Michael Grenfell on antitrust in the digital age

Lecture given by Michael Grenfell, CMA Executive Director – Enforcement, at the Competition Law Forum.

Michael Grenfell photo


The digitalisation of our economy, of our society, of our daily lives is, as we all know, having a huge effect on all aspects of human endeavour. It’s surely not surprising, then, that this should include our own field of endeavour, competition or ‘antitrust’ law and policy.

At the organisation I work for, the Competition and Markets Authority (CMA), we take a keen interest in digital markets. There are various reasons for this.

First – but perhaps not foremost – is that this is what the government has asked us to do. In its strategic steer to us updated last December, the government told us that we should “continue to focus on developments in new emerging markets, such as online digital market places and use of data … [and] … develop expertise in the impact new emerging markets are having on competition, innovation and consumer choice.”

Rather more fundamentally – and in line with this – we care about digital markets:

  • Because the digital sector, in and of itself, is a significant, and growing, part of the country’s economy – in 2014, it accounted for £118 billion gross value added (GVA), which is 7% of total GVA across all economic activities in the UK (1).

  • Because digital services underpin pretty much every other part of the economy. I’ve been trying to think of an area of business activity which doesn’t in some way rely on digital services. It’s not easy. We all do, and usually we’re fairly heavily reliant on them.

And, turning to the particular perspective of most of us in this room, a modern competition agency such as the CMA is bound to be interested in digital sectors because they raise so many new, and difficult, antitrust issues. This is not to say that existing antitrust law is incapable of dealing with them – thankfully, our competition laws are based on principles that are not merely ephemeral, and can be applied to economies in various different stages of development. Rather, digitalisation raises questions about how we apply competition law to new situations. Let’s take some examples.

Market definition is a staple of competition analysis in a range of contexts: to assess dominance, to assess the impact of mergers, to judge whether an arrangement is de minimis, and so on. Digitalisation raises important questions, particularly in distribution and retail markets. Put simply: are online services in the same markets as traditional services? Should we regard online services as competitive constraints on more traditional services? And is it meaningful to see this in such ‘binary’ terms? After all, where does ‘click-and-collect’ purchasing fit in? Or e-readers sold in bookshops? Or browsing on price comparison websites but buying in-store? The reality is that consumers use a mix of traditional and digital channels even in the same single purchase.

A second example of a tricky issue thrown up by digitalisation arises from the fact that markets can be distorted when online services, facing lower costs, ‘free-ride’ on the investments of traditional services. How should we deal with that – by allowing restrictions on online services, or by allowing the distortions of such free-riding?

Thirdly, ownership of digital platforms gives market power which can be abused by discriminating against those dependent on the platforms – that applies whether we are talking about hard platforms (raising issues of access to electronic trading platforms, and access to operating systems such as the work-server interoperability issue in the EU Microsoft case) or ‘soft’ information bases (alleged discrimination against downstream competitors is at the heart of many of the current Google cases).

In short, the digital economy raises a number of problems for antitrust law and policy. And they’re not easy. But before we get too gloomy, it is important to remember how digitalisation can enhance competition and offer solutions that help markets work better, to the benefit of consumers. Just to take a couple of examples of this:

First, to make a point that is obvious, but nonetheless of critical importance. In many markets, digitalisation offers a cheaper, more innovative, alternative to existing products. That is precisely what competition law and policy have always sought to achieve: choice, innovation, downward pressure on price.

And second, and more specifically, we have in recent years seen how new digital comparison tools such as price comparison websites and apps, can stimulate competition – in our major recent market investigations into energy and retail banking, a key part of the remedies to the adverse effects on competition were digital: in the energy investigation price and service comparison websites, and in the retail banking investigation the proposed ‘open banking’ app interface to enable bank customers to find the best deal suitable for them. And this autumn we have launched a market study into the role digital comparison tools can play in fostering competition.

Issue 1 – collusion and cartel conduct

First, let’s consider the most basic – and serious – kind of antitrust infringement is collusion between competitors to fix prices or carve up markets; cartel conduct in other words.

It should come as no surprise that the temptation to cartelise exists in online selling as much as in traditional markets. And in case there was any doubt, our infringement decision concerning the online selling of posters and frames via Amazon Marketplace in August this year makes clear collusion in online selling will not be tolerated. We found that 2 sellers had agreed not to cut each other’s pricing for online selling, and we imposed a fine. I should emphasise that Amazon was in no way a party to the infringement. We have followed our decision by launching, on 7 November, a programme of competition law compliance advice to online sellers, with online marketplace providers helping to disseminate our advice.

The case itself had another digital aspect: it was a digital device – automated repricing software – which was the tool for price coordination.

In that case, there was clear agreement between the parties to facilitate price coordination, with communications between them that they aimed not to undercut each other’s prices. But there may be cases in which, as Professor Ariel Ezrachi of Oxford University has pointed out, computing devices automatically and autonomously collude – because they “determine independently the means to optimise profit” (2), ie independently of their human programmers – and this can raise questions about key concepts relevant to the antitrust notion of an ‘agreement’ or a ‘concerted practice’, such as intentionality or a ‘meeting of minds’ or knowledge.

Issue 2 – resale price maintenance

Second, there is the question of resale price maintenance – that is, the practice where an ‘upstream’ supplier limits the freedom of downstream resellers, such as distributors or resellers, to discount. As a matter of EU law – and hence, under present arrangements, UK law – resale price maintenance is presumed to restrict competition and to infringe the prohibition on anti-competitive agreements (3). Resale price maintenance is also presumed unlikely to merit exemption from the prohibition (4).

But there are many people who argue that resale price maintenance may, in certain circumstances, be justified by efficiencies: for example, by giving retailers an incentive to make investments which increase the demand for the product.

Against this background, after the CMA’s decisions on 2 resale price maintenance cases in our first 18 months – sports bras in June 2014, where we decided that there were no grounds for action, and hotel online bookings, where we closed our investigation in September 2015 on grounds of administrative priorities – some commentators may have concluded that we are no longer interested in resale price maintenance.

Recent developments should have dispelled this notion. We continue to be interested in resale price maintenance. We will assess each case on its merits, but some of our recent cases make clear that resale price maintenance will not be allowed online any more than it is in retail shops – in particular our 2 infringement decisions in May this year in the bathroom fittings case, involving enforced online ‘recommended’ retail prices, and in the commercial catering equipment case, involving minimum advertised internet prices.

And we are concerned about other means in which suppliers control resale pricing, including in online outlets, such as ‘most favoured nation’ or ‘price parity’ provisions, which are the subject of our ongoing monitoring in relation to online travel agents and hotel bookings.

Issue 3 – online retailers competing with ‘bricks-and-mortar’ shops

Third, there is an aspect of online competition which raises perhaps the most challenging issues for competition decision-makers, but which is a hugely important implication of the way our economies are developing. This is the way that online retailers increasingly compete with ‘bricks-and-mortar’ shops.

One obvious issue raised by this is market definition. Should online retailers be regarded as a competitive constraint on bricks-and-mortar shops, or as being in a separate market? That question may depend whether we block or clear a merger, whether we find dominance for the purpose of the prohibition on abuse of a dominant position, whether we see agreements as appreciably restrictive of competition.

There is no single answer. As with all questions of market definition, it depends on the actual economic circumstances of the market at any one time. And the ‘at any one time’ qualification is important: these markets are fast-developing, and whereas in one year a critical mass of consumers might not consider buying online as a substitute for going shopping in a real physical store, that can change within just a few years (and it can be a very few years).

So we, and other competition authorities, will answer this question without dogma or preconceptions, according to the actual evidence, case by case, and from time to time. For example, in recent UK merger cases, we have found that:

  • for betting shops, in 2016, the constraint from the online channel on the retail channel is not sufficiently strong for these channels to form part of the same relevant product market (5); but

  • for video games sales, in 2008 bricks-and-mortar retailers and internet selling did not constitute separate markets (6).

And, under the prohibitions, in our recent decision finding an anti-competitive agreement in advertising estate agents’ services, we concluded that traditional bricks-and-mortar estate agents were a separate market and that online estate agents exercised “only relatively weak competitive pressure” (7).

But, more important than market definition, is the substance of the matter. What does online retailing do for competition – and for the benefits that competition is supposed to yield: choice, innovation, better value for money?

Most obviously – and importantly – online retailing offers real competitive choice. If we want to buy a book, or a radio, or a watch; if we want to place a bet on the horses; if we want to book a package holiday – all normal things we were doing before the internet came along – we no longer have to go to the high street or the shopping mall. We have a choice: we can go out shopping, but alternatively in the comfort of our homes we can buy, and we can shop around. So our choices are wider. And because there is more potential competitive choice, there are greater competitive pressures on price, quality, efficiency, innovation. More than that: online selling – because it does not require physical stores – is typically lower cost, which augments the downward pressure on price and the spur to efficiency. Of course, if you just buy online, you can’t so easily browse or ‘feel the quality’, and you can’t ask a nice shop assistant to advise you before you make your purchase. And that’s the beauty of choice and competition: you can choose the low-cost option online, or the ‘full-service’ option in-store.

But – and there is a ‘but’ – it’s not clear that this situation is sustainable, and there are risks that this new entry by online retailing takes away some of the choices we value. This arises from the ‘free-riding’ problem, or I should say the alleged free-riding problem, because many commentators think it’s overstated.

We’re all familiar with the phenomenon of browsing in a bookshop, finding a book that appeals (perhaps after taking advice from the shop assistant), and then going out and buying it on Amazon for a lower price. Back in 2009 there was that famous, and controversial, advert for the online electronics reseller Dixons.co.uk that appeared on railway platforms across Britain:

Step into middle England’s best loved department store … where an awfully well brought up young man will bend over backwards to find the right TV for you. Then go to Dixons.co.uk – the last place you want to go.

This is what people have in mind when they talk about ‘free-riding’: the online retailer – be it Dixons.co.uk or Amazon or whoever – gets the sale, because it has lower costs as a result of not having to operate physical retail premises and employ shop assistants who give advice etc. But the consumer made the purchase only after having browsed in the bricks-and-mortar store and received the shop assistant’s advice – so the online retailer has ‘free-ridden’ on the physical retailer’s investment in premises and staff, while the physical retailer hasn’t reaped the rewards of its investment because it has lost the sale to the online seller.

A survey conducted in the course of the European Commission’s current competition sector inquiry into e-commerce found that “45% of manufacturers indicate that free-riding, whereby customers benefit from services offered by brick and mortar shops to make their choice, but then purchase the product online, is common or very common” (8).

That is clearly not sustainable. If this kind of free-riding persists, competition and consumers lose out – it becomes loss-making to provide service in stores such as the shop assistant’s advice or indeed the facility to browse, stores then go out of business – we’ve all heard about the decline of the high street – and the bricks-and-mortar stores, and the services which were provided to consumers and valued by consumers, are lost. Consumer choice, and the quality of the retail experience, suffer. That is not what competition policy seeks to achieve.

In all this, there is a dilemma for competition authorities.

On the one hand, we should facilitate competitive pressure from online sales by taking action against restrictions on online selling that protect traditional bricks-and-mortar sales channels.

But on the other hand, as we’ve seen, free-riding by online sellers on the investment made by bricks-and-mortar stores can lead to loss of consumer choice and quality. That said, competition authorities and policymakers need to take an evidence-based approach to that assertion, as to everything else we do; in any situation, we can’t just accept such claims, and allow restrictions on competition, without real evidence that this is likely to happen.

In any event, as well as the ‘free-riding’ argument, there is an argument that, unless the supplier has market power, restrictions on online selling of the same products – ie restrictions on intra-brand competition – do not matter if there is sufficient inter-brand competition (which of course won’t always be the case).

EU competition law, which has been the basis of our own competition law for many years now, has dealt with this dilemma with an uneasy compromise, reflected in the EU’s vertical agreements ‘block exemption’ regulation and guidelines (9). The EU’s Competition Commissioner, Margarethe Vestager, recognised this in a speech on publication of the preliminary report of the Commission’s sector inquiry into e-commerce, acknowledging that “pricing restrictions can help to stop physical shops from disappearing” but insisting that it’s important that “consumers’ choice isn’t unnecessarily limited” (10).

Under this compromise in EU law, in the context of agreements between suppliers and resellers, many arrangements that discriminate against online selling are strictly prohibited as being ‘hardcore’ restrictions. These include obligations on a reseller or distributor:

  • to prevent customers in other territories viewing its website
  • to terminate consumer transactions on the internet once credit card data show addresses outside the distributor’s exclusive territory
  • to make no online sales – this is what people call an ‘online sales ban’, which I’ll talk about in a moment
  • to limit the proportions of their sales that they make online
  • to pay to the upstream supplier a higher price for products being resold online than for products being resold in bricks-and-mortar

But, on the other hand, EU law provides that it is permissible for a supplier:

  • to limit a reseller distributor from online advertising that is specifically addressed to customers in other distributors’ exclusive territories
  • to oblige a reseller to sell a minimum absolute amount (whether by value or volume) in bricks-and-mortar stores
  • to oblige a reseller to observe certain quality standards for resale over the internet – for example, that the reseller must have one or more bricks-and-mortar shops or showrooms as a condition for being a member of selective distribution system

I have said that online sales bans are strictly prohibited. These are provisions that prevent distributors from selling online at all, or place restrictions which have that effect. In June this year we at the CMA issued a Statement of Objections in a case setting out our provisional finding that a provision by a supplier of golf equipment preventing its retailers from selling its golf clubs online is an infringement of the prohibition on anti-competitive agreements. We will naturally consider all representations before reaching a final decision on this.

These kinds of ‘online sales bans’ are to be contrasted with ‘online platform bans’, which are provisions that prevent distributors from selling through certain specified platforms (Amazon Marketplace, for example, or eBay). The European Commission and various national competition authorities see such online platform bans just as a restriction on how a product is sold, but without closing off an outlet – and therefore as not being an ‘object’ restriction, but rather one to be assessed case by case according to its effects. And the European Commission has recently said that it doesn’t regard them as ‘hardcore’ restrictions which lose the benefit of the block exemption (11). By contrast, Germany’s competition authority, the Bundeskartellamt, has viewed an online platform ban as likely to be an ‘object’ restriction in the running shoes case (12), but German courts have been uncertain on the point; it should be resolved in the context of a case called Coty, concerning the sale of cosmetics on Amazon, which has been the subject of a reference in April this year from a German court to the EU Court of Justice.

How then to solve the free-riding dilemma? For businesses, there are alternative means to influence resale channels that comply more easily with competition law, such as through exclusive or selective distribution arrangements, in line with established competition law. In reality, major bricks-and-mortar retailers have mitigated the risk of online competitors free-riding on their investment by commercial strategies of themselves entering the digital market but with multi-channel propositions where consumers use both – like ‘click-and-collect’. Which I think counts as an example of competition working.

Developing law

Consideration of some of these issues is in flux with lively debate across the competition community worldwide.

The CMA is of course largely bound by EU precedent – by virtue of EU Regulation 1/2003 article 3, and, in our own domestic legislation, section 60 of the Competition Act 1998.

But, depending on the eventual arrangements on our exit from the EU, there might in future be more scope for the CMA to develop policy and thinking itself, reflecting current thinking – although we need always to recognise that there are real advantages, all other things being equal, in international consistency between competition laws.

The problem of speed

It is a commonplace that digital markets are fast-moving.

Products and services that were once unthinkable become the norm. And previously staple products become obsolete. Who now can say ‘the cheque’s been lost in the post’?

The same is true of competitive structures: only a few years ago, the technological giants were IBM, Microsoft, Yahoo. It’s not only that Google has largely eclipsed Yahoo, but that we have new companies offering new services – such as Amazon, Facebook, Twitter. Who now remembers Friends Reunited?

The implications for competition policy of fast-moving markets are hard to gauge.

On the one hand, they might suggest that competition authorities should be hesitant about intervening in digital markets:

  • Apparently dominant players are quickly superseded in ever-changing dynamic markets.

  • In any case, there is the reality that competition processes are necessarily painstaking and time-consuming – involving detailed reviews of voluminous evidence, protecting the procedural rights of all parties through access to the file, respect for business secrets, and so on. As a result, by the time competition authorities intervene to address the problem, it might well be too late. The markets concerned will have moved on.

But, on the other hand, a consequence of digital markets being fast-moving is that anti-competitive practices can create damaging barriers to innovation that would otherwise flourish – repressing and retarding the dynamism that can hugely benefit consumers.

This may occur, for example, where an incumbent refuses to grant potential new entrants access to ‘essential’ platforms, or where there has been discrimination by operators of search engines against competitors in respect of the prominence of displays on the search page (an issue that has arisen in various investigations into Google), or where the arrangements underlying competition price comparison websites and other digital comparison tools limit competition either between suppliers or between the comparison tools themselves. That means that, unless competition authorities act quickly to prevent such practices, we deny consumers the potential benefits of these new and flourishing markets and are left with players with significant market power – and often also power to control information, and, indeed, the way we live (many people say that about Google and Facebook, for example, and one can see the point).

There is no easy solution. Direct ‘ex ante’ regulation is often too blunt a tool, failing to anticipate and keep up with developments in fast-moving markets. Competition law measures – ex post – have the risks that I’ve identified. But where we at the CMA become aware of practices and arrangements that risk stifling new entry and innovation in these all-important sectors, and we think that there is scope for remedying them, then we will do so. I have talked about some of the measures we have taken so far – and, where speed is critical, it is open to us to use our interim measures powers, and we will do so where appropriate.

Other competition law tools

I have focused so far on ‘antitrust’ prohibitions on anti-competitive agreements and abuse of dominance, in our UK Competition Act 1998 and the EU equivalents.

But at the CMA we also have other tools to help ensure that our society enjoys the full competition and consumer benefits of digital markets – including our functions to conduct market studies and investigations, our merger control role of scrutinising mergers and acquisitions for their effects on competition, our powers to enforce consumer protection laws, and what we call our ‘advocacy’ role in public debate where competition and consumer issues are at stake.

I will say something about each of these in turn.

Market studies and investigations

Digitalisation was key to our analysis and remedies in the 2 major market investigations the CMA has conducted – both into significant, but traditional and non-digital, industries: energy and retail banking.

Energy market investigation

We launched our energy market investigation in the earliest days of the CMA acquiring its functions, and issued our final report and conclusions in June this year.

We regarded this work as central to our mission to promote competition for the protection of consumers. In the poorest households in the country, household spending on gas and electricity represents 10% of overall spending. And in the decade to 2014, domestic energy prices rose 75% in real terms.

Our investigation found low levels of ‘shopping around’ between energy suppliers. If consumers don’t shop around much, that weakens competitive pressures, to the detriment of consumers.

Digitalisation was highly relevant to the investigation’s analysis: for example, the planned roll-out of smart meters by 2020 can improve customer engagement, making people more aware of their energy costs and so perhaps more willing to shop around.

Digitalisation was also essential to the remedies – ie the measures that the investigation concluded were necessary to remedy the adverse effects on competition that had been found in the way the market operates. These remedies included: the establishment of a database of customers who have been on the standard tariff for at least 3 years, accessible to competing suppliers, coupled with an enhanced role for ‘brokers’ in the energy market, removing restrictions on their access to data.

Retail banking market investigation

In August, we published the final report and conclusions of our other major market investigation – into the supply of retail banking services to personal customers and to small and medium-sized enterprises.

This investigation also found the market to be characterised by low levels of shopping around – in any year, only 3% of personal customers and 4% of business customers switch their bank.

As with energy, so with retail banking: digitalisation was highly relevant both to the analysis and the remedies proposed in the investigation.

For example, in respect of the analysis, careful consideration was given to the extent to which digitalised bank services – online and mobile banking – reduces the need for competitors to have extensive bricks-and-mortar branch networks, which had traditionally been viewed as a significant barrier to entry and expansion. As for remedies, digital services were at their core, by way of the ‘open banking’ remedy: an ‘application programming interface’ standard through which banks are to be required to allow their customers to share their own bank data securely with competing banks using an open banking standard - to help customers find better-value services, and to enable new entrants and small providers to compete on a more level playing field. The idea is that, with open banking, apps can provide customers, on the basis of transactional information, with details of the current account that suits them best while authorised brokers and price comparison websites will have access to such information through the open banking interface.

Digital comparison tools market study

Price comparison websites – which are perhaps better described as ‘digital comparison tools’ (DCTs) because they relate to service quality as well as price – were also key to remedies proposed in the CMA’s market investigation report on payday loans, published in February 2015. Our current legal services market study is considering increased use of digital comparison tools to enable customers to compare offerings by providers of legal services.

Such tools are of potentially huge benefit to competition and to consumers. They make it easier for people to shop around between suppliers of goods and services, and to make informed choices. Because of this, suppliers need to be ever more competitive on price and quality if they are to thrive. In short, they sharpen competitive pressures, which is normally very much in the interest of consumers.

But these digital solutions to encourage switching can create problems. By way of example, concerns about ‘most favoured nation’ provisions, under which a supplier agrees with a price comparison website that it will not make its products available more cheaply on other online platforms, were a feature of our market investigation into private motor insurance on which we published our report in March 2015, and of our work, and the work of other competition authorities, concerning hotel online booking.

In September, we launched a wholly new market study which is entirely focused on digital comparison tools including price comparison websites. This recognises the huge benefits that digital comparison tools can bring for the competitive process and for consumers, but will examine concerns that have been expressed that they could work even more effectively, including (among other issues) about whether consumers would benefit from being made more aware of how DCTs earn money, and about whether arrangements between DCTs and the suppliers that sell through them might restrict competition.

Merger control

We have already seen how, in assessing whether a merger might substantially lessen competition, a key factor in many sectors is the extent to which digital services act as a competitive constraint on traditional services.

But I also want to mention another issue, which arises from the way that digital companies are so fast-growing. The issue is whether merger control regimes are able to capture many potentially anti-competitive mergers and acquisitions in digital sectors.

This arises particularly in merger control regimes which rely purely on turnover-based jurisdictional thresholds – which is not true of the UK domestic merger control regime, but is true for those mergers affecting UK markets that are subject to the exclusive competence of the European Commission under the EU Merger Regulation (currently at least).

The risk is where major tech companies, like Google or Facebook, acquire innovative, emerging companies with currently quite low turnover but which are the in the midst of, or have the potential for, rapid growth. In those circumstances, the acquisition may not come within the jurisdiction of the merger control regime – because the target’s turnover is too low – but it is potentially a matter of concern that it should escape merger control scrutiny where, as often happens in digital sectors, the target is liable to grow quickly and the acquisition thus stymies the competitive challenge from an independent new entrant.

In October the European Commission began a public consultation on whether to address this by revising its jurisdictional thresholds for digital mergers, so as to involve a criterion based on transaction value (13).

Consumer protection

The enforcement of consumer protection laws is one aspect of the CMA’s work (and indeed of my portfolio) that is ‘off the radar screen’ for most competition lawyers! (I know – I was one.)

The CMA is just one of a number of consumer enforcement agencies in the UK under the new ‘consumer landscape’ in place since 2013/14, along with local authorities’ trading standards services, the National Trading Standards Board, Citizens Advice, and so on.

In this new landscape, the remit of the CMA, in particular, is to use consumer protection to support competition. The government, when setting out its plans for the new landscape, said that the CMA should “retain consumer enforcement powers as remedies for use in markets where competition is not working appropriately due to practices and market conditions that make it difficult for consumers to exercise choice” (14).

A key aspect of this has been our work on online reviews and endorsements, in which we have secured public commitments from providers to prevent online reviews being misleading – eg through (i) fake positive reviews, (ii) suppression of negative or critical reviews, or (iii) undisclosed paid-for advertising that is indistinguishable from genuine reviews or endorsements.

Online reviews are a fantastic tool for consumers to exercise informed choice, and therefore sharpen and stimulate competitive pressures.

But that only works if they are honest, and consumers can rely on them.

Our work was designed to ensure this by rooting out – and deterring – dishonest and misleading practices.

Our work was critical to the effectiveness of this major digital stimulus to competition and to markets working better. And much of our other recent consumer protection work has been in digital markets: children’s online games, cloud storage, online car hire bookings, online purchases of tickets for events, online gambling.

Advocacy in public debate

A classic example of digital technologies challenging traditional service suppliers is in ‘private hire vehicle’ services – what you and I know as taxi and minicab services.

Companies such as Uber relying on mobile apps to serve customers obviously present a competitive challenge to traditional regulated taxi services.

A lack of regulation potentially has adverse effects for consumer protection, eg passenger safety.

Public authorities worldwide have been uncertain how to draw the balance.

In late 2015, Transport for London (TfL) consulted on proposed new regulations designed to protect consumer interests from exploitation, but which would also have made much harder new entrance like Uber to compete.

We very publicly responded in December 2015 saying:

We recognise that private hire vehicle passengers need the protection of appropriate regulation. But consumers also benefit from effective competition exerting downward pressure on prices and upward pressure on service quality and standards. Clearly there is a balance to be struck, but the general principle is usually that, in the interest of consumers, competition should only be compromised or restricted by regulatory rules to the extent that doing so is necessary for consumer protection. We are concerned that some of the proposals on which TfL is consulting would go significantly beyond this, it would impose regulation that excessively and unnecessarily weakens competition, to the overall detriment of users of taxi and private hire services in London.

TfL significantly scaled back its proposals – leaving London as one of the world’s major capital cities most open to digital competition in taxi and minicab services.

Similar challenges are arising in other cities, and we are responding appropriately, most recently, in September, in relation to the city of Sheffield.

Some conclusions

Digitalisation is here, whether we like it or not.

We do like it, on the whole. Digitalisation can bring huge benefits, including in offering greater choice, innovation, spurs to efficiency and downward pressure on price – exactly the kinds of objective that competition policy seeks.

It is growing, and no antitrust authority can ignore it. Indeed we use digital tools in our own work – in evidence and intelligence gathering; in monitoring social media, for example to hear about consumer experiences of an issue that we have identified as being of potential concern; and in communicating our work through social media, complementing more traditional channels such as press releases.

Digitalisation presents real challenges – eg on market definition – and real potential problems, about the balance between allowing online competition and avoiding free-riding, operators of ‘essential’ digital platforms abusing market power to squeeze out competition, etc.

But it also offers real opportunities – challenging existing structures, sharpening competitive pressures and offering new increased and innovative choice.

The opportunities are there for the taking. The challenge for competition authorities is to help society make the most of them.


  1. House of Commons Library Briefing Paper CBP 7610, Digital economy: statistics and policy, 2 June 2016, p3.
  2. Ariel Ezrachi and Maurice E. Stucke, ‘Artificial intelligence and collusion: when computers inhibit competition’, The University of Oxford Centre for Competition Law and Policy Working Paper CCLP (L) 40, April 2016.
  3. European Commission Notice – Guidelines on vertical restraints (2010/C130/01), May 2010, paragraph 2.
  4. EU guidelines on vertical agreements paragraph 3.
  5. CMA, A report on the anticipated merger between Ladbrokes plc and certain businesses of Gala Coral Group Limited, 26 July 2016.
  6. Competition Commission, A report on the completed acquisition by Game Group PLC of Games Station Limited, January 2008.
  7. Three Counties, Case CE/9827/13, May 2015, Annex C, paragraph 23.
  8. European Commission staff working document, ‘Preliminary report on the e-commerce sector inquiry’, 15 September 2016, paragraph 286.
  9. Paragraphs 52 to 54 of the guidelines (European Commission Notice – Guidelines on vertical restraints (2010/C130/01), May 2010).
  10. Speech by Margarethe Vestager, ‘Competition and the digital single market’, Forum for EU-US Legal-Economic Affairs, Paris, 15 September 2016.
  11. European Commission staff working document, ‘Preliminary report on the e-commerce sector inquiry’, 15 September 2016, paragraph 472.
  12. ASICS running shoes: Unlawful restrictions of online sales of ASICS running shoes, Bundeskartellamt (Germany), August 2015.
  13. European Commission DG COMP, Consultation on Evaluation of procedural and jurisdictional aspects of EU merger control, 7 October 2016.
  14. Department for Business, Innovation and Skills, Empowering and protecting consumers, April 2012, paragraph 57.
Published 15 November 2016