The past years have seen an extraordinary proliferation of sharing economy platforms, and the largest platforms have fast become global enterprises with market capitalisations that exceed competitors with “traditional” business concepts. These developments call into question a number of regulatory matters and pose new questions for national and European-level regulators.
Over the last few years, the sharing economy has gone from niche to mainstream. From its humble beginnings as Air Bed and Breakfast, which hosted conference attendees on airbeds in the home of its founders, Airbnb has become a global entity with a market capitalisation that exceeds those of “traditional” major hotel chains, while Uber, Didi and Lyft have made headlines by attracting investments in the range of several billion dollars.
The European Commission, based on a PwC study, estimated the size of the European “collaborative economy” at €28 billion in 2015.1 This figure is likely to become a common point of reference, even though it is actually not restricted to the sharing economy but also encompasses other forms of economic activity such as crowdfunding.2 Car ride-sharing and accommodation, the two major components of the sharing economy, accounted for €20 billion.3 While consensus has not yet emerged concerning the definition of sharing economy, and the term is often loosely applied to a number of very disparate business models and internet platforms,4 the common notion appears to run along the lines that the sharing economy is 1) peer-to-peer based rather than business-to-consumer or business-to-business based, and 2) based on rental or similar transactions/purchase of access rather than a permanent change of ownership/sale.5
Despite the uncertainties concerning the size and indeed the very conceptualisation of the sharing economy, it is clear that it has grown to become a substantial transformative force in the European economy that cannot be ignored by regulators. Accordingly, national and European policymakers are preparing to step in and evaluate and update regulation, but they are often confounded by the peculiar nature of the sharing economy, which circumvents many of the conceptualisations and ways of doing things that regulation traditionally presupposes. In the following, first the economics of the sharing economy are considered as well as the current state of play in terms of usage, and then the current regulatory challenges and developments that arise from the development and growth of sharing economy services are discussed.
The economics of sharing
The advent of sharing platforms stems from two developments that are both in turn consequences of digitalisation and internet technology becoming readily available for consumers and businesses.
Firstly, transaction costs have been massively reduced, since internet technology allows information to be exchanged virtually instantly at no cost. For instance, listing an apartment on Airbnb can be done at little cost, just as potential customers can easily browse the multitude of offerings at no substantial cost. Without internet technology, it would have been prohibitively expensive to advertise an apartment offered for short-term rent or lease to potential customers across the globe, and it would in most cases have been impractical for potential customers to search for individual apartments temporarily offered for rent to serve their needs for a short holiday. The Airbnb business model is thus simply not viable on a large scale without cheap information-exchange through the internet. People have always offered rooms or flats for rent, of course, but the market for short-term rentals was largely non-existent before the internet age.
Secondly, the intermediation of a neutral third-party platform helps to mitigate the risks and uncertainty concerning economic transactions with strangers; the platform builds the requisite trust for sharing economy services to function. Airbnb, for instance, acts as an intermediary and ensures that the money is not paid out to the landlord before the temporary tenant has had the chance to ascertain that the accommodation lives up to expectations. Moreover, it offers a guarantee to landlords in case the flat is somehow degraded while in the custody of the guests. Perhaps more importantly, intermediaries operate user-rating systems to help users avoid counterparties with a dubious track record. It is hard to see how, for example, people could be convinced to hand over their keys to complete strangers without an intermediary stepping in to provide this kind of framework to build up trust and weed out people with a history of misconduct.
The sharing economy as we know it is thus intimately interwoven with digitalisation. The qualitative change wrought by internet technology in this respect has translated into both qualitative and quantitative change: the amount of commercial peer-to-peer lending has increased from a trivial to a substantial size, and as the proliferation of sharing economy platforms testifies to, new sharing concepts continue to be developed by entrepreneurs eager to launch the next Airbnb or Uber.
In economic terms, there is an expectation that the development of the sharing economy may come to benefit society at large. The promise that the sharing economy holds is at least fivefold:6
- Sharing services may increase consumer utility by offering new kinds of services that were not previously available. For instance, Airbnb allows people to stay in more “authentic” accommodation than traditional hostels, some people prefer flats to a hotel room (so that one can cook, so that the whole family can live together, etc.), and whereas hotels have tended to cluster in certain areas, peer-to-peer accommodation offers a greater variety in terms of location. For instance, most New York City hotels are situated in midtown Manhattan, whereas Airbnb listings offer much greater flexibility in terms of which locations are available.7
- Sharing services may also allow for more effective resource allocation and utilisation. In the jargon of the sharing economy literature, “one does not want a drilling machine, but a hole in the wall”, and “one does not want a car per se but transportation”, so it is arguably wasteful for all households to own drills and cars that see very little use. Following this logic, it would be more economical for fewer people to buy cars and drills and instead simply rent the asset when the need arises. The resources saved from producing fewer cars and drills could instead be invested in creating other goods or services to increase consumer welfare (or analogously, people save money if they can rent instead of buying, which enables them to spend more on other goods or services). In economic terms, the sharing economy increases utility by shifting the budget constraints of individual consumers outwards. Likewise, the owners of cars and drills can choose to make extra money by renting out their assets, which increases their utility. Conversely, while the sharing economy is often expected to result in a “greening” of the economy, this is by no means an inevitable outcome and will depend on how the money that consumers save by shifting from owning to renting assets is ultimately spent.
- It has been argued that sharing economy services have positive societal externality effects by increasing social cohesion and third-person trust, which the social capital literature finds to be positively correlated with various indicators of well-being as well as economic growth.8 This effect arises because peer-to-peer commercial transactions lead people to interact with others, in many cases even with people from “outgroups”.
- Sharing economy platforms may increase productivity and economic efficiency by allowing for a more effective allocation of labour and capital/assets. For example, research based on a comprehensive study of data from the US has found that driver redundancies are much lower among Uber drivers than among regular taxi drivers.9 This is due to Uber’s use of price surging, which means that drivers are happy to adjust their working hours to meet demand. During peak hours, prices surge, leading more drivers to offer their services on Uber; conversely, supply declines as demand dwindles. Anybody who has tried in vain to get a taxi on a late Saturday night can appreciate why this increases consumer welfare. Moreover, according to this research, even though Uber operates with lower per hour prices than traditional taxi companies, the per hour remuneration in the US is in fact comparable.
- In some cases, there is nothing to prevent or prohibit traditional businesses from using the same techniques as sharing economy platforms. For example, apart from some statutory maximum prices in some countries, taxi companies could also adjust prices to demand, along the lines of Uber’s price surging, and they could use apps to ease access to taxi rides.10 However, sharing economy services have been the harbingers of innovation and new business models. As such, the sharing economy services are a strong force of “Schumpeterian” or innovation-driven growth that help unlock economic potentials and give agents more freedom over their consumer choices.
Reaping the productivity dividend
The rise of the sharing economy has led to widespread “disruption” in many industries as new business concepts challenge and often surpass traditional ones. Uber and Airbnb are often seen as the exemplars of the transformative power of digitalisation, and as many other industries are poised to be “disrupted” by new digital business models, many analysts believe that the modern economy is set to usher in a new era of productivity growth.11
But as analysts have frustratingly found, the empirics fail to conform with these expectations, as productivity growth – at least in the Western world – continues to stall. To paraphrase Robert Solow, we see platform economy services and new digital business models everywhere – except in productivity statistics. For regulators, one pertinent question therefore is whether something is blocking the digital economy and if by some regulatory action the long-expected productivity boom can be brought about.
At least three explanations can be given for this so-called productivity paradox.
One possibility is of course that this expectation is simply wrong. While few would argue that digitalisation has not brought any productivity gains, perhaps they have largely been exhausted already. And perhaps we tend to exaggerate the gains to be accrued from digital technologies. After all, as more and more office software for information-sharing has become available, employee time has been pushed towards marginal activities whose benefits for the company are small. For instance, setting up meetings with many participants used to be rather tedious but is now easy, thanks to commonly used software such as Microsoft Outlook; consequently, the threshold value a meeting must have before it is worthwhile to set it up has decreased. In the same vein, many employees complain about the large number of emails they receive with little valuable information. At a company level, digitalisation has spurred the growth of a number of new back office support functions, such as increasingly expensive ICT divisions, social media and communications experts, and many similar services. In short, based on the current productivity growth trajectory at least, it seems unwarranted to liken the current phase of digitalisation with another industrial revolution.12
Second, it is feasible that our current state of technological and digital development is simply not yet mature enough for it to translate into productivity gains that are felt at a macroeconomic level. For instance, as noted above, the US experience is that Uber provides greater productivity in the field of personal transportation but is too limited in scale to significantly affect the productivity level as measured across the entire economy. Of course, if the claims that digitalisation will (eventually) greatly enhance productivity levels are correct, we would expect such effects to accumulate over time as more and more industries experience disruptions and the process of digital innovation and transformation proceeds.
Third, there are well-known measurement problems that could lead us to insufficiently capture the true productivity effects of digitalisation. Qualitative changes are hard to adequately capture; price indices are developed by looking at the price of a basket of goods over time, but what if product quality increases so that they should also be worth more to the consumer? For example, the price of a personal computer has plummeted but its computing power has increased exponentially, so the true consumer value is not adequately captured simply by seeing how much cheaper a personal computer has become.13 In short, it could be that productivity is actually growing fast, but we have failed to pick it up statistically.14
Although sharing economy services constitute only a relatively minor subset of the total digital economy, examining sharing economy services may yield a part of the solution to the productivity paradox. More pertinently, from an economic point of view, if new digital business models and services indeed hold the potential to greatly increase productivity, we must ask what can be done by regulators and other actors to unlock it. This is a question that national governments and the European Commission are increasingly devoting attention to. Perhaps most forcefully, the Commission has made it a central policy objective to ensure a truly integrated Digital Single Market through regulatory harmonisation, since regulatory divergence and market fragmentation in digital fields are seen as major impediments to growth and business development in the EU.
Divergence and harmonisation in the sharing economy
Differences in the use of and the rules governing sharing economy services may well result in different approaches by national regulators. The British government, for instance, has been very forthcoming and declared its intent to make the UK the “global centre for the sharing economy”.15 Other governments have been less sanguine, especially in light of the resentment many labour unions and other political actors harbour against certain sharing economy services and, more generally, the presumed development of a “gig economy” of low-paid employees who will be unable to find the economic certainty and stability that full-time permanent employment provides.
Differences also arise at the level of the judiciary, as legal traditions differ among member states, which may cause the slowly developing jurisprudence concerning sharing economy platforms to develop in different directions across the EU. In the most visible example of this, Uber has been banned in a number of countries but continues to operate in others. Labour markets and labour law in particular differ vastly across EU member states, from the flexible Anglo-Saxon and Eastern European models to Scandinavian “flexicurity”, in which the state has little sway over labour market issues, to the continental and Southern European labour markets, where the state plays a pivotal role. As a result, regulatory and legal differences are bound to arise and perhaps increase over time as initial path-dependencies are established and affect the subsequent development of sharing economy regulation, as well as the actual development of sharing economy business models and concepts. Additionally, while there have been significant movements in the direction of greater European integration within labour legislation, e.g. with the service directive and more recently the working time directive, regulating the sharing economy is a comprehensive topic that impinges on many regulatory and legal aspects and touches questions more deeply engrained in the fundamental functioning of labour markets.
Interestingly, while there is considerable divergence in the extent to which sharing services are utilised across EU member states, these differences do not seem to mirror the general “digital divide” between more and less tech-savvy consumers and countries (see Figure 1). For instance, a Eurobarometer survey has shown the French and the Irish to be the most frequent users of sharing economy services, whereas Malta, Cyprus, Czech Republic, Finland and Portugal are least prone to do so. This variability suggests that the European economy may well be set on a path towards a confusing “patchwork” in terms of both regulation and usage across member states.
Utilisation of sharing economy services in the EU, by country
Note: Percentage of each country’s population indicating they have used such services, in response to the question “A collaborative platform is an internet based tool that enables transactions between people providing and using a service. They can be used for a wide range of services, from renting accommodation and car sharing to small household jobs. Which of the following matches your experience regarding this type of platform?”
Source: Eurobarometer 438: The use of collaborative platforms, March 2016, European Commission.
On the one hand, regulating the sharing economy in many ways seems like an obvious area where European-level integration will confer benefits. In the absence of common rules, sharing economy companies may find that different rules, regulations and interpretations apply to their business concept across member states. For example, one crucial question is whether a sharing economy operator (e.g. an Uber driver or an Airbnb host) should be considered an employee of the platform, a subcontractor of the platform or an independent contractor – and in turn, if deemed to be a contractor, whether the service provider would be seen as an independent business in his or her own right or simply a natural person. There is little reason to expect judges and regulators to come to the same conclusions across all EU member states concerning such matters. This divergence will make it harder for sharing services to grow and operate across borders and will generally hamper the development of the European sharing economy to the detriment of consumers and business.
On the other hand, the sharing economy touches upon a number of areas that are national competencies and in which member states and populations in some cases are reluctant to see legislation move to Brussels. As noted above, labour market laws and regulations are strongly rooted in national traditions and jurisprudence, which makes harmonisation with regard to sharing economy services difficult without more fundamentally harmonising labour market models. Regulating the sharing economy is therefore no easy task, and this is further compounded by the nature of the digital transformation that it is precipitated on.
Hitting a moving target
Regulating the sharing economy is a particularly daunting task because it is still evolving fast. Accommodation and personal transportation are arguably the only areas in which sharing economy services have gained significant traction, but the expectation is that many other areas will develop in the future. Regulators will benefit from a factual understanding of the sharing economy and its implications. For instance, an oft-used example to describe the regulatory challenges of the sharing economy is that it is unnecessary for households to own a drill that they rarely use if they can instead borrow one for a small fee. But if a drill is repeatedly rented out through sharing platforms, it will quickly get worn out, and the stipulated product guarantees might no longer apply. This is an obvious area that requires regulatory clarification and perhaps updating, but in fact it is probably a rather minor problem relative to other challenges posed by the rise of various kinds of platform services.16
While many observers of the sharing economy surmise that it will continue to grow and encompass new areas, it is very difficult to foresee in which areas it will stay a negligible niche concept and where it will gain traction. This means that regulation will invariably lag behind. Policymakers thus face the difficult dilemma of either being very specific, at the risk of impeding innovation and creating lock-in dependencies, or being less specific, at the risk of not properly addressing the issues in question and failing to provide enough clarity for the sharing economy to develop on an even regulatory playing field. Regulating the sharing economy is thus akin to hitting a moving target: it is a difficult task since one does not know how the future movement trajectory and pace will develop.
Regardless, the sharing economy has reached a size where regulatory action is called for. According to Eurobarometer data, 17% of all EU adults have used a sharing economy platform at some point (see Table 1). Sharing economy services are employed more frequently by the young, who tend to be digital first-movers. Another study of sharing economy users in one EU member state found that age and degree of urbanisation are the two strongest predictors of whether a household has used a sharing economy service or not, and that people with university degrees are more likely to have used such services than those with only primary school education.17
Utilisation of sharing economy services in the EU, by demographic group
Note: Percentage of each demographic group indicating they have used such services, in response to the question “Which of the following matches your experience regarding this type of platform?”
Source: Eurobarometer 438: The use of collaborative platforms, March 2016, European Commission.
Those sociodemographic groups that use sharing economy services less may well catch up fast. In one study, 77% of sharing economy non-users indicated that the reason they have not used sharing platforms was that they simply had not really thought about it.18 In short, there seems to be considerable potential for further growth in the sharing economy, which accentuates the need to ensure an adequate regulatory framework for sharing economy platforms.
Regulatory challenges and the risk of “patchwork Europe”
The rise of the sharing economy has garnered regulatory attention at both the national and European levels. There are several major issues of contention which regulators particularly need to address.
Employees or contractors
As mentioned above, one important question to resolve is whether sharing economy service providers (i.e. the Uber driver, the Airbnb landlord, etc.) are to be considered employees of the sharing platform or independent contractors. This is a crucial question, because employees have certain rights with regards to working conditions, paid leave under various circumstances, etc. Furthermore, they are entitled to remuneration for expenses, whereas independent contractors are not. For example, Uber drivers in the US have founded a union of sorts to centralise price bargaining with Uber, but if they are considered independent contractors rather than employees this would probably amount to an illegal cartel trying to coordinate prices between several business ventures. Several lawsuits concerning the status of Uber drivers have already been pursued in the US, but the legal differences are too large for American jurisprudence to yield any definitive answers for a European context.
Business entities or private individuals
If sharing economy providers are to be considered as independent contractors, we must in turn ask whether and under what conditions they should be considered business entities rather than simply private individuals. How this question is answered will have important regulatory repercussions, as businesses are regulated and must meet numerous administrative and regulatory requirements. For example, legislation concerning consumer rights and consumer protection generally foresees a relationship between individuals and businesses, but it may not apply to relationships between two individuals.19
On the one hand, it may seem excessive to expect a casual sharing economy service provider to understand and ensure compliance with a vast body of regulation, and it could well stifle the sharing economy if service providers are equated with businesses in all regards. On the other hand, if sharing economy providers are not considered businesses, and various industry and business regulations do not apply to them, then they might be construed as having an unfair advantage by dodging the rules that non-sharing operators must follow. Obviously, the expansion of sharing economy services will only benefit society if they truly represent something that is more efficient and yields higher customer utility – and not if their growth simply reflects that they are not held to the same requirements and standards as other businesses due to the regulatory ambiguity under which they operate.
The prevailing interpretation, and the one propagated by platforms fearing the consequences if they were to incur a vast mesh of legal responsibilities, may be that platforms are intermediaries only. It is well established, however, that platforms can be held liable for what they are used for under certain circumstances, so depending on the specifics, sharing economy platforms could conceivably face certain obligations that must be assessed on a case-by-case basis.20
Another range of questions arise concerning the nature of the contractual relationship entered into when somebody uses a sharing economy platform, and the liabilities and obligations that stem from the nature of the relationship for both the platform and the service provider. While some platforms are passive, others take a more active role. For example, most sharing platforms actively manage a trust-building user review scheme: Uber offers comprehensive guidance to drivers concerning the legal side of ride-sharing, while Airbnb intermediates in a way so that the rent is only paid out to the host once the guest has ascertained that the accommodation lives up to the agreed stipulations. Both platforms offer “guarantees” to users in case of accidents or other mishaps. This suggests that sharing economy platforms may in reality face more comprehensive liabilities and obligations.
While it is probably excessive to consider a sharing economy platform a contracting party and the actual service provider a subcontractor, this does not absolve the platform of any responsibility. For example, if a personal transportation platform failed to weed out a service provider with a criminal record, despite having set up procedures to do so, and the person committed a felony against a customer, it is reasonable that the platform might incur some legal responsibility. This follows existing jurisprudence on platform liabilities,21 but since sharing economy platforms are clearly different from other platforms (such as eBay) that previous rulings have covered, it is likely that their more extensive role implies more comprehensive responsibilities and liabilities.
A further issue raised by the advent of the sharing economy relates to the matter of insurance. As described above, the sharing economy works only insofar as some degree of trust can be established between strangers. In order to generate the requisite trust, some sharing platforms have begun to guarantee their users against mishaps. If the innocuous-looking tenant decides to thrash your apartment, Airbnb now guarantees to pay for damages up to a certain amount. It bears noting, however, that in case of disagreement between the platform and a service provider invoking the guarantee to demand compensation, this service would hardly amount to actual insurance, with the implied legal obligations on the part of the insuring company and with certain insurance industry standards to be observed and the possibility of calling upon neutral arbitration, etc.
Another major political issue with regards to sharing platforms is that of taxation. Income from sharing services should be declared and taxed, but one may well speculate that income accrued from sharing economy platforms in reality largely escapes the attention of tax authorities, as service providers choose not to report it. Tax avoidance is of course an endemic problem, but as the sharing economy rapidly expands, the magnitude of the problem is fast becoming non-trivial. Authorities will have to find ways to ensure proper oversight with these income streams.
On top of these difficulties and the legal uncertainties resulting from the sharing economy, there are also concerns regarding negative externalities. For instance, home-sharing leads to a regular flow of new people moving into a neighbourhood or apartment block, which can create a sense of estrangement and make permanent residents uneasy. In most cases, the new temporary tenants are tourists on holiday and tend to be noisier than permanent residents. City councils in some cities have decided to step in and restrict the use of home-sharing platforms. Other similar negative effects may conceivably arise from other types of sharing economy services, which could necessitate regulatory attention to resolve.
Regulatory bifurcation would obfuscate and hamstring the development of a European sharing economy. The Commission is now beginning to tackle the integration of audiovisual markets and e-commerce in the EU under the aegis of its digital economy initiatives. Much can certainly be done to regulate platform services. It has long been a concern of European policymakers that the EU appears to be lagging behind the US when it comes to digitalisation. It is often noted in Brussels that the EU has failed to produce a truly global IT company (although SAP and perhaps BlaBlaCar could arguably be seen as exceptions to this claim). Meanwhile, the US can boast of Facebook, Apple, Google and, indeed, Uber, Airbnb and a dozen other companies shaping the new digital economy. Moreover, while European countries do generate many major tech start-ups such as Skype, they tend to be acquired by large American corporations before they can grow to global dominance. It is therefore a concern that Europe will once again prove less adept at reaping the gains of digitalisation with regard to the development of a sharing economy, especially as a patchwork of different national regulations and legislation may hinder the development of pan-European sharing economy services.
For national and European regulators, then, there is much to be done, and the stakes are considerable. Regulation must strike the difficult balance between being too loose and vague, in which case it would not bring about the much-needed clarity and the even regulatory playing field, and being too stifling and heavy-handed, in which case it would thwart the development of the European sharing economy. Failure would prevent the EU from fully reaping the economic rewards that the sharing economy offers.
- 1 European Commission: A European Agenda for the collaborative economy, Brussels June 2016.
- 2 R. Vaughan, R. Daverio: Assessing the size and presence of the collaborative economy in Europe, PwC, 2016.
- 3 Airbnb and Uber are the two major components of the sharing economy, while other types of sharing economy services remain small and limited to niche markets. See for example A. Sundararajan: The Sharing Economy, Cambridge, MA 2016, MIT Press, p. 14.
- 4 For example, Sundararajan concludes, perhaps somewhat laconically, that “I am unaware of any consensus on a definition of the sharing economy.” See A. Sundararajan, op. cit., p. 22.
- 5 M. Munkøe: Deleøkonomien på fremmarch, Copenhagen 2015, Dansk Erhverv.
- 6 For a useful discussion, see also V. Demary, B. Engels: Collaborative Business Models and Efficiency – Potential Efficiency Gains in the European Union, Impulse Paper No. 7, Cologne Institute for Economic Research, 2016.
- 7 A. Sundararajan, op. cit., p. 122.
- 8 See, for example, R.D. Putnam: Bowling Alone, New York 2000, Simon & Schuster, for a useful summary and introduction to the literature on trust.
- 9 J. Cramer, A. Krueger: Disruptive Change in the Taxi Business: The Case of Uber, NBER Working Paper No. 22083, 2016.
- 10 Of course, price surging may work better with a part-time workforce that supplements a normal income by driving a couple of hours from time to time in peak hours than having it as a full-time job that needs to bring in a living wage each month and drivers who prefer guaranteed hours and guaranteed prices or income levels. Here we focus on the economics of the sharing economy and leave the political implications aside.
- 11 E. Brynjolfsson, A. McAfee: The Second Machine Age, New York and London 2014, Norton & Company.
- 12 See for instance K. Schwab: The Fourth Industrial Revolution: what it means, how to respond, World Economic Forum, Annual Meeting, 2016.
- 13 The measurement problem is not a constant but is accentuated by the process of digitalisation since the qualitative properties of electronics and digital services are harder to measure and change more rapidly than more traditional consumer goods.
- 14 Andrew McAfee argues that low growth depresses demand for high-productivity labour, which retards the movement of labour to high-tech industries where digitalisation allows for productivity growth. See A. McAfee: How to reconcile great technology with lousy productivity, 15 May 2015, available at http://blogs.ft.com/andrew-mcafee/2015/05/15/how-to-reconcile-great-technology-with-lousy-productivity/.
- 15 Government of the United Kingdom: Move to make UK global centre for sharing economy, 2014, available at https://www.gov.uk/government/news/move-to-make-uk-global-centre-for-sharing-economy.
- 16 L. Gansky: The Mesh, New York 2010, Portfolio Penguin, develops a useful framework that distinguishes between high and low value assets that may see extensive or rare usage. Sharing concepts are economically more viable for high value, low usage assets.
- 17 M. Munkøe: Deleøkonomiens vækstpotentiale, Copenhagen 2016, Dansk Erhverv.
- 18 Ibid.
- 19 This is an area where national legal practice may differ and where there is generally some uncertainty as to whether non-business natural persons may in fact be covered by legislation that covers certain kinds of transactions relevant to the sharing economy. See for example A. de Streel, J. Bravo: Specific liability issues raised by the collaborative economy – professional service, Impulse Paper for DG GROW of the European Commission, 2016, CRIDS/University of Namur, especially the discussion on p. 40.
- 20 See A. de Streel, J. Bravo, op. cit., for a useful exposition.
- 21 See in particular L’Oréal SA and Others v eBay International AG and Others, Case C-324/09, 12 July 2011.