity from fines and reduction of fines in cartel cases’.

According to Article 1, the Directive seeks to ensure the right for anyone who has suffered harm caused by an infringement of the competition rules to claim full compensation, which is a codification of the right for compensation established by the Court’s case law, and the responsibility for each Member State to ensure this. Concomitantly, it emphasizes the importance of an effective public enforcement of cartels, referring inter alia to leniency programmes. The Directive thus seeks to establish a common standard in all Member States for the interaction between public and private enforcement of the competition rules at the EU level, stressing that to achieve an optimal enforcement of the competition rules this interaction has to be taken into consideration. To assure an effective enforcement of the competition rules, different – and to a certain degree “opposing” – interests of parties need to be protected when competition authorities publicly enforce competition laws and later on when private actors file claims for damages before the courts. By “opposing interests” this abstract specifically refers to the conflict that arises in a private follow-on damages suit resulting from a cartel infringement between a leniency recipient and the victims of that cartel activity seeking compensation. The risk of follow-on damages claims after having received lenient treatment from the competition authorities undermines the effectiveness of a leniency programme. To understand these conflicting mechanisms, the abstract will start by explaining the rationale of a leniency instrument.

The Commission’s leniency programme enforces Article 101 TFEU. Article 101 prohibits “all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the internal market”. The Notice is only directed at “the most serious violations” of Article 101, which accordingly are “secret cartels”, cartels being defined as “agreements and/or concerted practices between two or more competitors aimed at coordinating their competitive behaviour on the market and/or influencing the relevant parameters of competition […]”. The secret nature of these infringements demands for non-traditional enforcement measures like leniency programmes. Because these infringements are very difficult to detect and because the cartel participants often go to great lengths in maintaining and keeping the cartel secret, the Commission’s leniency programme is based on a ‘carrot and stick’-strategy. Cartel participants are profit driven, meaning that their incentives are oriented towards what gives the highest economic gain. This economic orientation, in addition to the secrecy, explains the “carrot and stick”-strategy, where the balance between creating deterrence through tough sanctions on the one hand (the stick) and offering on the other hand a lucrative leniency regime (the carrot) is vital for the detection and prevention of cartel activity. The risk of follow-on damages claims threatens this interaction by undermining the ‘carrot’-effect as it diminishes the attractiveness of the leniency programme by adding to the ‘stick’-effect. The value of receiving immunity from fines from the competition authorities will decrease if the leniency applicant still risks damages claims in front of the courts. Despite the fact that the object of these damages is to compensate the victims of cartel activity, the damages will be perceived as adding to the fines escaped by the immunity recipient during the public enforcement. The controversy of juxtaposing damages with fines in the European enforcement tradition will be briefly commented on at the end of this abstract.

The impression of this thesis’ studies so far is that the Directive has tried to handle the conflict by outlining a compromise between the conflicting interest of an immunity recipient and the victims of cartel activity which arguably leads to both sides losing. This thesis therefore argues that the Directive’s solution might not be the way to go, and that it seemingly fails in seeking the objective of an optimal balance between public leniency enforcement and private follow-on damages enforcement. On the contrary, the outcome seems to be neither optimizing the public leniency enforcement nor the private follow-on damages enforcement of the competition rules. The abstract will outline the rules on disclosure and liability in the Directive as these are of particular interest and illustrating of why the Directive might not have found the solution.

Concerning disclosure of documents, the PhD-chapter analyses the case law on disclosure of evidence in the Commission’s file before the Directive’s entry into force, emphasizing what seems to have been different views of the Commission and the Court concerning the protection of the immunity recipient’s interests on the one hand versus protection of the victim of cartel activity’s interests on the other. The Court arguably was more in favour of protecting the latter and therefore promoted a pragmatic approach to disclosure of leniency information. However, Article 6 (6) of the Directive now states that “for the purpose of actions for damages, national courts cannot at any time order a party or a third party to disclose […] leniency statements”. Further, Article 7 (1) provides that “Member States shall ensure that evidence in the categories listed in Article 6 (6) which is obtained by a natural or legal person solely through access to the file of a competition authority is either deemed to be inadmissible in actions for damages or is otherwise protected under the applicable national rules”. According to Article 2 (16) of the Directive “statements” are defined as

an oral or written presentation voluntarily provided by, or on behalf of, an undertaking or a natural person to a competition authority or a record thereof, describing the knowledge of that undertaking or natural person of a cartel and describing its role therein, which presentation was drawn up specifically for submission to the competition authority with a view to obtaining immunity or a reduction of fines under a leniency programme, not including pre-existing information“.

The prohibition does not regard “pre-existing information” which according to Article 2 (17) is defined as “evidence that exists irrespective of the proceedings of a competition authority, whether or not such information is in the file of a competition authority”. The fact that leniency statements are exempted from disclosure at all times is said to preserve the incentive for undertakings to provide information to the competition authorities as it hinders the leniency applicant from becoming an easy target for damages claims. This thesis agrees with the fact that revealing this type of information could hinder the effectiveness of a leniency programme. However, to assure both efficient leniency enforcement and efficient private enforcement of follow-on damages claims, maybe the solution is not to have detailed rules on disclosure. Instead, maybe focus should be on changing the rules on liability outlined by the Directive.

The immunity recipient is according to paragraph 11 of the Directive only jointly and severally liable for the losses the undertaking has caused to its own direct and indirect purchasers and/or providers. If the cartel has caused harm towards others in addition to the cartel’s providers or customers, the leniency recipient is only responsible for a relative part proportionate to the losses directly linked to the undertaking. Only when the claimants cannot obtain full compensation from the other participants, will the immunity recipient be responsible for the whole loss. The loss shall according to Article 3 of the Directive be fully recovered. Member States are obliged to follow the European principle of full compensation, including compensation of overcharges according to Article 12. According to Article 10 (3) limitation periods shall be at least five years, starting to run from the moment the infringement has ceased and the claimant “knows, or can reasonably be expected to know” of the behaviour, that this behaviour constituted an infringement of competition law which caused harm to the claimant, and the identity of the infringer, see Article 10 (2).

The point of this discussion is to highlight the fact that even if the liability of an immunity recipient is limited, this liability read together with the rules on quantification of harm and limitation periods implies that the immunity recipient still risk facing damages claims of an unknown number and size. This creates uncertainty, and has made this thesis question how attractive the compromise outlined in the Directive is not only for the immunity recipient, but also for the victims of cartel activity. The leniency applicant on the one hand still risk an unknown number and size of claims in front of national courts, whereas the victims of cartel activity on the other hand are denied access to evidence.

That is why this thesis has looked into an alternative solution to the one outlined in the Directive. This proposal, inspired by among others a comment by Paolo Buccirossi, Catarina Marväo and Giancarlo Spagnolo on ‘Leniency, damages and EU competition policy’, is to further limit the immunity recipient’s liability for damages, possibly abolish it, and give the victims of cartel activity full access to the Commission’s file. This would mean revealing the identity of the immunity recipient, which initially constitute a disincentive for the cartel participants to cooperate with the authorities. However, this disincentive will be outweighed by the reduced liability which will make the carrot juicier and the stick more painful if the cartel participant does not confess. With this solution, the cartel participants not cooperating with the competition authorities face a greater risk as jointly and severally liable with the disclosure of the leniency information. Full access to the Commission’s file, including leniency statements, will on the victims’ part facilitate the ability to obtain compensation.

This chapter of the PhD-thesis will also as already mentioned contrast the Directive to the US system, highlighting the issue of classifying damages more or less as a sanction in relation to the immunity recipient, juxtaposing damages with fines. This is questionable in the European enforcement tradition where damages are perceived as a compensatory measure. The classification problem therefore adds to the traditional role of damages in European private enforcement. The US does not follow the principle of full compensation. In the US, claimants are entitled to sue for three times the losses suffered. This is called ‘treble damages’, providing the damages with a punitive element. Concerning the interaction between leniency and follow-on damages, section 213 of the Antitrust Criminal Penalty Enhancement and Reform Act of 2004 ‘detrebles’ damages for the immunity recipient, in addition to providing an exemption from joint and several liability. This means that the immunity recipient is only liable for the actual damages attributable to its own conduct, and that rather than being liable for three times the damages caused by the entire unlawful cartel, the immunity recipient only risks single damages. Furthermore, the US has a different antitrust enforcement tradition than in Europe, relying more on private enforcement and on public criminal enforcement. The US system is therefore to a higher degree based on deterrence and punitive elements than what traditionally has been the case in Europe. Despite these fundamental differences, but also because of these differences, seeing as the US is considered to be the ‘pioneer’ in antitrust enforcement, this thesis finds it interesting to contrast the EU system with the US system with the purpose of suggesting changes and improvements de lege ferenda to the interaction between the public leniency enforcement of secret cartels and follow-on damages claims arising from that same cartel activity.

This chapter will conclude with a summary of its findings and a discussion of its most relevant aspects. Still a work in progress, any feedback is very much welcome.


Ingrid’s presentation takes place from 13:00-13:45 in the Thomas Paine Study Centre, Room 0.1. Please note this seminar is part of a double-session which includes an additional presentation by Ignacio Herrera-Anchustegui from 13:45-14:30. 

‘Private v public sanctions: the case of cartels’

On Friday 10th July, we rounded off our Summer Seminar Series in style with the dream pairing of Franco Mariuzzo (CCP and ECO) and Peter Ormosi (CCP and NBS) presenting ‘Private v public sanctions: the case of cartels‘, a joint project with Antonios Karatzas (Warwick). An abstract for their presentation can be found below.


Economics literature on the relationship between private and public sanctions has led to a consensus that private (market-based) sanctions can act as an important deterrent to corporate misbehaviour inasmuch as they internalise the social costs of these offences. On the other hand public sanctions are needed where the harm caused by the offence is not internalised (for example because the damaged party remains oblivious to the offence); in these cases the amount of the sanction should equal the un-internalised social cost. We provide further empirical evidence to this literature by looking at cartels, using a novel methodology that enables us to directly control for the magnitude of the private sanction in our model. This allows us to empirically study the substitutability between fines and reputation for firms that have been found guilty of cartel illegal conduct.

We measure reputation with a composite of the frequency of media exposure and the intensity of the media content. Media exposure is captured by a count of sources that document a firm’s illegal behaviour in the cartel and intensity is measured by a sentiment analysis of the text of the published news. We employ an event study technique over a sample of about 300 public companies that belong to about 100 cartels detected by the Competition Commission during the period 1990-2012. Our results confirm that public and private sanctions are not perfect substitutes, and that the intensity of the media content has a larger punishment effect than exposure; we call this “quality for quantity” effect. In addition we find that the effect of reputation is larger for firms that belong to cartels with non-atomistic customers.

This is the final seminar of the Summer but we will return in September for another excellent series of interdisciplinary presentations on competition policy and regulation. We will be announcing the programme for the Autumn Seminar Series in due course and you can keep up-to-date by visiting our designated seminar pages on the CCP website.

‘In Gov We Trust: Voluntary Compliance in Networked Investment Games’

Only two seminars remain in this semester’s CCP Seminar Series but they have most certainly been worth the wait. This week’s penultimate seminar on Friday 3rd July sees Natalia Borzino (CCP and ECO) presenting her research entitled ‘In Gov We Trust: Voluntary Compliance in Networked Investment Games‘, which she has undertaken with Enrique Fatas (CCP & UEA) and Emmanuel Peterle (University of Goettingen). Natalia is a PhD Researcher and Associate Tutor in the School of Economics at UEA. In addition to CCP, Natalia is also a member of the Centre for Behavioural and Experimental Social Science (CBESS) at UES. An abstract for her paper can be found below.


We conduct a controlled laboratory experiment to investigate trust and trustworthiness in a networked investment game in which two senders interact with a receiver. We investigate to what extent senders and receivers comply with an exogenous and non-binding recommendation. We also manipulate the level of information available to senders regarding receiver’s behaviour in the network. We compare a baseline treatment in which senders are only informed about the actions and outcomes of their own investment games to two information treatments. In the reputation treatment, senders receive ex ante information regarding the average amount returned by the receiver in the previous period. In the transparency treatment, each sender receives ex post additional information regarding the returning decision of the receiver to the other sender in the network.

Across all treatments and for both senders and receivers, the non-binding rule has a significant and positive impact on individual decisions. Providing senders with additional information regarding receiver’s behaviour affects trust at the individual level, but leads to mixed results at the aggregate level. Our findings suggest that reputation building, as well as allowing for social comparison could be efficient ways for receivers to improve trust within networks.

The seminar takes place from 13:00-14:00 in the Thomas Paine Study Centre, Room 1.03.

‘The Damages Directive fails to deliver – Can it be fixed?’

After a short interval for the Annual Conference last week, the CCP seminar series returns on Friday 26th June with the our good friend Sebastian Peyer (University of Leicester) presenting ‘The Damages Directive fails to deliver – Can it be fixed?‘. Seb is a Lecturer in Law at the University of Leicester and an alumnus of CCP and the UEA Law School. He specialises in the area of the private enforcement of competition law and his research interests also encompass empirical legal studies, law and economics and comparative law. An abstract for his seminar can be found below.


The EU Damages Directive came into force in December 2014. It seeks to ensure the effective private enforcement of competition law rules by facilitating claims in the courts of the EU Member States. However, the proposed measures do not address pressing issues such as claim aggregation or the funding of claims. Instead, the Directive introduces complex rules regarding access to information and joint and several liability.

This paper investigates the goals and key features of the EU Directive on antitrust damages actions. It demonstrates that the EU framework, if implemented without further changes, is unlikely to encourage more claims and create the envisaged level playing field. The Member States, however, could devise a more balanced system of private actions if they are willing to regulate private antitrust actions beyond the Directive’s remit.

The seminar takes place from 13:00-14:00 in the Thomas Paine Study Centre, Room 1.03.

CCP Annual Conference 2015 Live Blogging Session 6: The Value of Information

Steve Tadelis (University of California, Berkeley, USA @BerkeleyHaas) talked about the challenges faced by market platforms in the presence of reputational externalities and biased feedback. Reputation mechanisms used by platform markets suffer from two problems. First, buyers may draw conclusions on the quality of the platform from single transactions, causing a reputational externality. Second, reputation measures may be coarse or biased, preventing buyers from making proper inferences.

Steve explored the limits of reputation mechanisms, their impacts on the marketplace, and ways in which a platform designer can mitigate these adverse impacts. By using an unobservable measure of seller quality, Steve demonstrated the benefits of the approach of a large-scale controlled experiment with eBay data. He claimed that platforms can benefit from identifying and promoting higher quality sellers. By highlighting the importance of reputational externalities, he also charted an agenda that aims to create more realistic models of platform markets.

Nick Anstead @NickAnstead (London School of Economics and Political Science) presented a paper that addresses a number of questions raised by the development of data-driven campaigning techniques. The recent General Election in the UK saw a very data-intensive campaign. Parties harvested massive amounts of information from both traditional and more modern sources, seeking to use them to better target the voters predicted to be more central to their success.

Nick revealed the source of the datasets used by political parties and he highlighted the reasons for the adoption of new methods. He discussed the effectiveness of the ways in which the parties used the data in the recent UK elections. Finally, he mentioned the effect of such methods on citizen participation and the quality of democracy.




CCP Annual Conference 2015 Live Blogging Session 5: New products – New Regulation?

Franco Mariuzzo (Centre for Competition Policy and School of Economics, University of East Anglia @UEA_Economics) presents his work on mobile applications, which is a fast-growing market whose proper regulation deserves attention.. In September 2014 more than 1.3 million apps were available in the Apple and the Android app stores. This market has low barrier to entry and has a high degree of competition.

He argues that strategic versioning is profitable for application developers because they will be able to use it to attract new customers, leading to continuous growth of downloads. A stylized theoretical model is presented to describe why and when updates should be released by developers in order to maximise profits. Then an empirical test is conducted using data on the top 1000 apps on Apple and Google Play stores for 5 European countries over 2 years. His finding confirms that, in general, updates to applications boost downloads and are more likely to be released when the app is experiencing a drop in downloads. He cites the finding as evidence that developers use updates strategically in order to “resurrect” their applications.


John Street (Centre for Competition Policy, University of East Anglia @Politics_UEA) discusses creativity and copying in the digital age. John states that there is a parallel music industry between digital and analogue, which runs in both complementary and conflicting ways. Differing considerations apply in each. The increasing use of digitisation and the fight between streaming companies and record companies is expanding.

John looks at how “creativity” and “copying” are distinguished in the musical world, by exploring the politics of the words. Having looked at the manifestos produced for the recent general election, John shows how the politics of digital markets controls how we see copyright. There are competing claims, economic, public policy and natural/moral rights. These three principles leads to different outcomes, which is supported by legal commentary.   John asks what the ends should be. The answer to this question leads to the adoption of a certain copyright basis. John argues that even within national systems, different opinions abound.

John also looks at the legal cases, the choices of artists and their decisions in making music. Having delved down into the decision making process itself, to determine what is “original” and what a “copy” is, John, with the help of case studies, argues that the decisions of all actors in the field are political in various ways. This extends to who makes the decision. Is it a judge, a jury or an expert? Each have different approaches and views as to what is copying and what is not.

John states that these issues have important implications for the regulation of the digital music market. John argues that a good copyright system can be identified by what kinds of political principles inform its rules, how decisions are made and who is involved in, and influences the system.



CCP Annual Conference 2015 Live Blogging Session 4: Trust in Digital Markets

Alex Chisholm (Competition and Markets Authority @CMAgovUK) opens up Session 4 with his presentation on “Data and Trust Concerns in Digital Markets: What are the Concerns for Competition and for Consumers?”. A transcript of his speech will be available to download from the CMA website later today.

Speech delivered at 11th CCP Annual Conference

CMA Chief Executive Alex Chisholm delivers speech on “Data and Trust Comcerns in Digital Markets”

Jonathan Porter’s (Ofcom @Ofcom) presentation mainly covers the economics of personal data and data privacy, the practical issues with informed consent in an on-line word and the potential implications for Ofcom’s work on the ‘Internet of things.’ He discusses the application of both the standard economic framework and behavioural biases to the analysis of data privacy concerns. He talked about the impact of technology on privacy and the economics of privacy to consumers and firms. The key features are (i) the market for personal data (ii) the market for privacy can be thought of as two sides of the same coin leading to different trade-offs between costs and benefits and (iii) the positive/negative externalities from data transmission.
When consumers are irrational, the market will then not be able to provide the efficient level of data privacy. The informed consent is a key concept that has underpinned the legal framework for personal data processing. In practice, people do not read the privacy policies and 55% of the internet users spend 15 seconds reading the privacy policy statement. This can be due the difficulty in understanding the ‘legal jargon’ privacy policy. He added that understanding could be improved by using software tools: browser add-ons, different formats, visual prompts, “smarter” notifications and Just-in-time “nudges”. Building awareness and effective communication to improve consumer engagement are some of ways that might improve the chance that consumers act upon information.


Greg Taylor @EconTaylor (Oxford Internet Institute) rounds off Session 4 by considering how the existence of contracts between firms and intermediaries affect, on the one hand, the quality of advice received by consumers and, on the other hand, firms’ incentives to invest in improving the quality of their products. It is important to consider these dynamics as, in many industries, consumers rely on advice and recommendations from an intermediary when choosing between competing products. Moreover, competition concerns lie in the fact that intermediaries in many industries are not independent of the market. Indeed, an intermediary may be owned by a firm in the market, thus raising concerns regarding its objectivity.

Greg takes us through the paper which he has co-written with his colleague Alexandre de Cornière. They use a ‘quality competition model’ that involves one intermediary and two firms who choose how much to invest. Greg explained the contrasting behaviours of an informed and uninformed consumer when selecting an intermediary firm. In Nash Equilibrium, profits are zero. He then considered the situation in the context of vertical integration. The non-integrated firm (without intermediaries) experiences a greater incentive to invest in low quality products than products of high quality. However, the integrated firm will have a greater incentive to invest in order to benefit from economies of scale. Greg cites the Google case as an example of this application.

In terms of price comparison, under vertically integrated situation, in equilibrium, (i) all prices increase, (ii) consumer welfare ambiguously falls, and (iii) the endorsed firm is always the ‘worst one’. Where quality is the main dimension of competition, the integrated firm has a high incentive to invest, whereas the non-integrated form will have a low incentive. In this instance, no consumers can benefit from becoming informed when the intermediary’s profit is maximised.



CCP Annual Conference 2015 Live Blogging: Start of Day 2

Welcome to the start of the second day of the Centre for Competition Policy’s 11th Annual Conference on Competition in the Digital Age.


CCP Annual Conference 2015 Live Blogging Session 3: Platforms and Price Relationship Agreements

Kai-Uwe Kuhn (University of Michigan, USA @umichECON) shed light on recent cases involving platform intermediaries, e.g., hotel booking and price comparison websites, by disentangling the effects of best price clauses offered by those platforms. By exploring the demand elasticity effect, substitution effect and entry-deterrence effect of best price clauses, Kai-Uwe commented on the anticompetitive impact that best price clauses bring about. Best price clauses could lead to higher prices since they tend to lower the levels of competition between service providers such as hotels. But since the best price clauses only affect end consumer substitution rather than seller substitution, the investment incentives are not necessarily undermined. We are in a dynamic environment in which there is an increasing trend of integration between booking and price comparison websites. Whilst issues in these two-sided markets remain complex, Kai-Uwe suggested that one could approach the problem by examining incentives under each separated vertical relations.


Matthijs R Wildenbeest (Indiana University, USA @KelleyIndy) presented work co-authored with Babur De los Santos (Indiana University), regarding the impact of vertical price restraints in the market for e-books. Their analysis is based on a suit filed by the Department of Justice in 2012 against five e-book publishers and Apple. In 2010, these publishers simultaneously adopted the agency model, which allowed them to directly set e-book prices. Amazon, who at the time followed an aggressive pricing strategy, was forced to adopt the agency model as well, along with other retailers.The suit resulted in a settlement that prevents the publishers from interfering with retailers’ ability to set e-book prices.

Using a unique dataset of daily e-book prices for a large sample of books across major online retailers, Matthijs and Babur estimate the effect of the return to wholesale model on retail prices. Their analysis reveals that e-book prices for titles that were previously sold using the agency model decreased by 18% at Amazon and 8% at Barnes & Noble. They found a stronger effect on the fiction genre, which is a more competitive book market. Their results illustrate a market where upstream firms prefer higher retail prices than downstream retailers. These findings help shed some light on the effects of the agency model on pricing. Matthijs suggests that they could be applied to other markets using agency models.