The privacy & competition law overlap: co-operation between enforcement agencies?

Last week, the European Data Protection Supervisor (EDPS) released an Opinion on coherent enforcement of fundamental rights in the age of big data (available here). It builds on a Preliminary Opinion issued by the EDPS in 2014, which aimed to launch a debate on how to apply the EU’s objectives and standards in areas such as data protection, consumer protection and competition more holistically. 

Recognising that the Commission’s wide-ranging Digital Single Market strategy presents an opportunity to launch a new, coherent approach, the EDPS makes recommendations (amongst others) for: (i) how merger controls should take personal data into account, and (ii) a voluntary network where regulatory bodies can share information (a Digital Clearing House). 

Is personal data an asset that should be considered in mergers?

The EDPS Opinion considers that the largest web-based service providers (Google, Amazon etc., some of the biggest companies in the world) “owe their success to the quantity and quality of personal data under their control as well as to the intellectual property required to analyse and to extract value from these data”.  And it’s true that gaining access to customers’ personal data has been a significant factor in some of the big tech acquisitions of the last couple of years (Facebook purchasing WhatsApp for example, or Microsoft’s pending acquisition of LinkedIn). 

In a speech in March this year (here), Commissioner Vestager highlighted the fact that data is an asset, and that it can be a company’s assets rather than turnover that make it an attractive target.  She warned that important deals which warrant review may be missed under the current system, as the acquisition of a company with access to – as yet unmonetised or undervalued – data may not meet the Commission’s turnover test (as with Facebook/WhatsApp, which only fell within the Commission’s remit due to Facebook’s Article 4(5) request).

The EDPS supports greater scrutiny of acquisitions of this sort, and recommends that the expertise of independent data authorities should be utilised to consider the effect of such acquisitions on consumer welfare. 

Is privacy a competition law issue?

Commissioner Vestager downplayed the importance of privacy and data for competition enforcement in a speech in Copenhagen on 9 September (text here).  She noted that “our first line of defence will always be rules that are designed specifically to guarantee our privacy” and that “we shouldn’t be suspicious of every company which holds a valuable set of data”.  However, she did leave the door open for competition enforcement action in this area, recognising that a company in control of a unique set of data may be able to use it to shut rivals out of the market.

The EDPS Opinion also considers the interface between competition and privacy, but with a particular emphasis on personal data.  It speculates that in the near future machine-learning algorithms may be able to exploit differences in consumers’ sensitiveness to price (identifiable from their personal data), enabling firms to segment the market into each individual consumer and charging according to his or her willingness to pay.

Should such an issue arise, it would prompt concerns from data protection authorities about whether personal data was being used in an appropriate way, and from competition authorities about the effect of such use on consumers and the market. 

Surely it makes sense for these authorities to share expertise on these matters?

Digital Clearing House

Even before machine-learning algorithms take over, it’s clear that there are occasions where competition and privacy overlap, and where regulators can help one another.  This already happens on occasion.  The EDPS points to examples such as: 

  • The French competition authority’s interim decision in September 2014 that GDF Suez had abused its dominant position by using personal data collected when it was a state monopoly to later offer a promotion on an open market. 
  • The UK Data Protection Authority advised the CMA on its proposal to invite households who had not switched energy suppliers for three years to opt out from having their details shared with rival suppliers.
  • Germany’s competition regulator, the Bundeskartellamt is currently investing Facebook’s privacy policies with input from a number of other national authorities – as we reported here.
The EDPS seeks to build on this kind of co-operation, proposing a voluntary network of contact points in regulatory authorities at national and EU level who are responsible for regulation of the digital sector.  Such a network could discuss the most appropriate legal regime for pursuing specific cases or complaints, and could potentially use data protection and consumer protection standards to determine theories of harm relevant to merger control and exploitative abuse cases.

From a competition law perspective, this is not uncontroversial: the relevance of other laws to the competition regime has been rejected on a number of occasions in the past.  Introducing privacy standards could open the floodgates to a need to consider, for example, environmental considerations, or industrial or social policy.  Added to which, there would doubtless be a number of practical challenges to setting up such a network – the first which springs to mind is persuading the diverse authorities involved to listen to one another!

The Brexit shaped spanner in the works

It’s too early to tell what appetite there is across Europe for a Digital Clearing House, but any UK involvement may obviously be affected by Brexit.  Aside from the politics involved, UK authorities may have to apply different legal frameworks to the rest of Europe (see our competition blogs on Brexit here and here, and our colleagues’ blog on the data protection implications here).  We’ll also have to wait and see if the CMA shares the view of the EDPS on the importance of personal data.

Either way, we expect there to be significant developments in this area in the future.

Patent Licensing and the Internet of Things – a Solution?

The Internet of Things (IoT) – a term used to describe the interconnectivity of electronic devices via the internet or wi-fi – is no longer an entirely new phenomenon.  Smart fridges, meters, watches and countless other connectable gadgets which have the ability to store and exchange data have been at the forefront of discussions by tech experts over the last few years.  The next wave of additions to the the IoT includes driverless cars and smart cities – and more as yet unimaginable changes may follow.

Issues such as security and safety, data protection and regulation have added a dash of reality to the otherwise positive picture of the IoT.  Nevertheless, an increasing number of companies are incorporating some form of connectivity into their business plans.  The most recent Ericsson Mobility Report for example, forecasts that there will be approximately 28 billion connected devices by 2021, of which 16 billion will be related to the IoT. 

However, one area that may pose a significant barrier to companies wishing to break into the emerging market of the IoT is in the arrangements to be put in place for the licensing of relevant patents and software.  The market for connected devices could be at risk of being ‘held up’ by IP disputes.  One question that has not yet been comprehensively answered is how makers of connected devices can acquire the licences necessary for their IoT products in a simple and efficient way.  

A possible solution has recently been introduced by a new licensing platform called Avanci.  Backed by Ericsson, Qualcomm and Royal KPN (among others), Avanci builds on the traditional idea of the patent pool. It aims to offer flat rate licences on FRAND terms for a collection of standard essential wireless patents, with the aim of removing the need to negotiate multiple bilateral licences.  If it works, this could speed up the expansion and uptake of the IoT.

The uptake of this initiative is yet to be seen. Its success is likely to depend upon a number of factors including:

  • Whether the licensors are able to find a mutually agreeable pricing structure;
  • Whether the price offered is acceptable to device manufacturers; 
  • The number of patent holders offering their backing to the initiative; and
  • The willingness of manufacturers to take a licence without forcing the patentees to resort to litigation and potentially costly FRAND disputes;
  • How the platform deals with the relationship between its prices and those applicable in any pre-existing bilateral deals.
It is already evident that some aspects of Avanci’s pricing may be controversial – for example, royalty rates will remain fixed regardless of how many patents are added to the platform.  This may sound like good value for licensees, but will it offer sufficient incentives for new licensors to join and make the platform a genuinely one-stop shop?  Or does it suggest that the early prices are likely to be rather high, to allow headroom for further patents to be added.  The use of fixed prices per device rather than percentage rates could also be contested by manufacturers of lower value devices, in particular if they are staring down the barrel of a patent infringement suit.

We will be continuing to monitor this fascinating space…

E-Commerce Sector Inquiry: Digital Content

Last week, the European Commission published its Preliminary Report in the e-commerce sector inquiry. The Report focusses on two main areas: goods and digital content.  In each case, the Report surveys the responses received to the requests for information sent over 
the past 15 months and sets out the Commission’s preliminary findings.  We reported on those findings on goods here, and now examine the preliminary conclusions on digital content, which focus in particular on audio-visual and music products.

Contractual restrictions in licensing agreements

Based on the market data received, the Commission concludes that contractual restrictions, in terms of licensed transmission technologies, timing of releases and licensed territories, are the norm in digital content markets.  Exclusivity is also widespread and can be granted along one or more of a number of different dimensions. For example:

  • Technological restrictions: Rights may be split according to method of transmission (e.g. satellite, online, mobile), whether content was streamed, downloaded or watched on a standard TV set; licences may also cover ancillary/usage rights on features such as catch-up services or use of multiple screens.
  • Temporal restrictions: This includes the use of “release windows” which can be the subject of complex negotiations with significant price differences depending on the period secured.  
  • Territorial restrictions: Here the concerns in relation to digital content closely mirror those identified in relation to goods (as we have discussed here).  ‘Geo-blocking’ is common, but causes are multiple.  For example, the Report finds that the main reasons why digital content providers do not typically make their services available in more than one territory are: (a) the cost of purchasing content for new territories, and (b) that the rights for the content are not available for licensing in some territories. Even those digital content providers that do make their services available in more than one Member State often offer different catalogues in each Member State, normally because they are unable to obtain licences for all of the Member States in which they are active. 
Where geo-blocking was used, many of the agreements submitted to the Commission contained clauses enabling the right holder to monitor the implementation of geo-blocking measures and to suspend distribution or even terminate the agreement if the measures weren’t implemented to its satisfaction. Almost 60% of digital content provider respondents are contractually required by right holders to geo-block, although the percentage varies considerably between licensing business models and Member States. In Italy, for example, only a minority of respondents reported that geo-blocking occurred, whereas in the UK the majority did so.

Overall, contractual restrictions of these kinds are found to be prevalent in a number of sectors investigated, and are often included in contracts of long duration.  (An exception is noted in relation to music content, where less use is made of exclusive licensing.)  The Commission notes the difficulties to which this can give rise for new entrants and smaller players.  The same is true of certain prevalent payment structures, such as requirement for advance payment, minimum guarantees and fixed/flat fees.  All of these are said to make it difficult to compete with large established providers.  (The results do also indicate that certain flexible payment arrangements have been used for certain types of digital products, which allow for payments proportionate to the number of users and facilitate competition. The Commission indicates that it is likely to encourage the wider use of such payment mechanisms, as they might promote risk sharing and streamlining of incentives along the supply chain.)

Lost in translation? 

As every good competition lawyer knows, contractual restrictions do not always translate into restrictions on competition.  The conclusion announced by the Commission, that it “will assess on a case-by-case basis whether enforcement action is necessary to ensure effective competition” is thus hardly surprising.  The question for companies active in the licensing or distribution of digital content will be to translate the Commission’s preliminary conclusions into a concrete risk assessment as to the status of existing licensing agreements and practices.

This is of course only a preliminary report, on which comments are requested.  Past sector inquiries (such as that in the pharmaceutical sector) suggest, however, that amendments after the first report are likely to be around tone and details rather than on the substance.  The data gathered by the Commission will already be under intensive analysis and case teams may already have been put together to start to pursue individual cases.  

It appears likely that such cases will be one of two main types:

  • Geo-blocking cases in which a breach of Article 101 is identified, akin to the ongoing Hollywood Studios investigation (as to which, see here), the outcome of which is likely to turn on eventual review by the EU courts (or potential further commitment decisions of the type entered into by Paramount).  
  • Abuse of dominance cases based around foreclosure of new entrants.  In principle, such cases could be brought against either content providers or rights holders, depending on the source of market power.  The Commission will want to try to focus such cases around the use of contractual provisions (e.g. the use of long-term exclusivity) rather than on refusal to license per se, where the existing law is likely to make new cases in this complex field very difficult.
It is early days, but we question how many abuse of dominance cases of this kind are likely. The Report suggests that many of the perceived market access problems arise from widespread structural issues, rather than the conduct of individual undertakings.  As content markets are currently likely to be national in scope (due to language requirements, if nothing else), it may be for national competition authorities to take the lead in this.  The Commission can of course also sponsor new legislation, as it has already done in relation to geo-blocking as it relates to goods (see here).  If that is on the agenda, a number of rounds of further consultation can be expected.  Stakeholders are invited to submit responses to this particular Report by 18 November 2016. 

EU Court to rule on ability of luxury brand owners to control online distribution

Following a dispute in Germany between perfume and cosmetics manufacturer Coty and one of its retail distributors (Parfümerie Akzente), a German court has sought clarifications on the proper application of the EU competition rules in respect of online distribution to the Court of Justice of the EU ("CJEU").  The case will likely decide how much control luxury brand owners have over distribution of their products on online platforms such as Amazon or eBay.  Whilst not a party to the German case, Amazon.com has recently sought permission to intervene in the case in order to ensure the views of third party online platforms are heard.

The original dispute arose in Germany after Coty sought to prevent Parfümerie Akzente from making sales through Amazon’s market place. 

On 18 July 2016 the Frankfurt Court of Appeals requested the CJEU to provide a preliminary ruling on whether a restriction by luxury brand owners on the use of online platforms is compatible with Article 101 of the Treaty on the Functioning of the European Union (“TFEU”).
 
Questions to the CJEU

The CJEU has been asked to consider the following:

  • Is the use of selective distribution systems by luxury brand owners to protect the ‘luxury image’ of their products compatible with Article 101(1) TFEU?
  • Is a general ban on online platforms compatible with Article 101(1) TFEU, even if the platform meets the criteria of the selective distribution system?
  • Does prohibiting the use of online platforms constitute a restriction by object under Article 101(1) TFEU as it restricts customer group retailers can sell to and their ability to make passive sales?

For luxury brand owners, the ability to have some control over the retail environment, whether online or offline, is an important consideration when setting up a selective distribution policy.  The nature of the products concerned will also be relevant in assessing whether restrictions are permissible.  Whether Amazon will be able to argue its case will however depend on whether the Court grants it permission to intervene. Demonstrating sufficient interest has historically been a difficult hurdle for companies to overcome, particularly where they have not been involved in the proceedings before the national court.  

Background to online selective distribution

The European Commission Guidelines on Vertical Restraints (“the Guidelines”) allows a supplier operating a (qualitative) selective distribution system to impose equivalency requirement restrictions on authorised distributors in respect of online versus offline sales.  For example, obligations on authorised retailers to meet equivalent criteria in respect of online product presentation and sales advice as applies to their bricks and mortar sales.  However, Coty was seeking to impose an outright ban on the use of third party platforms (such as Amazon marketplace) and no doubt sought solace in the Guidelines which do specifically permit restrictions on sales through third party internet sites where those sites display the platform’s name and/or logo (Guidelines, paragraph 54). 

In a number of Decisions on selective distribution agreements, the German Bundeskartellamt ("FCO") has taken a dim view of prohibitions on online sales.  For example, the FCO’s investigation into Sennheiser’s selective distribution system resulted in the removal of a prohibition on sales on third-party platforms.  In that case Amazon itself was already an authorised distributor but the ruling opened up the possibility of sales being made over third party platforms, for example Amazon Marketplace.  In its investigations into Adidas and ASICS the FCO also found that general prohibitions on sales via third party platforms in selective distribution systems restrict intra-brand competition, and harmed small and medium-sized distributors. 

Conclusion

Brand owners and online retailers will be watching this case with interest as the CJEU will clarify a central question of whether it can ever be acceptable under competition law to restrict internet sales in order to protect brand image.




Competition and Regulation in Digital Markets

On Friday 9 September 2016, I attended the International Conference on Competition and Regulation in Digital Markets, a smoothly-run conference hosted by the University of Leeds School of Law. It featured a host of well-informed speakers from a variety of backgrounds. 

The keynote speech was given by Dr Andrea Coscelli acting Chief Executive of the CMA (the full text of which is available here). He flagged the CMA’s interest in digital tools, both as an aid to the investigative processes and also as part of any remedial actions, referencing the remedies proposed following the recent CMA investigations into the UK energy and retail banking markets. 

The speakers raised a number of interesting questions faced by courts and regulators in the fast-moving world of digital technology.  For example:

  • How can a static market definition be successfully applied to markets filled with constantly changing products?
  • How can regulators know when to act in markets that are still developing so quickly?
  • Is there a real risk that dominant digital companies will be discouraged from innovating further due to antitrust fears? 
  • What will the impact be of automatic algorithms capable of doing shopping on behalf of consumers?

Another feature of the event was the frequent differences between the perspectives of the economists and lawyers present. For example, there was some debate about the relevance of price substitution to market definition (the conclusion was that the focus should be on competitive restraints) and the extent to which vertical restraints are capable of restricting competition.

A development that nicely encapsulates some of the issues identified above is the Commission’s recently proposed overhaul of telecoms regulations (which we reported on here).

Some of the new rules apply to web communications services such as WhatsApp and Skype, thus extending the regulatory burden beyond its previous parameters.  One of the speakers at the conference, James Waterworth, pointed out that services like WhatsApp aren’t limited by the bottleneck of requiring a physical telephone line in the same way that traditional telecoms providers are. They operate in a completely different way, so why should they be regulated in the same way? 

After all, what happens if the services develop in the same way as the instant messaging service WeChat has in China, by including additional functionality such as video games and the ability to transfer money to other users within the app? Is that still a communications service that should be regulated in the same way as a company providing a landline?  Or is it a banking service that will fall within financial regulation?  Is adding layers of regulation consistent with the Commission’s exhortations to EU Member States to “aim to relieve operators from unnecessary regulatory burden, regardless of the business model adopted”? (See the Communication on the “collaborative economy” published in June 2016.)

There are many questions here, and not so many answers (yet…).  We’ll be keeping a close eye on how developments in this area play out.

Matthew Hunt

Connecting the DSM dots

European Commission releases proposals on copyright and telecoms as part of Digital Single Market strategy

Today the European Commission released proposed directives, regulations and initiatives on telecoms and copyright, aimed at modernising EU law in these areas. This is part of the Commission’s push to deliver on the Digital Single Market strategy.  

The Proposed Directive establishing the European Electronic Communications Code will be of particular interest to consumers and telecoms providers alike. In a bid to ensure that telecoms legislation is in line with modern technology, it extends regulation beyond traditional service providers. In certain areas, the rules will also apply to those providing consumers with “number-independent interpersonal communications services”, which encompasses newer chat and calling apps such as Facebook Messenger and Skype. 

In particular, Article 40 imposes an obligation on Member States to ensure that providers’ networks are secure and that they have appropriate systems in place to manage network security risks. The aim is to allow consumers to benefit from consistent and secure provision of services. This will undoubtedly have an impact on the way in which owners of chat and calling apps manage their network security. 

Another interesting development lies in the area of interconnectivity and interoperability of services. Article 59 grants national authorities the power to require service providers to have interoperable services. It applies in justified circumstances; the proposed legislation outlines the need for this where users require access to emergency services or “end-to-end connectivity between end-users is endangered”. Crucially, this will also apply to chat and calling apps.

For consumers, the proposed changes will be largely beneficial. The focus on network security is reassuring and the provisions on interoperability should improve the quality of services generally. However, these changes may come at a cost. Providers of chat and calling apps could choose to charge for apps which are currently free, in an attempt to offset the costs of compliance with the new regime. 

For those in the telecoms industry, this is perhaps a welcome development. Telecoms has traditionally been heavily regulated and increased competition from online communications service providers has been fierce, particularly as these services have not been subject to the same stringent regulations. The Commission’s move to legislate in this area is a step towards levelling the playing field. The questions are whether the proposals are enough to encourage telecoms companies to continue investing across Europe without inhibiting investment by over-the-top service providers that will now be subject to regulatory limitations. We will have to wait and see. 

Breaking news – General Court confirms Lundbeck decision and fine

The General Court of the EU has upheld the European Commission decision fining Lundbeck and a number of generic companies in relation to patent settlement agreements.  At the time of writing, the full text of the decisions has not been published.

What we do know

  • The Commission’s Lundbeck decision found a restriction of competition by object only.  The recent trend towards a more restrictive interpretation of the ‘object’ category (which we discussed in the context of patent settlements here) has not prevented this novel finding being upheld by the General Court.
  • Would-be generic entrants are therefore held to be potential competitors of the patentee (Lundbeck), despite the existence of patent protection held by Lundbeck.  The fact that they had possibilities for entering the market, including through an at-risk launch, is regarded as a form of potential competition.  (Having been brought up to respect the blocking power of patents, this is something I expect to find troubling for some time to come…)
  • The fine has been upheld in full – no credit has been given for the novelty of the decision.
It is still unclear how closely the General Court has followed the Commission’s reasoning – to judge by the press release, it appears likely that the legal analysis is closely aligned.  (See our discussion of the decision itself here.)  Other than for the parties themselves, the judgment will be of immediate interest for the parties to the Paroxetine litigation in the UK: as reported here, an appeal of the CMA’s decision in that matter is due to take place before the Competition Appeal Tribunal early next year.  This decision is likely to be welcome news to the CMA… Meanwhile, companies entering into agreements settling patent litigation will need to continue to pay very careful heed to the competition rules when deciding on the terms of market access for generic products.

The General Court’s press release is available here.

Sophie Lawrance

Proposed changes to the US Antitrust Guidelines for the Licensing of Intellectual Property – a transatlantic perspective

Just over a week ago, the US Federal Trade Commission and the Department of Justice’s Antitrust Division published a proposed update to the current US Antitrust Guidelines for the Licensing of Intellectual Property (the IP Licensing Guidelines). Issued in 1995, the IP Licensing Guidelines are the US equivalent of the European Commission’s Technology Transfer Guidelines, and set out the US agencies’ antitrust enforcement policy on the licensing of patents, copyright, trade secrets and know-how.  Whereas the Technology Transfer Guidelines were introduced in 2004 and updated in 2014 (our contemporaneous comments on the changes can be found here), the IP Licensing Guidelines have been in effect for rather longer.  The FTC and DOJ have asked for comments on the proposed update from interested parties – including lawyers, economists, consumer groups and the business community – by 26 September 2016.

According to this FTC press release, the US agencies’ aim is to “modernize the IP Licensing Guidelines without changing the agencies’ enforcement approach with respect to intellectual property licensing…”.  The proposed update thus retains the basic analytical framework of the 1995 guidelines.  As FTC Chairwoman Edith Ramirez puts it, the updated guidelines “reaffirm our view that US antitrust law leaves licensing decisions to IP owners, licensees, private negotiations and market forces unless there is evidence that the arrangement likely harms competition”.

The proposed revisions do, however, take into account developments in US case law that have occurred in the last 20 years or so.  One of the most significant such developments was the US Supreme Court’s decision in Leegin Creative Leather Products v PSKS (2007) that resale price maintenance (RPM) agreements should be assessed under the rule of reason, rather than being treated as per se illegal.  The proposed amendments to the guidelines reflect this change in thinking.  Although the Leegin case arose in a sale of goods context, the US agencies take the view that the Supreme Court’s analysis of vertical pricing restrictions applies equally to price maintenance in IP licences.  The rule-of-reason treatment of RPM by the US courts and agencies stands in contrast to the stricter EU approach: RPM is treated as a ‘hardcore restriction’ in the Commission’s Technology Transfer Block Exemption.  Another area in which the US and EU agencies take divergent approaches is in relation to the treatment of obligations to pay royalties after the expiry of the licensed IP right. The revised US guidelines refer to the recent Supreme Court case of Kimble v Marvel (2015), in which it was confirmed that post-expiry patent royalties are unenforceable.  In contrast, the Commission’s Technology Transfer Guidelines state (at paragraph 187) that “the parties can normally agree to extend royalty obligations beyond the period of validity of the licensed intellectual property rights without falling foul of Article 101(1) of the Treaty”.

Yet in other respects, US and EU attitudes to the relationship between IP licensing and competition law are very similar.  Both view IP licensing as typically pro-competitive and good for innovation.  The draft revised text of section 2.3 of the US guidelines states: “Licensing can allow an innovator to capture returns from its investment […] through royalty payments from those that practice its invention, thus providing an incentive to invest in innovative efforts”.  This echoes the view expressed by the Commission in paragraph 9 of its Technology Transfer Guidelines: “[…] licensing as such is pro-competitive as it leads to dissemination of technology and promotes innovation by the licensor and licensee(s)”.  In addition, the function of the antitrust “safety zone” outlined in section 4.3 of the US guidelines is akin to that of the “safe harbour” established by the Commission’s Technology Transfer Block Exemption: both seek to provide a degree of certainty for licensors and licensees alike and use market share thresholds.

In the final analysis, the proposed revisions to the US guidelines are arguably most striking for what they omit to deal with.  They do not address, for example, any issues relating to the licensing of standard essential patents (SEPs) or (other than in a single, brief paragraph) settlement agreements.  This is perhaps surprising given that both the FTC and the DOJ have paid considerable attention to such issues in recent years.  While the Commission’s Technology Transfer Guidelines also do not address these subjects exhaustively, they are in this respect arguably more helpful to practitioners and businesses than the US equivalent.  Where licensing agreements relate to both EU and US markets, however, it remains important for those involved to have regard to both sets of guidance.

Online poster seller in the frame for algorithm-centred price-fixing cartel

A few days ago, the CMA issued a decision confirming that two Amazon Marketplace vendors had fixed prices by using and configuring “commercially-available automated repricing software”.  The vendors, Trod Limited and GB eye Limited, which sell licensed sport and entertainment merchandise adorned with images of popular stars like Justin Bieber and 1D, colluded to offer online shoppers the same prices for products and co-ordinate price changes. 

Although the CMA accepts that online pricing tools can “help sellers compete better, for the benefit of consumers”, thereby benefiting competition, in this instance, the parties applied these tools to illegally fix prices at an artificially high level.  The CMA fined Trod £163,371, reflecting a 20% discount for Trod’s admission of liability and co-operation during the investigation.  GB eye, however, obtained 100% immunity for ‘whistle-blowing’, reflecting the CMA’s continued support for companies who come forward first.  

The CMA opened its investigation last year following a dawn raid at Trod’s premises and the home of one of its directors.  The CMA’s searches were co-ordinated with searches carried out by the British police on behalf of the US Department of Justice who were investigating the same conduct for sales through Amazon’s US Marketplace.  Following the dawn raid, the DoJ prosecuted both Trod and its director for price-fixing.  Trod accepted liability in the US as well, pleading guilty a few days ago (the DoJ has not yet provided details of any sanctions), but the director is still awaiting trial.  This provides an important reminder of the wider ramifications that anticompetitive behaviour can have, resulting in criminal sentences as well as civil fines and/or director disqualification.  

The CMA's investigation is yet another example of the authorities’ focus on digital markets, complementing the Commission’s e-commerce sector inquiry (see here and here).  The case follows in the footsteps of the CJEU’s Eturas decision which established the potential for liability for participants in a platform which fail to distance themselves from automated pricing updates (see here).  While the price-fixing agreement itself is hardly novel, the use of software to implement the agreement is more ‘innovative’ – and will doubtless not be the last such case to come before the competition authorities. 

This summer’s (not so) light reading – the CMA’s published Paroxetine decision (GSK/generics)

Some 6 months after issuing its infringement decision against GSK and a number of generic companies, the CMA has released a non-confidential version.  This comes in at a weighty 717 pages.  

Other than the grounds of appeal (on which we reported in the final paragraphs of this post), this is the first chance for companies and their advisors who weren’t involved in the proceedings to see the approach the CMA has taken, and to compare it with the current Commission approach.  First impressions are that the CMA has closely aligned itself with the Commission’s patent settlement decisions, such as Lundbeck**. The CMA and the parties will therefore be particularly keen to see the General Court’s forthcoming judgment in that case – indeed, the case management directions set down by the Competition Appeal Tribunal in the appeal proceedings against the CMA’s decision require the parties to prepare submissions on the relevance of the GC’s judgment to the case.

For those who aren’t keen on such weighty holiday reading, but can’t stand the suspense, below are a few pointers to the parts of the CMA’s legal reasoning which may be worth dipping into:

  • Paragraphs 1.3  1.20: A high level summary of the decision for those who only have an appetite for some light reading.
  • Paragraphs 3.65 – 3.84: The CMA’s view of patents, expanded upon at paragraphs 6.19-6.22.  The Windsurfing case law on the ‘public interest’ in removing ‘invalid patents’ is key: patents are treated as ‘probabilistic’ (although the term isn’t used) and are not guaranteed to be valid. Like the Commission, the CMA treats legal challenges to patent validity as part of the competitive process, and argues that the market is ‘in principle’ open to generic entry after expiry of patent protection over an API.  
  • Paragraphs 4.17 – 4.26: Overview of the market definition section which finds that, while other antidepressants may be substitutable for paroxetine, consumption patterns suggest that the actual competitive constraint is limited.  For market definition geeks, the full analysis is at paragraphs 4.29 – 4.97.  It is notable that paroxetine’s position within the ATC features only briefly, with the focus being on actual competitive constraints, including a ‘natural events’ analysis to look at the relative impact of generic entry in relation to the candidate competitor molecules (such as citalopram – the subject of the Lundbeck decision), and entry by generics of paroxetine itself (see para 4.73 in particular).
  • Once the narrow market definition is established, there isn’t much suspense as to the dénouement of the dominance ‘chapter’ (paragraphs 4.98 – 4.127).  In this context, the section on why the PPRS does not constrain pharmaceutical companies’ dominance is again unsurprising, but perhaps worth a read (paragraphs 4.124 – 4.126).
  • Paragraphs 6.1 – 6.9 and 6.204 – 6.206 contain a summary and the conclusion of the ‘object assessment’ under Article 101/Chapter I: while generally Lundbeck-esque, the reference to “the effective transfer from GSK [to GUK/Alpharma] of profit margins” strikes me as a novel way of expressing an old idea.
  • Paragraphs 7.1 – 7.3, 7.61 – 7.62 and 7.114 – 7.115 contain the summary and conclusions of the effects assessment under Article 101/Chapter I.  Even though the agreements were actually operated in the market, the CMA has confined itself to looking at their ‘likely’ effects – presumably to try to account for the fact that the outcome of the discontinued litigation is unknowable. It also concludes that the agreements assisted GSK to “preserve its market power” (paragraphs 7.63 – 7.64 and 7.116 – 7.117).
  • Leading on from that conclusion, paragraphs 8.1 – 8.3 summarise the case on abuse of a dominant position.  Central to the abuse case is the concept of inducement by GSK.  The allegations span not only the agreements in respect of which fines are issued under Article 101, but also an agreement with IVAX (for those with time on their hands, Annex M seeks to explain the discrepancy). GSK raised a number of objective justification arguments, notably around its right legitimately to defend its patent rights and to defend the company’s commercial position.  Paragraphs 8.61 – 8.67 reject these arguments, in particular on the basis that the conduct was not ‘competition on the merits’ (as per AstraZeneca) and that the conduct “went beyond the legitimate exercise of its patent rights to oppose alleged infringements”.  
  • Finally, and again for the more technically minded, at paragraphs 10.43 – 10.53, the relevance of the Vertical Agreements Block Exemption is dismissed, on the basis that the agreements were between potential competitors rather than being true ‘vertical’ arrangements.  At paragraphs 10.54 – 10.97, the parties’ Article 101(3) exemption arguments are also dismissed (spoiler alert: the exemption criteria are not found to have been fulfilled).  One curiosity is the lack of an infringement decision in relation to the agreement between GSK and IVAX.  This was held to benefit from the (now repealed) UK-specific Competition Act 1998 (Land and Vertical Agreements Exclusion) Order 2000 (now repealed).  In other words, that agreement is treated as vertical, unlike those between GSK and each of the other generic companies, even though the decision recites that IVAX did have plans to launch its own paroxetine generic.  The difference appears to be based on the context in which the agreements were reached: whereas the agreements with GUK and Alpharma related to the settlement (deferral) of litigation, that was not the case for the supply deal agreed with IVAX.  This is addressed at paragraphs 10.36 – 10.47 and in Annex M.
The paragraphs listed above focus on the legal analysis.  Those who prefer their reading less dry will want to look also at the descriptions of the agreements, and will note in particular that the ‘settlements’ considered in the decision did not finally resolve the litigation, but rather deferred it for the duration of the agreements entered into by GSK and the generic companies. Those who like tales of retribution will wish to read about the calculation of fines in section 11 – note that GSK received separate fines in relation to each of the agreements and the abuse of dominance.

The appeal hearing before the CAT is due to start next February, and to last for around a month.  By that time, the General Court will have issued its rulings in the various appeals against the Commission’s Lundbeck decision – which will doubtless be another weighty 
read for the Autumn.

** For more on Lundbeck, please see here (the abridged version) or here (the full analysis).