First published on the Global Competition Review website, 2 December 2016

Big Data can be described as large volumes of data, produced at high speeds from multiple sources, whose handling and analysis require new and more powerful processors and algorithms . We consider that Big Data can be dealt with as part of the usual competition law analysis and does not warrant special attention by way of new regulation. However, it is our view that competition authorities need to have greater understanding of Big Data and how it works in order to have the greatest effect in combatting anti-competitive conduct relating to Big Data.

Big Data has been considered in several areas of law, particularly data protection, consumer and privacy law, but as yet, competition authorities have not intervened. However numerous competition authorities around the world have begun studying Big Data to consider the possible role of competition law. This article considers recent studies by European National Competition Authorities (“NCAs”), sectoral regulators and the European Commission (“EC”).

Theories of harm


There are several theories of harm surrounding Big Data. In merger control, the EC has focused on whether the combined data set of the merged entity might be unique, and whether access to this data is in any way essential for competitors to be able to compete on an even playing field. Commissioner Vestager expressed her view that the standard for unique data was high and only data sets that could make it 'almost impossible for anyone else to keep up' will be problematic. The EC considered this in Google/DoubleClick and found that the combined data collections did not give the merged entity an advantage as competitors continued to have access to similar data. The data was not unique, so the merging entity was not found to have a competitive advantage.

In the recent merger of Facebook/WhatsApp , the EC investigated two further theories of harm, by which Facebook could strengthen its position in online advertising: (i) introducing advertising on WhatsApp, and/or (ii) using WhatsApp’s user data to improve the targeting of Facebook’s advertising activities . The EC found that there was no barrier to entry as 'there are currently a significant number of market participants that collect data alongside Facebook’ .

Network effects, which arise when a product becomes more valuable as more people use the product, were discussed in Facebook/WhatsApp. The EC found that they ‘may raise competition concerns in particular if they allow the merged entity to foreclose competitors and make it more difficult for competing providers to expand their customer base’ .

The Facebook/WhatsApp merger further attracted commentators’ attention as it did not need to be notified to the EC since WhatsApp's turnover did not meet the financial thresholds under the European Merger Regulation (“EUMR”). The case was voluntarily notified by the parties . This raised a question around whether the value of the parties’ data should be taken into account for the purposes of the EUMR, for instance giving the data a financial value such that an asset-based test could be used or alternatively introducing a ‘value of the deal’ test . This is currently the subject of a review by the EC of the EUMR.

A potential issue with a deal value test is that the value of the transaction could cause issues where the value might fall just above or below the threshold and parties seek to avoid notification by varying the price, or currency fluctuations mean that in one week the merger would need to be notified and the next, notification would be unnecessary. Further, the EUMR review is also looking at whether ‘industry-specific’ rules should apply alongside a transaction value test to ensure that the thresholds are not too broad in their application.

Currently, the EC is reviewing the merger of Microsoft/LinkedIn which was notified to the EC on 17 October 2016. The questionnaire sent out by the EC focuses on whether the data could be replicated from other sources . The parties submitted commitments to the EC on 23 November 2016, but these related to the bundling of products rather than access to data.

Abuse of Dominance

Another theory of harm surrounding Big Data is foreclosure of competition. This theory of harm is that dominant players with access to greater volumes of data or unique data sources may exclude competition by restricting access to this data.

If a particular data set is indispensable to other players and represents an ‘essential facility’ with no alternative products and technical, legal and economic obstacles that make it impossible or unreasonably difficult to compete in the downstream market, then licencing the data on fair, reasonable and non-discriminatory (“FRAND”) terms may remove the potential for foreclosure. This remedy could address this theory of harm.

The Groupe Spécial Mobile Association (“GSMA”), a global association representing the interests of mobile operators, has also noted the ‘potential for anticompetitive effects’ when gathering Big Data in its recent report on the Digital

Ecosystem . This report advocated that ‘data per se need not raise competition concerns’ and authorities should instead be focusing on what is done with that data. The GSMA recommended that authorities look at 'actual substitution patterns' in online markets, and suggested that NCAs need to understand the dynamic effects and efficiencies that Big Data can provide better, such that the consumer benefits are considered fully.

Anticompetitive Agreements

Another concern for the competition authorities regards the anticompetitive exchange of information between competitors through Big Data. Greater volumes of data and more transparency in a market mean more opportunities for tacit or explicit collusion. It is the content of the data exchanged which might concern the competition authorities most, for example, if the data contains pricing strategies, customer information or references to market sharing. This would be the same under standard competition analysis demonstrating that there is no need for additional regulation in this area.

This article was written by Sarah Turnbull, David Riley and Rachel Carr

First published on the Global Competition Review website, 2 December 2016

Authorities may also need to consider the possibility that computer programming could be used to take part in anticompetitive conduct, for example, using a common algorithm to fix prices. ‘The Commission has earmarked more than half a million euros to analyse companies’ use of algorithms’ demonstrating their desire to greater understand the collection of Big Data. In August 2016, the CMA issued a decision on this point, demonstrating that even complex algorithms can be dealt with using existing competition law rules .

Approach and developments at the European Member State level


In 2016, the Financial Conduct Authority (“FCA”) issued a feedback paper on Big Data relating to retail general insurance and noted that ‘Big Data does not appear to raise or create barriers to entry’ . Following the Call for Input, the FCA decided not to launch an in-depth market investigation on the basis that it agrees with the proposition that Big Data does not warrant any special attention. That said, the FCA will continue to monitor this area including working with the UK

Information Commissioner (responsible for data protection issues) to gain the greater understanding that we encourage.

The Competition & Markets Authority (“CMA”) produced a statement in which they considered competition law issues in Big Data. This report focused on whether the data in question is substitutable or ‘widely available for collection’ . If a data set is unique but it can be substituted by another data set or collected by another company, then this will not act as barrier to competition. The CMA uses this substitutability analysis for product-based competition law issues and this should not change for data issues. The scale of the data set may affect its substitutability, but that will be determined on a case-by-case basis.

The CMA recognised the importance of working ‘with other authorities to track new developments in the collection and use of consumer data and to ensure an integrated approach to enforcement and regulation’ .

The CMA’s acting Chief Executive Officer also considered whether the current rules work in relation to mergers. He noted that the UK’s share of supply test has allowed the CMA to review mergers in which a target company has no or low turnover . He recognised that Big Data may present some novel issues and that the CMA ‘might need to apply the rules in new contexts, but that doesn’t mean [the CMA] need a completely new set of rules’ .

This view was broadened to apply to more than just mergers in the CMA's response to the EC’s Preliminary Report on the E-commerce Sector Enquiry. In this report, the CMA's outlined its view that ‘the existing principles-based competition law framework is sufficient to tackle competition concerns’ in the digital economy . We agree with this view and the CMA’s statement that ‘nonetheless it is important for competition authorise to continue to seek to understand better the way in which competition operates in the online sphere’ .

France and Germany

The French and German NCAs issued a joint paper that focused on data as a source of market power, data reinforcing market transparency and data-related conducts that may cause competition concerns. Notably, the report took a different view to the EC on privacy policies, stating that they should be considered whenever they are liable to affect competition. Having highlighted their concerns in the joint report, the French NCA has since launched a sector enquiry into Big Data .


In addition to the joint report, the Bundeskartellamt (“BKA”) has considered how dominance should be considered for digital companies in a further paper published in June 2016 . It has stated that a review of market dominance should focus on ‘the existing innovation competition in the market and the potential competition from innovative businesses’. The BKA further notes that the following specific areas should be assessed alongside market dominance: direct and indirect network effects ; economies of scale that may result in barriers to entry; the prevailing types of use on the opposite market side (single-homing/multi-homing) and the degree of differentiation; the access to data, particularly considering the volumes and speeds at which data is now available; and the innovation potential in digital markets. Having considered all of these factors, the BKA concluded that ‘control of data is not per se an indication of market power... but this can play an important role' and, similarly to the EC, determined that each factor must be assessed on a case-by-case basis. Since issuing the report, the BKA has begun an investigation into Facebook to determine whether Facebook have abused their market power by infringing data protection rules . In order to find that there has been an infringement, the BKA are required to establish (i) Facebook’s dominance, (ii) an infringement of data privacy rules and (iii) a causal link between both. This investigation reinforces the view of the BKA that competition law should be dealing with breaches of privacy law.


The Dutch NCA began an investigation in September 2016 into how consumer data may give companies market power and to what extent online platforms may harm competition. It appears that the Dutch NCA is not currently looking at privacy concerns, but this has previously been brought to the attention of the Second Chamber by the Public Prosecutor in the context of Big Data . It is therefore possible that we could see privacy concerns surface in this investigation.

EC’s position and Data Protection developments

As part of the EC’s Digital Single Market initiative, the EC is aiming to reform EU data protection rules such that they are suitable for a digital age. The new legislation which will come into effect in May 2018 , will require businesses to provide consumers with more information on how data is handled, inform consumers about data breaches and have increased accountability when processing personal data.

Commissioner Vestager recently observed that ‘we don’t need to look to competition enforcement to fix privacy problems’ , but should privacy be a standalone objective of competition law? The German NCA’s view in its investigation into Facebook appears to think so. Case-handler at DG Competition Cyril Ritter recently considered data protection as an online service quality issue. Ritter considered a situation in which one website championing privacy merges with one which does not, then the ‘loss of privacy as an element of quality could be important in the future' . The CMA have said the ‘starting point is that businesses should compete over all issues... including privacy and data collection practices' .

The European Data Protection Supervisor (“EDPS”) recently issued an opinion , which included recommendations on updates to merger control rules. Similarly to Ritter, the EDPS discussed ‘privacy as a factor of quality’ such that where a consumer’s privacy is compromised, they suffer detriment. In addition, they focused on the 'significance of consumer welfare' in mergers of digital companies and suggest that the lack of monetisation of data renders the turnover thresholds inadequate. The EDPS further suggests that the EUMR should be interpreted such that it protects individuals’ rights to privacy and data protection . The EU’s Digital Single Market Strategy is considered the ‘right opportunity for the EU to work coherently towards these goals’ . The EDPS has also suggested a Digital Clearing House to encourage regulators to share best practices in this area. We encourage this approach as it is likely to improve understanding of Big Data amongst regulators and competition authorities.


We consider that work by the various NCAs and regulators to improve their knowledge of the area is key, as the ‘growth in the collection and use of data, and the complexity of data markets make the role of regulators increasingly challenging’ , and we encourage knowledge sharing between regulators as this area continues to develop. . It is our view that new regulation is not necessary as competition authorities already have the necessary tools to tackle anti-competitive conduct relating to online trading and Big Data.

However, there is a risk of Member States approaching this issue differently, particularly in relation to privacy concerns. Indeed Commissioner Vestager considers that 'there's a strong case for new EU rules as part of the answer’, and even suggests putting a proposal on the table early next year .