Andrea Coscelli: staying ahead in a digital marketplace
Andrea Coscelli, CMA Acting Chief Executive, explores how the CMA is responding to the digital evolution of the modern marketplace.
As I am sure we will hear several times during the course of today’s hugely interesting agenda, disruption is a key feature of the new economy. As digital ways of doing business have proliferated, we have seen an exponential increase in the rate of technological change and in the speed with which new ways of doing business have taken hold.
I would like to open the day with some thoughts on how, in competition terms, such digital disruption can be central to both competitive ‘illnesses’ and their cures. I will explore the role that we, the Competition and Markets Authority (CMA), alongside other competition enforcers, play in that digital evolution; reflecting in particular on the way in which we have used and will continue to use our full range of competition and consumer protection tools, to seek to drive – and remove impediments to – better competition, choice and consumer trust in online markets.
So what should we make of all this disruption?
From a competition economist’s point of view, cycles of disruptive change in markets are often regarded as beneficial. The development of new or innovative ways of doing things – often emanating from academic institutions like this – helps stimulate further innovation, opens up new markets, and forces incumbents to ‘up their game’.
And, as consumers, we can see the benefits as well in the choices available to us: compared to even a few years ago, we now have near instant access to a huge range of digital products or services which previously we might not have even become aware of, let alone been able to benefit from.
But whether you consider a particular change to be good or bad, disruption can evidently be unsettling. Traditional businesses can suffer sudden reverses in fortunes if they are not able to adapt quickly enough to new sources of competition or evolving consumer preferences, at times leading to restructuring and redundancies. While it is the demise of famous high-street names – and the painful job losses that come with such collapses – that inevitably attract the headlines, the same process is taking place daily for small businesses without benefiting from the same media coverage. At the same time, as individuals and consumers, we are taken out of our comfort zone: what was previously familiar can now seem rather different, and rules we took as read may no longer apply:
Can we be sure that the taxi we book through our smartphone app will get us home as quickly as the cab that, in the past, we’ve always just hailed on the street?
Can we be confident that the products we are buying from sellers in Bratislava, Bremen or Brussels and the service they provide, will meet our expectations in the same way as those we get from our nearby high street seller who has served the local community for generations?
Can we trust the information that’s available to us on the internet to help us decide whether those sellers are good or bad?
And these are some of the questions and challenges that, as policy makers, we also have to think just as hard about. If old rules and concepts are being challenged, how should competition authorities and governments react?
How do we make sure not just that we keep up with the pace of change, but also that we help facilitate that change where it promises to benefit consumers and guard against efforts – by the disruptor or the disrupted – to exploit consumers or exclude competitors?
How do we distinguish a novel way of delivering a service from a potentially harmful practice covered up in a ‘digital veneer’?
Should we set the ‘rules of the road’ in advance, or should we let new services become established and only step in when problems arise?
And if we do choose to step in, are our own rules, policies and processes ‘fit’ for the digital age?
These issues have engendered significant discussion and debate among policy makers and stakeholders in the past few years. In particular, in May 2015, the European Union launched its wide-ranging Digital Single Market Strategy, to which the CMA has contributed its views. This strategy sets out a range of measures to grow the digital economy across Europe and to ensure both better access to digital goods and services, and a legal and regulatory environment in which digital networks and innovative services can thrive. And closer to home, the EU Select Committee in the House of Lords recently carried out a complementary inquiry looking specifically at the competitive impact of the online platforms that have become such a feature of the e-commerce landscape. The CMA was one of a wide range of respondents – including governmental bodies and businesses – that gave evidence. In April, the committee issued a detailed report highlighting both the benefits digital platforms can offer and various potential issues to which their conduct or market position can give rise, along with its recommendations for how such concerns might best be guarded against.
While certainly not the only piece in the puzzle, promoting competition and consumer protection, with which we at the CMA are principally concerned, is clearly central to any drive to foster open and innovative markets that work for both customers and businesses. But we need to do so in a way that still allows businesses scope to respond in new and innovative ways to evolving technologies, markets and consumer demands – setting the framework for businesses to compete freely, and thus avoiding some of the possible risks of ossification and hard-wiring of the status quo that can arise from seeking to put in place detailed ex ante regulatory rules.
And that is something that I believe is reflected in all aspects of the way in which the CMA operates.
It can be seen in our strong, prominent public advocacy on achieving the balance between allowing new competitive entry from ‘disruptive’ technologies while retaining adequate consumer protection: for example, in the past year, in response to regulatory proposals in minicab markets in London and Liverpool. We have argued for legislative and regulatory frameworks which allow new or disruptive business models that harness the opportunities of the interconnected digital world to develop, while ensuring consumers remain adequately protected by regulations that are truly necessary (and no wider than necessary) for such protection.
It can be seen in the way we identify potential issues, be it through our use of social media to track trends and engage directly with consumers, or through our investment in forensic intelligence capabilities to enable us to capture and process digital evidence ever more effectively.
And, most tangibly, it can be seen in the direct interventions we have made in those areas where we have found evidence of harm to the competitive process or directly to consumers.
And it is on those interventions that I plan to focus my comments today. Because if we are to warn against the risks of inflexible regulation, and to advocate instead for a central role for competition authorities in promoting innovative and open digital markets, then we have to ‘put our money where our mouth is’. We have to ensure that our interventions are effective. That means taking care that we don’t jump in too early and risk chilling potentially pro-competitive innovation; while equally working to ensure that where we do intervene, that intervention does not come too late for businesses struggling against well-funded opponents or for consumers. It means recognising on the one hand the role of regulatory schemes like the patent system – which on their face limit competition, but provide an important incentive to encourage firms to invest in innovation – while, on the other hand, stepping in where companies seek to exclude competition beyond the legitimate scope of such regulatory protections.
I hope that by reflecting on just some of the actions we have taken of late, I will show how we have embraced that challenge.
I will start with our recent competition enforcement, particularly as it neatly illustrates the quite different ways in which competition in a market can be affected by the emergence of digital tools or means of selling.
It’s worth noting at the outset that we receive a wide range of complaints across a variety of sectors, and must necessarily prioritise. Looking back, and given this, I think that the number of investigations we have undertaken recently involving online sectors reflects not only our desire to play an important role in the digital economy, but also the significant impact that we believe anti-competitive activity in those sectors can have on consumers, and thus the equally significant scope for our intervention to bring about benefits and avoid harm.
So, in recent months, we have imposed fines that total almost £3 million against suppliers of commercial catering equipment and of bathroom fittings for imposing restrictions on retailers who were using the internet as a distribution channel. Specifically, we were concerned about restrictions on price competition in those online channels. The precise restrictive measures employed by the supplier in each case were slightly different:
In the commercial catering case, ITW Ltd, a supplier of commercial fridges, imposed a ‘minimum advertised price’ policy restricting the price at which retailers could advertise ITW’s product online, and enforced this policy with threats of higher wholesale prices or even a complete cessation of supply for ‘non-compliant’ retailers.
The bathroom fittings case did not involve a ’minimum advertised price’ policy, but rather a supplier, Ultra Finishing Ltd, threatening – again with higher wholesale prices or withheld supply – those retailers who did not price at or above the Ultra’s ‘recommended’ online price for its products.
But the key is that the outcome in each case was the same: the suppliers’ actions limited the ability of retailers to set their own prices online and, therefore, in practice to offer lower prices to end customers.
We appreciate that it can be challenging for a supplier of a product to ensure that bricks and mortar retailers through whom they sell their products continue to have the incentive to, for example, bear the cost of maintaining a showroom or a display for those products. But there are legitimate ways to achieve this, without relying on practices that seek to frustrate the opportunities that the online economy provides for retailers to compete to provide consumers with the best value products and services.
We are currently investigating similar concerns around the sale of golf clubs: specifically that Ping, one of the best-known golf club manufacturers, prevents its authorised retailers from selling Ping golf clubs online, potentially in breach of competition law? We issued a statement of objections to Ping in June, and will duly consider its response to our allegations and any justifications put forward. And we are similarly progressing other, earlier stage cases. We are of course sensitive to the fact that most businesses want to comply with the law: which is why we’ve followed up our interventions in the cases I have mentioned earlier with a range of compliance activities, aimed at providing accessible materials and information to help businesses to understand the law and when their actions risk falling foul of it.
But it is not just instances of businesses seeking to frustrate the use of new technologies that we face. Equally, we see examples of those looking to use digital technologies as tools to facilitate anti-competitive ends. In July, following a CMA investigation, Trod Limited, an online seller of merchandise and other products, admitted having agreed with one of its competitors, GB eye Limited, that they would not undercut each other’s prices for posters and frames sold on Amazon Marketplace in the UK.
What makes the case particularly relevant for today’s purposes is that the agreement was put into effect using automated repricing software that adjusted each manufacturer’s prices automatically to ensure that they didn’t undercut each other. As competition authorities, we need to stay alive to the potential for such software to be used in ways which seek covertly to distort markets, particularly in an age of increasingly ‘intelligent’, ‘self-learning’ technologies and algorithms. As the then US Assistant Attorney General Bill Baer recently commented following an investigation by our counterparts at the Department of Justice into the anti-competitive use of algorithmic software: ‘We will not tolerate anti-competitive conduct, whether it occurs in a smoke-filled room or over the internet using complex pricing algorithms‘. That is a sentiment which we in the CMA very much share.
But the ramifications of such behaviour by online operators are not just felt in the specific markets in which they are implemented. Rather, they infect other markets, not least by undermining consumers’ trust in online markets.
And that trust is vital if the digital economy is to continue to develop and meet its full potential. As consumers, we want to be sure that ’what we’re seeing is what we’re getting‘, and all the more so where the goods or services are new or unfamiliar to us.
Take cloud storage services for example.
The ability easily to store, back up and access our data in large volumes can evidently be of great value to us as consumers: it can protect that data from loss – say, if our smartphone breaks or is stolen – and allow us to share it with other people or across different devices.
But those services are also unfamiliar to many consumers and can therefore leave them vulnerable when deciding which ones to use. When – often with a single click or tap – we upload our information to the cloud, do we in fact know what are we really agreeing to? Are the terms we’ve just signed up to even fair? Do they make it harder for us to switch suppliers (thereby also weakening competitive pressures)? Having recently reviewed these services, the CMA has found evidence that, at present, some terms do indeed appear unfair, with certain storage providers’ contract terms and practices potentially breaching consumer protection law. It is at that point that we feel it is right to step in: we have already taken action that has led several providers to commit publicly to change those terms to make them fairer to consumers, and we are continuing to work in this area to ensure that the terms that providers offer comply with the law.
The market for online apps is another good example, and one in which the CMA and its predecessors were among the pioneers of action – now replicated on a coordinated basis internationally – to address concerns.
Apps have evidently proven hugely popular with consumers: the average person in the UK now has around 50 apps on their smartphone or tablet, the vast majority of which are – initially at least – available free of charge. But you may also recall the press stories in the past of people receiving credit card bills for thousands of pounds as a result of in-app purchases that they – or their children – had made on their digital devices, often seemingly inadvertently. Particularly to the extent that some of those apps appeared to be marketed at young children, who one can see may be less able fully to appreciate when their use of an app was incurring further charges, the scope for consumer harm and mistrust is apparent. The CMA, and other enforcers internationally, took action to make sure that customers would not, for example, be charged when using apps without their explicit consent. In response to the pressure that this brought to bear, Apple and Google agreed to change their payment authorisation systems, to reduce the chance of customers inadvertently incurring additional charges and so to avoid a repeat of the ‘horror stories’ of the past. This is a clear example of targeted intervention by competition and consumer authorities having a positive, lasting impact on the way an important digital market has subsequently evolved.
But trust isn’t just about the end product or service in itself. In a world where over half of adult consumers in the UK now consult an online review before making a purchase, the need for such trust extends to all aspects of consumers’ online purchasing ‘journey’, including trust in the information those reviews provide. That is why we recently took action against various online review sites for practices that we thought had the potential, contrary to consumer protection laws, to mislead consumers and to damage trust. Those practices included posting fake reviews and failing to disclose when social media posts and other content were in fact paid-for promotions. Again, our action has led to various companies formally agreeing to change their practices, and we expect other providers to review their own behaviours and ensure they are in line with the law.
We are committed to continuing to pursue such cases and to taking robust action where we identify traders using unfair terms or practices. It is not fair on those businesses that do comply with the law if there are insufficient sanctions for those that don’t, and there is a risk that their incentives to comply may reduce. For that reason, we continue to believe it would be highly desirable for the law to be changed to provide powers for the courts or the CMA, in appropriate cases, to impose civil fines on businesses that break unfair trading laws. We think that such an ability would provide an important additional deterrent against unlawful practice: it would be consistent with the powers we have available for violations of competition law, and would complement our existing powers to restrain future conduct and seek redress for victims, while being more flexible and proportionate than our ability to take criminal law action.
Before I move on, I would note that in discussing these issues, we should not mistakenly assume that safeguarding consumers is somehow wholly distinct from the promotion of competition: there are evident cross-overs between the two, something which questions regarding the veracity of online reviews effectively demonstrate. The trust that we as consumers derive from having confidence in the reviews that we read is not just important for us in deciding what to do: it is an integral part of the competitive process in that market. Better-informed choices by consumers serve to intensify competitive pressures, which in turn stimulates suppliers to strive to improve the quality and value-for-money of the goods and services they supply. And where that trust breaks down, where consumers disengage from transacting or make misinformed choices, this can similarly skew or reduce competitive tension between firms, potentially impeding the dynamic competitive evolution of markets.
To assess the multitude of issues that I’ve already discussed today and their likely impact on both consumers and competition, it is evidently vital that, as competition authorities, we properly understand how those markets work and, crucially, how consumers engage in them. How they search for, evaluate and compare particular goods and services, for example. We need to go beyond what consumers ‘say’ they will do, and look at what they ‘actually’ do, including by using behavioural economics and other empirical evidence to demonstrate the true ways in which we all in fact respond when faced with new products or services.
The CMA’s markets powers are well-designed for this purpose, enabling us to look broadly across a sector, and to use that wide insight to then take action targeted on areas in which, and using the tools through which, it is most likely to be effective. The enforcement action that I’ve just mentioned in relation to cloud storage and online review services followed on in each case from an earlier cross-sector assessment. Our analysis of each sector shone a broad light on issues that were prevalent across the markets in question and the potential effects of those issues on consumers. But it also uncovered various practices by specific businesses that we considered warranted direct enforcement action against those individual firms. Last year, we also carried out a call for evidence relating to the commercial use of consumer data, including in particular the data-driven markets that characterise the digital economy. Again, that work highlighted those situations in which consumers were most at risk from losing out or competition was most likely to be impeded, and has helped to create a framework for analysis that we, and others, can use when assessing competitive and consumer effects in heavily data-driven markets.
The success of these pieces of analysis has encouraged us to undertake similar efforts in future in other segments of the online economy. In the next few weeks we will launch a review focused on ’Digital Comparison Tools‘ – price comparison websites mainly, but also other similar apps and digital tools that enable consumers to compare and perhaps switch different providers of a particular good or service, not just on price but also on quality.
In fact, price comparison sites provide a nice example of the potential for disruptive digital services to be used both to invigorate a ‘sleepy’, uncompetitive market, and conversely as a mechanism in some cases to protect the ‘new incumbents’ from further entry or the next cycle of competition.
When price comparison websites (PCWs) emerged in earnest a decade or so ago, their arrival helped to inject significant competition into a number of markets. Those sites – many of which have grown to become incredibly widely-recognised brands – have enabled consumers to more easily compare prices and policies, encouraged them to switch between suppliers to get a better deal – especially for goods or services where there might not otherwise be an obvious ‘trigger point’ that would spur them to do so, and enabled new providers to enter without first having to build up a significant brand or web presence, as they might otherwise have had to do.
The market for private motor insurance is evidently one market in which PCWs now play a prominent role. Indeed, the entry of PCWs into that market again very much initially helped drive increased competition between existing insurance providers. So far, so good, from a competition authority’s point of view. But that market, into which the CMA carried out a market investigation, also showed the problems that can sometimes still arise: our investigation found that the market had become characterised by agreements between insurers and price comparison sites that prohibited insurers from making their products available more cheaply on other online platforms, restricting competition and further entry by new PCWs, and leading to higher car insurance premiums overall.
But while mindful of such instances, we should not ignore the significant potential for digital tools to be the cure for ‘sickly’ markets: not least because, as competition authorities, we can incorporate those tools into our interventions to drive the more effective competitive outcomes we wish to see.
Two recent, but quite different, examples from our work demonstrate this well.
Most recently, to address concerns raised by our market investigation in the UK retail banking market, we set out a wide-ranging package of remedies, including a requirement on banks to implement an ‘Open Banking’ standard by early 2018. This Open Banking – based on the sorts of Application Programming Interfaces (APIs) that, for example, allow you to log into other online accounts using your Facebook account – will enable customers to share their data securely with other banks and third parties, and to manage accounts with different providers through a single digital app. We believe this will have a significant beneficial effect, enhancing customers’ ability to take control of their funds and to compare products to find the one that suits them best. And by mandating it to be put in place within the next 18 months, our remedies ensure not only that such Open Banking will be rolled out, but also that that roll-out and adoption will happen more quickly than we believe would have been the case without our intervention.
That same focus on exploiting the full potential of online tools to empower consumers and enable better informed choice is also evident in several of the remedies proposed earlier this year after our investigation into the UK energy market. In particular, those remedies are intended to foster the conditions for price comparison websites to play a more active role in helping customers find the best offers for them, for example, through the ability to access meter data to enable customers to search instantly for deals.
Before I conclude, I am keen to touch on our merger control responsibilities. Merger control is another area where recent market developments have generated considerable debate among stakeholders and politicians across Europe, as to whether existing regimes are able to assess the competitive impacts of the actions – in this case the acquisitions – of businesses active in the digital economy.
In the specific mergers context, much of that debate has centred on wholly ‘turnover-based’ thresholds for determining which mergers require formal clearance in different regimes. The concern is that these risk allowing mergers involving target companies with no or limited turnover to avoid scrutiny, even if those targets may still have a significant competitive impact on the marketplace overall – say through a particularly innovative business model, or simply because they have still to start to monetise the market position that they have built up.
The risk of potentially harmful acquisitions slipping through the cracks is particularly acute in digital markets that are characterised by innovative new companies with the potential to change the game: an impression created in particular by acquisitions by the likes of Google and Facebook of innovative, emerging companies in the midst of, or with the potential for, rapid growth.
But further concerns have also been raised as to the digital ‘fitness’ of other aspects of current merger control processes. Some have worried, for example, that certain authorities’ substantive analysis of the competitive effects of a merger is too focused on price effects. The fear is that such a pure ‘price focus’ may fail to fully capture effects on other metrics of competition, including quality and innovation, particularly in digital markets where the goods or services at issue are – on their face at least – provided free of charge to the end consumer. And others have noted the challenge for authorities of assessing the relevant post-merger counterfactual in markets which are very new or in which new online products and offerings may shortly be – or may already be – fundamentally and rapidly changing the dynamics of competition.
We may hear some of these points made during the panel session later today. But from a UK-centric perspective, I think that the system we have here, and the experience we have built up, leave us well placed to avoid some of the risks to which I have just alluded.
The fact that our jurisdictional thresholds look not only to turnover, but also to the parties’ combined shares of supply, has meant that we are able to review important mergers involving targets with no or low turnover – as with Google’s purchase of Waze, or Facebook’s acquisition of Instagram.
And our substantive test – the ‘substantial lessening of competition’ standard – and the economic analyses we apply in assessing it, have shown themselves through a number of past cases involving fast-moving and dynamic markets, to be very capable of effectively assessing a range of competitive metrics, including impacts on innovation and new online entry.
That ‘digital’ experience is evident even in our very recent caseload. On a number of recent occasions we have considered in detail the extent to which the emergence of online services and sales channels impacted the likely competitive effects of the merger. We don’t apply a one-size-fits-all approach to that question, but rather consider each market on its own terms. Indeed, that is something well demonstrated by two cases that we have just reviewed: ICAP’s sale of its voice and hybrid broking and information businesses to Tullett Prebon, a competing broker of financial products, and Wiggle’s purchase of fellow cycling equipment retailer Chain Reaction.
In the latter, we found that the nature of online competition in the sector meant that there was high price transparency, little consumer brand loyalty, easy product and supplier comparison, and non-exclusive supply chains, and in turn that the remaining providers in the market, although smaller, were able to exert a competitive constraint beyond that which their size would suggest.
While in the former, having analysed in detail competitive effects across 20 overlapping product categories, we identified competition concerns in only one category (broking of oil products), where – in contrast with wider industry trends towards ‘electronification’ – use of alternative, electronic trading platforms for that particular product remained limited. Yesterday, we accepted undertakings offered by the companies that address our specific concerns in that area, and cleared the merger.
There are, of course, several other examples to which I could refer, all of which reflect a theme that runs through much of what I’ve said this morning. That is the sense that, fundamentally, the underlying principles on which our interventions have to date been founded are just as applicable in the new economy as they were in the old, and have been shown to provide the flexibility we need to adapt them to the new business models and new practices that we find. We might need to apply the rules in new contexts, but that doesn’t mean we need a completely new set of rules.
There is a risk that we are ‘dazzled by the new’; that we consider anything with an ’e‘ or an ’i‘ in front of it to be somehow special and requiring a fundamentally different set of rules. Or that we see the acquisition of a start-up tech company as necessarily needing to be scrutinised wholly differently from past mergers involving nascent companies or R&D-led activities.
But we need instead to take a step back: competition authorities and competition rules have shown themselves capable of effectively addressing emerging technologies and rapidly changing markets in the past, and I am confident that they will do so again in the future.
I’d stress though, that that confidence should not be mistaken for complacency, or as indicative of a view that there is not more that we can do, or have to do, to ensure that our rules and practices are able to address the challenges of today’s, and tomorrow’s, markets. The digital evolution continues, and, if recent history is any guide, will only gather pace in the future. We need to make sure that our practices and interventions continue to match that evolution stride for stride. This requires boldness and creativity:
We need to leverage the advantages that come from having both competition law and consumer law powers and be willing boldly to use our full range of tools, and consider alternative approaches and outcomes, to bring about benefits or to end harm more quickly.
We need to find ways to further build our own digital ‘expertise’ to help us better assess competitive dynamics not only of today’s digital markets, but also those of the markets that may emerge in future were we to intervene – or not intervene.
We need to recognise that the increasingly fluid and borderless nature of those markets means both: a) that problems UK consumers face are liable to be equally keenly faced by consumers overseas and b) that we can have greatest effect by working in co-operation and sharing insights with other enforcement bodies in the UK and overseas. This will be the case, whatever specific ‘post-Brexit’ institutional arrangements emerge from the government’s upcoming discussions with our European counterparts.
And we need to continue to think creatively about how best we can use the power of digital tools ourselves, whether:
as part of our investigative processes, identifying areas for possible intervention and gathering digital evidence;
as part of our remedial actions to address problem; or
even simply to explain our actions, to make businesses and consumers aware of their rights and responsibilities, or to spread the ‘gospel’ of the societal benefits of competition.
But as well as boldness and creativity, rapid change also requires prudence, rigour and robustness to reduce the risk of authorities intervening in ways that serve inadvertently to undermine or inhibit such competitive evolution, whether by protecting the narrow private interests of incumbents reluctant to adapt to new competitive challenge, or by creating new uncertainty as to the bounds of the law.
Walking that digital tightrope is the challenge we face. It is an important one, and it will be a constant one. But it is one that we stand ready to continue to respond to and to meet.