University of Amsterdam
Faculty of law
The threat of powerful sharing economy platforms in EU
Competition law: The Airbnb case.
A comparative case study on the impact of Airbnb’s business model on the hotel sector in France and Europe.
LLM Thesis
Name: Zoé Paponneau
Email : zoe.paponneau@gmail.com Supervisor: Pr. Rein Wesseling
Master program: European Competition Law & Regulation (International and European Law)
Student number: 12557730 Words count: 12 936 words Date of submission: 19/07/2020
p. 2 Acknowledgements
I would like to thank the master’s program director Pr. Dr. Annette Schrauwen, who allowed me to follow this enriching program, and my supervisor, Rein Wesseling. Also, I would like to express my gratitude to the University of Amsterdam, Ms Cseres, program coordinator, and the Competition Law Department in particular. Finally, I would like to thank my family, my friends, and my boyfriend for their never-ending support - and especially Claire-Emeline Auduc for her critic but kind and useful eye on my work.
Statement of Originality
This document is written by Zoé Paponneau, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Law is responsible solely for the supervision of completion of the work, not for the contents.
p. 3
Abstract
‘Now we are fit to talk about a transformative economic philosophy that alters the rules of the game in an increasingly visible way’ said Elena-Mădălina Vătămănescu and Florina Magdalena Pînzaru in 2018, while analysing the new business models of sharing economy platforms as Airbnb and Uber. Indeed, the XXIst century is the century of new technologies, progress, and of a constant growth sometimes difficult to follow. It is characterized, among other things, by the development of the so-called sharing economy which has boomed in 2017. It was born from a need to find peer-to-peer solutions in a capitalist world in constant evolution. Today, online platforms are at the core of our societies and have been an efficient solution to allow the connection between peers. However, this movement and change in the economic model make it difficult to define. This thesis has for scope to highlight this opposition between online and offline markets which has been one of the biggest stakes of the XXIst century and to clear the impact of the former on the latter. It is meant to identify how Airbnb’s business model must be characterized, especially its functioning on the relevant market of French short-term rental market, but also the impact it has on neighbouring markets. EU and French legal frameworks will be examined to determine if Airbnb could be taking advantage of a legal tension between the European legal orders. Finally, it is necessary to analyse if competition law could apply, or if new criteria are necessary for a better assessment of the effects of platform economy. A complex question taking into account the complexity of multi-sided markets in relation to the sharing economy.
p. 4
Table of content
INTRODUCTION ... 6
PART I – WHAT IS THE BUSINESS MODEL OF AIRBNB ? ... 10
1–HOW DOES AIRBNB ACTION QUALIFIES ON THE MARKET OF SHORT-TERM RENTAL? ... 10
1.1 HOW TO DEFINE THE RELEVANT MARKET? ... 10
1.1.1 The market of short-term rental ... 10
1.1.2 The complex assessment of the relevant market in multi-sided markets ... 11
1.2. DOES AIRBNB ACTION QUALIFY AS A SERVICE? ... 12
2–TRADITIONAL HOTELIERS VS AIRBNB = DOES THE PLATFORM PREVENT THEM FROM COMPETING ON THE MARKET OF SHORT-TERM RENTAL? ... 13
2.1 HOW DOES CONSUMERS’ TRUST IN THE PLATFORM REDUCE THE POSSIBILITY FOR TRADITIONAL ACTORS TO COMPETE? ... 13
2.1.1 The rating system as a trust factor ... 13
2.1.2 Reduction of the number of goods on the market : the consequence of this increasing trust 14 2.2 DOES AIRBNB’S ACTION REALLY ANSWER A SPECIFIC DEMAND? ... 15
2.2.1 Peer-platform: a new business model ... 15
2.2.2 A demand actually created by the platform ... 16
PART II – EU LAW BEYOND COMPETITION: DOES AIRBNB BENEFIT FROM A TENSE LEGAL FRAMEWORK? ... 17
1–IS THERE A LACK OF REGULATION AT THE EU LEVEL, NECESSARY TO ENSURE EQUALITY BETWEEN ECONOMICS OPERATORS? ... 17
1.1 THE SERVICES DIRECTIVE: TOO BROAD TO BE EFFECTIVE? ... 17
1.1.1 The content of the services directive ... 17
1.1.2 The outcomes of the services directive in EU legal systems ... 18
1.2 IS THE E-COMMERCE DIRECTIVE SUFFICIENT TO FRAME THE DEFINITION OF A SERVICE IN A VIRTUAL ENVIRONMENT? ... 18
1.2.1 The content of the e-commerce directive ... 18
1.2.2 The Airbnb Ireland case: the complex application of a non-exhaustive directive ... 19
2-DOES FRENCH PROTECTIONISM THEREFORE RESTRICT THE FREEDOM TO PROVIDE SERVICES, AND COMPETITION, BY AN EXCESS OF REGULATION? ... 21
p. 5
2.2 WHAT ARE THE CONSEQUENCES OF FRENCH LAW ON ESTATE AGENTS REGARDING THE COMPETITION BETWEEN
AIRBNB AND THE TRADITIONAL HOTELIERS? ... 21
2.3 IS THIS CONSTANT FIGHT BETWEEN FRENCH REGULATION AND AIRBNB ALLOWS A RIGHT CONTROL OF THE PLATFORM? ... 22
PART III – IS COMPETITION LAW SUFFICIENT TO CONDEMN AIRBNB REGARDING THE COMPLEXITY OF THE MARKET ON WHICH IT OPERATES? ... 23
1–COULD WE APPLY ARTICLE 102FEU TO A POSSIBLE DOMINANT POSITION ON THE SHORT-TERM RENTAL MARKET? ... 23
1.1 DOES THE PLATFORM HAVE A DOMINANT POSITION? ... 23
1.1.1 The reality of the market: a multitude of competitors ... 23
1.1.2 The complexity of a two-sided market ... 24
1.1.3 What are the consequences of spill-over effects in the platform economy? ... 25
1.2 IS THERE AN ABUSE IN AIRBNB’S CONDUCT? ... 26
1.2.1 Airbnb’s action: an exclusionary conduct? ... 26
1.2.2 Airbnb’s action: an exploitative conduct? On the application of article 102 (a) TFEU ... 27
2–DOES THE RELATION BETWEEN AIRBNB AND ITS PEERS COULD ALLOW THE APPLICATION OF ARTICLE 101TFEU? ... 28
2.1 AIRBNB AND THE USERS: A VERTICAL AGREEMENT? ... 28
2.2 HUB-AND-SPOKES AND THE INTERPLAY BETWEEN UPSTREAM AND DOWNSTREAM MARKET: A POSSIBLE CONDEMNATION OF AIRBNB FOR RESTRICTIVE VERTICAL AGREEMENT? ... 30
PART IV – TOWARDS NEW CRITERIA TO QUALIFY PLATFORM’S RESTRICTION OF COMPETITION? .. 31
1–HOW TO ASSESS AIRBNB’S ACTION WHEN NO QUANTIFIABLE MARKETS ELEMENTS ARE RELIABLE? ... 31
1.1 IS IT NECESSARY TO CONSIDER NON-MONETARY CONCERNS?ON THE ASSESSMENT OF INDIRECT NETWORK EFFETS. 31 1.2 THE GERMAN EXAMPLE: A DATA-BASED ANALYSIS TO QUALIFY AN ABUSE? ... 32
2-THE FUTURE OF PLATFORM CAPITALISM: IS THERE A HOPE OF CONDEMNING AIRBNB IN THE FUTURE? ... 34
2.1 DOES THE DEFINITION GET CLEARER WITH TIME? ... 34
2.2 WHAT WILL BE THE STATE OF THE MARKET OF SHORT-TERM RENTAL IN THE UPCOMING YEARS? ... 34
CONCLUSION ... 37
LIST OF ABREVIATIONS ... 39
TABLE OF CASES ... 40
p. 6
Introduction
‘Now we are fit to talk about a transformative economic philosophy that alters the rules of the game in an increasingly visible way’ said Elena-Mădălina Vătămănescu and Florina Magdalena Pînzaru1 in 2018, while analysing the new business models of sharing economy platforms as Airbnb and Uber. Indeed, the XXIst century is the century of new technologies, progress, and of a constant growth sometimes difficult to follow. It is characterized, among other things, by the development of the so-called sharing economy which has boomed in 2017. It was born from a need to find peer-to-peer solutions in a capitalist world in constant evolution. This movement and change in the economic model make it difficult to frame, but the sharing economy can be more or less defined as the use of technology to facilitate the exchanged access of goods or services between two or more parties2 . The sharing economy allows actors on the market to share space on a specific platform (with Airbnb, the actors are the hosts), which allows consumers to look for the same product at one and only place. This is the reason it is often referred as the “platform economy”. Indeed, the sharing economy has given rise to the development of various platforms in different facets of our daily lives, from housing to transportation, food or even clothes. Its fast growth is particularly recent and explains the place of venture capitalism in this sector, as investors appear particularly willing to put money in those popular structures, the so-called platforms. Defining a platform is complex, however, after the public consultation of the European Commission on platform regulation3, it resulted a definition of platform as an undertaking operating in two (or multi)-sided markets, which uses the Internet to enable interactions between two or more distinct but interdependent groups of users so as to generate value for at least one of the groups. Certain platforms also qualify as Intermediary service providers.”
Three factors could characterize sharing economy according to Richardson4. First, the central role of digital platforms in the establishment of those companies; second, a peer-to-peer
1 ‘Knowledge Management in the Sharing Economy: Cross-Sectoral Insights into the future of Competitive
advantage’, Introduction, Vol. 6, Editors Elena-Mădălina Vătămănescu, FlorinaMagdalenaPînzaru ©Springer International Publishing AG 2018.
2 Derek MILLER, ‘The sharing economy and how it is changing industries’, The balance SMB, June the 25th 2019. 3 This public consultation about the ‘regulatory environment for platforms, online intermediaries, data and cloud
computing and the collaborative economy’ was held on the 24th of September 2015 as part of the Digital Single Market initiative launched in May 2015.
4 ‘The crazy new world of the sharing economy’, Constantin Bratianu in Knowledge Management in the Sharing
p. 7 functioning, and finally access-based services. Those factors are the basis of the structure of what can be called ‘economy of platforms’, and explain its huge success.
The concept of Airbnb itself, comes from the name of the B&B model (which means bed and breakfast, a type of accommodation where the host offers a room and the breakfast for a certain sum of money) and from how the idea came up. Indeed, the founders rented airbeds on the floor of their San Francisco’s apartment when all accommodations were booked for a design event in the city. Then, the start-up was created in 2008 and is active on the market of short-term rental throughout the world. In 2009, they received lots of funding from business angels and venture capitalists and began to charge both hosts and travelers, which allowed them to make profit and to create the company that it is today. The first model was based on the idea of offering a place to stay by sharing an accommodation or renting one that is not used by the landlords, to make ends meet. It takes place on an online marketplace, which is a virtual place for commercial dealings. It is meant for hosts and travelers to meet, and all of the transactions are processed through the platform owner.
The relevant market on which this thesis focuses is the one of short-term rental in particular. Indeed, Airbnb is not active on the market of long-term rental, at least not in Europe thanks to the limits imposed by Member States. The rental period is usually considered to be lesser than 6 months. However, short-term rental is usually extended from a few days to weeks and is characterized by a high recurrence5. Furthermore, those limits establish some boundaries for online platforms within the EU, but this legal framework does not seem sufficient at the national level nor at the European level. Indeed, against the background that the sharing economy generates positive externalities thanks to network effects arising from this model, it disturbs the economic sphere in a such a way that the regulatory framework does not adapt anymore. The issue of Airbnb is the one of the sharing economy in general: the notion is too broad to find a clear definition and a legal framework to which regulation would apply. It is the stake of the XXIst century. European directives might also have been too large to determine what is a service or not, allowing Airbnb to benefit from this lack of framework and to escape the qualification of a practice restricting competition. Maybe the issue could lie in the emerging model of multi-sided markets that has given rise to market failures and emerging monopolies, which could be
p. 8 an issue for Airbnb, and which makes platforms difficult to catch. This complexity brought the question of online platforms at the center of our digital economies. Thus, those concerns have come to the fore of Member States’ competition policies – something that we definitely see less at the European level. However, the experience of Member States reacting to this business model is progressively about to induce a change in the EU legal order. Indeed, the actual Commissioner Margarethe Vestager recently presented a proposal of regulation6 from the European Commission. Thus, the EU wants to introduce an ex-ante regulation of digital platforms and is thinking about a new tool that could help preventing platforms from tipping the market. A proposal necessary and which brings the issue to the fore.
This phenomenon results in cases all over the world reacting to this aggressive business model, accused of eliminating traditional actors from the market in different countries. Some talk about a ‘history of disrupting traditional sectors’, which reflects how those platforms are causing harm to the traditional sectors. Indeed, the increased efficiencies authorize those undertakings to pass-through value to their customersand supply chain partners7. In France in particular, it has been proved that the model was very harmful and induced the withdrawal of offers from the traditional market of short-term and long-term rental. In Paris, 20 000 accommodations were withdrawn from the traditional market and proposed on the platform8. Consequently, not only it is foreclosing the market, but it also has spill-over effect on the market of long-term rental and it is the reason why rents have been raised that much in Paris.
However, the EU began to answer – and is even more reacting today with the project for a new ex-ante regulation of digital platforms – when observing that some countries brought Airbnb in front of national courts. In France, Airbnb business model has induced a lot of tensions and a need to regulate. It provoked a judiciary saga between February 2018 and December 2019 to determine whether the platform would qualify as a service, and consequently be obliged to request the special authorization that traditional hoteliers have to gather. This judicial relevance could be explained by the fact that France is a priority market for Airbnb in Europe.
6 ‘Single market – new tool to combat emerging risks to fair competition’, proposal for a regulation, European
Commission, public consultation opened until the 8th of September 2020.
7 Derek MILLER, The sharing economy and how it is changing industries, The balance SMB, June the 25th 2019 8 Location « Airbnb » : les règles du code de la construction et de l'habitation à l'épreuve de la directive «
p. 9 The assessment of Airbnb’s action is very touchy because of the complexity of peer-platforms functioning. They are too global to be caught, although it can have the effect of restricting competition in certain countries, like in France. The country asks for a stronger protection of the traditional sector, whereas the EU refuses to tighten the definition of peer-platforms elements and to harmonize the issue throughout Europe. This dichotomy results in a constant opposition between the national authorities and EU legal orders. However, online platforms also generate efficiencies. They are at the core of our modern economies and are likely to maintain in the future. Therefore, the following question can be raised “How do online platform affect traditional business models and markets and what are the limits to what competition law can do about that?”, to study the impact of Airbnb’s business model on the French hotel sectors.
p. 10
Part I – What is the business model of Airbnb ?
1 – How does Airbnb action qualifies on the market of short-term rental?
1.1 How to define the relevant market?
1.1.1 The market of short-term rental
Airbnb is a platform that provides a virtual marketplace for tenants and tourists to meet, so the former can offer its good to the latter. In most of European countries, limits have been imposed and the platform can be held liable for not respecting the competing market. Indeed, Airbnb is supposed to stay in the framework of short-term rental, which means no more than 6 months9. In France, and in Paris particularly, owners cannot rent a place for more than 120 days. This aims at preserving the neighboring market of long-term rental but also to avoid important consequences. Those practices in fact, have resulted in the raise of the monthly rent in cities where the platform is very active, such as Paris. By renting their accommodations through the platform, owners tend to withdraw them from the market of long-term rental. Inhabitants had to move out for other cities where the rent is affordable while facing a lack of offers in the market of long-term rental. Indeed, it had an influence on the prices of the accommodations remaining.
For years, the market of short-term rental throughout the EU has been principally composed of traditional hoteliers, youth hostels, gites and other types of accommodations. The arrival of sharing economy platforms such as Airbnb in 2008 has entirely disturbed the market by imposing the platform as a standard. This is also the result of the development of the Internet in the last two decades which has made the access to the platform easier, thus referencing a lot of accommodations options.
Consequently, at least now, Airbnb is active on the relevant market of short-term rental. However, it is difficult to set a clear geographical market due to the international aspect of the platform which develops without borders. For the clarity of the demonstration, we will focus
p. 11 on the French market of short-term rental, or other Member States markets when specified, or the European market of short-term rental.
1.1.2 The complex assessment of the relevant market in multi-sided markets
Although Airbnb is active on the market of short-term rental, it is difficult to clearly determine the relevant market where the platform is active, when it also has spill-over effects on the market of long-term rental. The reason why it is so difficult to frame the relevant market, is because those marketplaces allow buyers and sellers to connect online. Therefore, it is an international marketplace, and will constitute a two-sided market. Indeed, most of those markets with network effects are two-sided. Actually, taking into account both sides of the market is their success recipe10. Benefits come from the interaction of both sides on the platform. Those specific types of markets are built on asymmetric price structure (which is not in accordance with the model of perfect competition). It means that the price is not the same on both side of the market. An article11 by Economics Nobel Prize Jean Tirole and JC Rochet actually shows that sharing economy platforms chose to apply a specific business model where one side of the market will represent a “profit center”, whereas the other side of the market will either be the “loss” leader or at least financially neutral. A structure which reflects the Airbnb business model, where consumers pay for the accommodation but also for the service, whether the offerors do not have to pay anything for offering their accommodations on the platform. To define and assess the relevant market, all sides of the market and their interactions must be taken into account. However, this reality is difficult to frame because of the international dimension of sharing economy platforms, which is the consequence of the fast growth of internet platforms.
10 Platform Competition in two-sided markets, J.C. ROCHET, J. TIROLE, Journal of the European Economic
Association, Volume 1, Issue 4, 1 June 2003, Pages 990–1029
p. 12 1. 2. Does Airbnb’s action qualify as a service?
According to the Services Directive12 a service “means any self-employed economic activity,
normally provided for remuneration, as referred to in Article 50 of the Treaty”.
Firstly, Airbnb is a platform, and therefore not a physical actor, which makes the notion of employment harder to apprehend. It is therefore necessary to take a deeper look at the other conditions. Secondly, the CJEU has defined the economic activity13 as any activity consisting in offering goods or services on a relevant market. Thus, Airbnb is proposing a number of goods on the market of short-term rental, hence engaging in an economic activity. However, this criterion can already be discussed. Airbnb does not offer goods as it does not offer the accommodations itself. Instead, the platform offers a service by allowing owners and users to meet.
Finally, when Airbnb was still a new startup, remuneration was not relevant for the owners. However, the development of the platform has given rise to a need from the directors to earn money from offering this marketplace. Now, Airbnb takes a specific fee on each booking made. This appears as a remuneration the platform is getting from offering services.
However, the Commission declared that distributive trades including retail and the wholesale of goods and services were considered as services covered by the Directive. It has given rise to a flourishing review from the CJEU to determine if Airbnb was actually a service, and thus fell under the scope of the Services Directive.
First, in 2018, the CJEU ruled that retail trade constitutes a service in the sense of the Directive, the term services being widely interpreted14. Then, in the Airbnb Ireland case15 the CJEU ruled that Airbnb was an information society service; therefore, it falls within the scope of the Services Directive. As a consequence, it does not need to register as real estate agents have to do in France, to get the authorization to practice.
12 Directive 2006/123/EC of 12 December 2006 on services in the internal market, article 4 (1).
13 CJEU, June 16th 1987, Commission v. Italy, aff. 118/85, pts 7 and 8 ; CJEU, September 21st 1999, Albany, C-
67/96, pts 82 to 85.
14 The EU Court of Justice holds in response to a request for a preliminary ruling by the Netherlands that the
definition of "service" in the Services Directive includes retail trade in goods (X BV), Caroline Korteweg, Coen
Drion | Concurrences | N°89965 - January 2018
p. 13 2 – Traditional hoteliers vs Airbnb = does the platform prevent them from competing on the market of short-term rental?
2.1 How does consumers’ trust in the platform reduce the possibility for traditional actors to compete?
2.1.1 The rating system as a trust factor
A recent OECD study16 has shown that the trust consumers had in the platform participated in its success. Airbnb and similar peer-platform actors are more entrusted by consumers than traditional actors on those markets. A lot of factors induced this phenomenon. Although these platforms sometimes have a relative interpretation of consumers rights, the study shows that they generate more trust for three reasons : they are offered at a cheaper cost than traditional services; they are more practical because of the multitude of internet users (45% of internet users use peer-platform markets according to the survey), and offers are more numerous and offered on one and only place. In first position, the low prices of those transactions compared to the price of most of the hotels for example are the main reasons of this success. In second position, the study points out the easy opportunity to make money as a big incentive for users to offer their accommodation through the platform. It is not surprising, since the platform also represents an easy way for middle-income families to make ends meet.
Moreover, this trust phenomenon has increased in the recent years thanks to the development of the rating system which gives user an overview of the quality of the accommodation. This significantly increases the trust they have in the platform, because they feel safer. Indeed, it makes user fell like the platform is controlling the quality of the offers. However, the study highlights a big issue: most of the users (80%) who were disappointed or had issues did not report it to the rating system. Another troubling aspect lies in the fact that platforms have discretionary powers in managing redress between peers. Consequently, it gives a wrong overview of the quality of the accommodation. Indeed, the trust users have in the system is actually based on a wrongful appreciation. Moreover, the issue was raised recently17 that those
16 Trust and power in (peer) platform markets – comments on an OECD survey report on trust issues, European
Commission.
p. 14 platforms are not concerned by consumers’ rights. Even if consumers are quite aware of this problem, they rather give up on their rights than being unable to access the platforms.
2.1.2 Reduction of the number of goods on the market : the consequence of this increasing trust
This rating system, tempered in a discretionary manner, has tough consequences for the traditional accommodation industry. Indeed, a certain amount of goods (accommodations) are transferred to the platform. In some countries, hotels nights are also offered on the platform, but Airbnb has difficulties to impose this model in Europe. The increasing number of offers is parallel to the reduction of the number of goods on the traditional market of short-term rental, as the online marketplace has become the new standard. However, a reduction of the number of goods on a market is correlative to the raise of prices, which contributes, inevitably, to make the traditional sector less attractive for consumers. That is due to the reduction of competition in the market, even if it is necessary to maintain a certain equilibrium between consumers’ and producers’ welfare. Secondly, in France, it results from high commissions in the sector of traditional hoteliers, which contributes to set those prices high.
The large number of goods are transferred to the platform to enhance accessibility to the public, and in some countries, hotels nights are about to be included in the goods offered on the platform too. This results in the elimination of the traditional hoteliers on the market of short-term rental with the creation of a platform marketplace where the same goods are rented. This is why 20 000 housing possibilities have been withdrawn from the market in Paris, which has resulted in a raise of the rent for long-term rental apartments and accommodations. The interdependence of those related markets is the outcome of this situation. Indeed, the action of Airbnb does not only have an influence on the market of short-term rental, but also has a spill-over effect on the market of long-term rental, by contributing to reduce the offer on this market. Consequently, it contributes to raise the monthly rent for inhabitants in the capital18.
18 Location « Airbnb » : les règles du code de la construction et de l'habitation à l'épreuve de la directive «services»
p. 15 2.2 Does Airbnb’s action really answer a specific demand?
2.2.1 Peer-platform: a new business model
The sharing economy is characterized by those new platforms. However, they are active in sectors where traditional actors were already prospering. Therefore, it is important to know if those new undertakings are really answering a new particular demand or if it is acting on a segment of consumers already active in the traditional markets.
The sharing economy platforms, such as Airbnb, propose a larger range of choices. Indeed, they gather all the short term offers. This disadvantages hoteliers for example, as they can only offer their own set of rooms. However, it is also what makes the platform different from those actors: it only allows the reunion of offer and demand at a larger scale. The platform also proposes new prices, from the cheaper to the most expensive type of accommodations, allowing consumers to have an overview of the state of the market.
Furthermore, it has been shown that Airbnb did not only allow the reunion of both offer and demand but was also an answer to a demand for a complement to private revenue. Indeed, average class households are able to rent unused rooms or properties to collect money when it is difficult to make ends meet. This phenomenon is called the social finality of the model, giving modest and poor families the possibility to earn a rent from an accommodation they do not use. However, not only does the platform allows a diversification of revenue possibilities, but it also creates kind of a synergy between the effects that result from the model19. Indeed, this two-sided market also allows the collection a commission through the interaction between offerors and consumers, to collect a large number of data for third parties and to create value through the users proposing new offers on the platform. Ronald Coase called it the externalities of
tasks20, meaning that everyone actually brings some more value.
Moreover, peer-platform markets are by nature disruptive21, meaning that they aim at correcting a bug or a politic, so the scope is to propose a new way of answering an already existing demand.
19 Pourquoi les plateformes multifaces triomphent sur le marché, S. Tran, Harvard Business Review – France, 7th
May 2019.
20 The firm, the market and the law, R.H. COASE, Chicago, University of Chicago Press, 1988.
21 Peer-to-peer accomodation networks : pushing the boundaries, Sara Dolnicar, © Goodfellow Publishers Ltd,
p. 16 That is what Airbnb has been doing by developing new conditions for transactions, an easier access to a lot of accommodation information, and allowing a new range of lower prices.
2.2.2 A demand actually created by the platform
The issue resulting from Airbnb’s action is actually deeper. Its functioning is actually based on the analysis of data collected from its competitors. Those data are used to make the platform the standard based on user’s anticipations22. However, it is not the case for actors of the traditional market.
Moreover, those platforms set the transactions’ conditions and have full discretion concerning the way they manage redress between peers23. This is why the OECD refers to a possible “blind faith” that those platforms are developing. The platform collects data thanks to platforms’ users and therefore, could use control over data about supply and demand side of the market to maximize the market expansion and allow a higher volume of transactions through different mechanisms. By acting this way, platforms are pushing sellers to be on the online marketplace if they want to reach their consumers, because the platform, Airbnb here, has been seen as the standard for renting accommodations. For this scope of setting a standard, the platform will use first, the ranking system, because as I have demonstrated before, this is a trust enhancement system for consumers. Moreover, the pricing guidance and the personalized marketing and searching results act as strong tools for the platform to set the model as a standard. There, cookies and third-party tracking will be particularly useful.
It has additionally been shown that hosts on the platform used consumer segmentation to target the consumers more effectively 24. Indeed, the Airbnb marketplace could be divided in two segments that results from the number of offers. On one side consumers looking for shared rooms, on the other side consumers seeking to rent an entire place. Using those data, hosts are capable to target more actively specific consumers that could be interested in their offers. Thus, they are more able to book their accommodations. It raised an issue which is the one of Airbnb
22 Trust and power in (peer) platform markets – comments on an OECD survey report on trust issues, European
Commission.
23 Ibid [21], p.7.
24 C. LUTZ, G. NEWLANDS, Consumer segmentation within the sharing economy: The Airbnb case, Journal of
p. 17 in general, does the platform ‘plays’ with users’ anticipation to make it the standard in term of short-term rental? It will be necessary to look at the increasing returns of adoption25, meaning the development of particular services by the platform. But the efficiency of this practice tends to its visibility, reason why it should be easily caught by competition authorities in theory.
Part II – EU law beyond competition: does Airbnb benefit from a
tense legal framework?
1 – Is there a lack of regulation at the EU level, necessary to ensure equality between economics operators?
1.1 The services directive: too broad to be effective?
1.1.1 The content of the services directive
The services directive26 of 2006 aimed at defining what is called a service, to ensure a proper enforcement of the freedom of services in the Internal market. It states that a service ‘means
any self-employed economic activity, normally provided for remuneration, as referred to in Article 50 of the Treaty’. However, experience has shown that this definition, and the directive
more generally, seems too broad to ensure an easy and effective qualification of a service. The application of the services directive was more complex than expected when it had to deal with disruptive economy. On January the 30th of 2018, the CJEU ruled that the term “services” includes the retail trade in services and applies to domestic situations. Consequently, there is no need for a cross border element anymore. This ruling allows the acceptance of Airbnb as a services platform.
25‘Protecting or diffusing a technology platform: tradeoffs in appropiability, network externalities and
architectural control’, M. A. SCHILLING, in Platforms, markets and innovation, edited by A. Gawer, 2009.
26 Directive 2006/123/EC of the European Parliament and of the Council of 12 December 2006 on Services in the
p. 18
1.1.2 The outcomes of the services directive in EU legal systems
The conclusion of the CJEU on 30th of January 2018 had consequences in European legal orders. In France, it has recently resulted27 in a stay of proceedings on the 17th of January 201928. The French tribunal was unable to rule without an examination of the case by the Court of Justice of the European Court of Justice (CJEU). Indeed, the Cour de cassation raised six prejudicial questions to the CJEU, concerning the application of the directive to the platform and consequently, the regime of the change of use of an accommodation from usual rental to touristic rental. In fact, a civil society was rented a good for short-term rental purposes through Airbnb. However, it needed to change the use of the accommodation in accordance with French law (and particularly article L.631-7 of the code de la construction et de l’habitation). Today, the answer is still pending as the Court is waiting for the answer of the CJEU concerning the compatibility of this provision with the services Directive and the freedom of services.
However, the European Court of Justice has a blurred definition of a service and it rarely imposes fines on Airbnb. In January 2019, the CJEU decided that the Belgium government applied unjustified restriction to the platform. Indeed, Belgium strengthened the conditions for individuals to rent their accommodations (for examples, privacy curtains at each window, a power switch at the entrance of each room). The European Commission denounced a wish from the government to restrict the rent of accommodations through the platform. They talked about a lack of proportionality in the measures taken by the government, although it is a European principle.
1.2 Is the e-commerce Directive sufficient to frame the definition of a service in a virtual environment?
1.2.1 The content of the e-commerce directive
The e-commerce Directive29 came into force on the 8th June of 2000 (transposed in August 2002 in UK law) to provide a clearer and exhaustive framework for information society services
27 Following a judicial saga since 2017, see Tribunal de grande instance de Paris, 7th March 2017, n° 15/12082 28 Cour d’appel de Paris, ch. 2, 17th January 2019, n°18/14327.
29 Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of
information society services, in particular electronic commerce, in the Internal Market ('Directive on electronic commerce').
p. 19 (ISS). It means that it concerns “services normally provided for remuneration at a distance, by
means of electronic equipment for the processing and storage of data and at the individual request of a recipient of the service30”. Even if this legal text was enacted ten years before the
creation of the platform, it would certainly apply to Airbnb. It supposes of course, that Airbnb has been recognized as an ISS and thus provides services. Indeed, the directive applies to online intermediaries, defined through the activities of mere conduit, caching and hosting.
According to the European Commission31, the main objectives of the directive were to remove obstacles within the EU internal market for cross-border online services, allowing more legal certainty, a neutral and balanced legal framework, and finally developing competition between EU service providers. The 3rd article of the Directive is particularly important in our case, because it states that no Member States (MS) should put obstacles to the freedom to provide information society services providing from another MS within the EU. Some exceptions still exist and are conditioned to the respect of the proportionality principle.
1.2.2 The Airbnb Ireland case: the complex application of a non-exhaustive directive
The e-commerce directive considers that an activity can be regarded as hosting only if the intermediary – in this case, the platform - does not have either knowledge or control over the content offered. It means that it is rather a technical intermediary with a passive role. This has been confirmed by the CJEU in two cases, Google/LVMH 32 and L’oreal/Ebay33. The issue is that Airbnb can be seen as an active actor because it regulates the offer and has an impact on how accommodations are offered through the platform. Thus, it would escape the qualification of ISS and fall outside the scope of the directive. This question led to the Airbnb Ireland case34 last December, for the CJEU to rule on the possible qualification of Airbnb as an ISS.
Two questions were asked by the French Court to the CJEU. First, does the provision on the freedom of services in the e-commerce directive apply to Airbnb Ireland UC, meaning when
30 E-commerce Directive, Thomson Reuters – ‘Practical law’, 15th July 2002, Resource ID 8-101-7661.
31 EU Regulatory framework for e-commerce, Denis Sparas, WTO Workshop Geneva, 17th of June 2013,
Communication of the European Commission.
32 C-236- 238/08, Google v. LVMH, 23rd of March 2010, European Court of Justice. 33 C-324/09, L'Oreal v. eBay, 12th of July 2011, European Court of Justice.
p. 20 the services are provided through a platform hosted in Ireland? Second, and consequently, do the rules that apply to real estate agent in France (set in the 1970’s Hoguet law) could be opposed to Airbnb Ireland when it is acting within the French territory?
The CJEU answered that the services offered by the platform were information society services to which the e-commerce directive applies. Indeed, Airbnb only allows potential tenants to meet with landlords that offer short-term rentals possibilities. Thus, the CJEU ruled that it was only an intermediation service. Moreover, it means that the notion of ‘country of origin’ will apply when determining where the platform should be liable. To the second question, the CJEU concluded that the provisions of the Hoguet law for real estate agent (authorization regime, ect, …) cannot apply to the platform because the French government did not notify the Commission before, whether the services directive specifically mentions it.
This decision seems very singular. Indeed, Airbnb is more than active and has a real impact on the conditions to which the accommodations are offered through the platform. It gives indications on the rent’s price, collects payment, provides help to create ‘experiences’, between other things. Once again, it seems like the CJEU chose the easy solution to avoid dealing with the issues that comes from the international development of Airbnb, and of sharing economy platforms in general. If the CJEU strictly applied the wording of the directive, associated with the European Commission’s communications, it surely would have concluded that Airbnb was not only an intermediate between tenants and landlords, but an active actor on the platform that orientates the way services are provided.
To conclude, this situation could be explained by the fact that the European Union is often slow – even reluctant – to regulate certain sectors that needs to be regulated. Indeed, it is also the EU’s task to ensure that the market remains competitive. However, this goal cannot be attained without a regulation that keeps the short-term rental market competitive despite the fast rise of alternative way of renting them, that are platforms. This is why, on the 2nd of June, a new proposal for a regulation has been presented35. It aims at proposing new rules for the control of digital platforms and a new tool to prevent platforms from tipping the market is in discussion. It would be based on a market investigation to determine which behavioral or structural remedies could be better adapted to the situation.
35 ‘Single market – new tool to combat emerging risks to fair competition’, proposal for a regulation, European
p. 21 2 - Does French protectionism therefore restrict the freedom to provide services, and competition, by an excess of regulation?
2.1 Airbnb, service or regulatory intermediary ?
On the 19th of December 201936, the CJEU ruled that Airbnb is actually an intermediation service and not a service. Thus, it does not need to register as a real estate agent and to obtain a professional card37. It does not have to follow the authorization process that hoteliers and real estate agents have to follow when proposing short-term rental offers. The issue in this case was to determine whether Airbnb was an ISS or a regulatory intermediary in the sense of the Hoguet law. As a consequence, it would have meant that Airbnb, as for every other legal intermediary, needs to submit a request to the French government to get the authorization and the professional card to exercise. The provisions of the Hoguet law have been considered as very restrictive and in 2017, therefore the Ahtop association (association for professional tourism) tried to apply the provisions of this law to the platform that it considered was acting as an estate agent without license. Thus, disfavoring traditional hoteliers. However, the Court did not consider that in the Airbnb case, it does not appear that it forms an integral part of an overall service whose main component is an ‘accommodation service’38. Indeed, the platform is offering a list of available short-term accommodations that matches the criteria set by the user, which allows landlords to book more easily those accommodations with potential renters. For this reason, the CJEU ruled that this service cannot be regarded as merely ancillary to an overall ‘accommodation service’39.
2.2 What are the consequences of French law on estate agents regarding the competition between Airbnb and the traditional hoteliers?
French regulation has sensible consequences for the short-term rental market since it might cause the elimination of smaller competitors who cannot compete. Indeed, the first step taken was the ALUR law40 in 2014 (enacted in 2016) and the ‘Airbnb decree’, both meant to precise
36 Case C-390/18, Airbnb Ireland, 19th December of 2019, European Court of Justice.
37 “Aibnb n’a pas besoin de carte professionnelle”, M. THIOYE, Dalloz actualite Europeen – IP/IT et
Communications – Immobilier, 21st of January 2020.
38 A regulatory conundrum in the platform economy, case C-390/18 Airbnb Ireland, A. CHAPUIS-DOPPLER, V.
DELHOMME, 12 February 2020, European law blog.
39 Case C-390/18, Airbnb Ireland, 19th December of 2019, European Court of Justice, para 53-54.
40 LOI n° 2014-366 du 24 mars 2014 pour l'accès au logement et un urbanisme rénové, called loi ALUR for a better
p. 22 the notion of the ‘digital Republic’ and the conditions for the registering procedure41. This decree also made it clear that Airbnb is liable for the hosts and has to make sure that they respect the conditions set in the law. The platform should follow a strict cooperation with public powers. Later, smaller platforms and undertakings became more vulnerable with the enactment of the Elan law42 of 2018, that has strengthened the rules and the sanctions around the accommodation’s rental activity. On the other side, Airbnb is still thriving and has not been impacted in a large sense by the laws.
Moreover, Airbnb is still playing with European law. The local authorities are the only one who can strike the balance between tourist accommodation and housing. However, the services directive could be a tool in the platform’s favor. Indeed, it says that an authorization regime, as set in French law, could only be applied if justified by a general public interest reason, and with due respect of the proportionality principle. Consequently, the platform could escape the registration with the French authorization regime, whereas traditional hoteliers have to follow those rules. Not only do those provisions reduce the competition between the platform and traditional actors - which enables it to be dominant on the short-term rental market - but they also represent a bigger risk for competition. Indeed, Airbnb could acquire a monopoly.
2.3 Is this constant fight between French regulation and Airbnb allows a right control of the platform?
In 2018, several proceedings have been led to determine if French law, in particular the Loi
Elan (November 2018), was in accordance with the services and the e-commerce Directives.
This law enacted sanctions for the lack of registration number, which did not exist before. Those trials resulted from a heavy lobby from the Ahtop association and the hotel industry43 to enact more regulations against the huge development of Airbnb. The government answered with the 120 nights limitation and a tourist tax that must be added to the rental price. However, this was not enough. In addition, those provisions made the landlords more liable, since they could be
41 For example, the most criticized face of the law is the limitations it imposes on accommodations, which can
only be rent up to 120 nights not to be considered as an estate agent activity, except in cities of more than 200k inhabitants where they always need to register.
42 ‘LOI n° 2018-1021 du 23 novembre 2018 portant évolution du logement, de l'aménagement et du numérique’,
called loi ELAN, law for the development of housing, interior design and digital.
43 ‘Airbnb and its second largest global market; can France rein the travel giant in?’, A. LEDSOM, Forbes.com,
p. 23 fined, whereas it did not change the platform’s situation. On March the 5th of 2019 the city hall of Paris brought Airbnb to the Court and lost the case for lack of proof. Indeed, the locals which are meant to be rented for a short-term, repeatedly, need to be declared so they can get a registration number, necessary to exercise their activity. However, 84% of Airbnb’s accommodation do not respect the French regulation on this point.
The issue has particularly raised the point that this regulation makes the situation worst for traditional actors because it imposes more obligations they must follow to stay on the market. Moreover, it favors the accommodation platform. In this article44, the author points out the fact that French law is so protectionist that landlords would feel “more protected” using Airbnb. Indeed, those last laws and decrees happened to favor tenants. They have been granted more rights regarding, in particular, the renewal or the termination of their contracts, whereas owners have been in a situation forbidding them from terminating those contracts “unless they can make a formidable case for why they need their apartments back”. This is a proof, that those provisions seem too strict to allow a fair framework for the short-term rental market.
Part III – Is competition law sufficient to condemn Airbnb regarding
the complexity of the market on which it operates?
1 – Could we apply article 102 FEU to a possible dominant position on the short-term rental market?
1.1 Does the platform have a dominant position?
1.1.1 The reality of the market: a multitude of competitors
Article 102 of the Treaty on the Functioning of the European Union (TFEU) prohibits any abuse of dominant position, affecting trade between Member States. The first step towards a possible sanction of this position lies in the difficult task of defining the relevant market. Indeed, there can only be an abuse of dominant position if the undertaking has an impact on a relevant market - also on the trade between Member States. The so-called SSNIP test is generally used for this
p. 24 scope. Its purpose is to determine the smaller and the more precise relevant market where the undertaking could apply a “small but significant non-transitory increase in price”. It means that if the potential monopolist could increase its prices by 5% for a period around 12 months on the market without customers switching for another market, thus, we have found the relevant market. However, the SSNIP test is difficult to apply to the sharing economy model. Indeed, can the SSNIP test be applied when an undertaking is active on different market at the same time?
When a platform is tipping the market, it is more likely to become a monopoly45, because of the strengthening of its dominant position. However, one of the issues relates to market shares. Airbnb is particularly present in the short-term rental market. Nevertheless, it is not the only undertaking active in the accommodation sector. We can name HomeAway, Couchsurfing, Booking, as examples. In addition, sharing economy platforms compete side-to-side with undertaking which do not use platforms but are active in the same sector through the internet. Some like Booking and Expedia are even more difficult to frame because of the very diverse range of services they offer. Consequently, the market is shared with a multitude of other actors (founding principle of the perfect competition model). Thus, it is difficult to determine if the platform is dominant looking at market shares.
1.1.2 The complexity of a two-sided market
The platform economy is characterized by virtual marketplaces where buyers and sellers can connect. This is why we talk about a two-sided market. This kind of market structure is based on an asymmetric price structure. All sides of the market and their interactions must be taken into account to define and assess the market. However, it is particularly difficult to assess the anti-competitive effects of a practice in multi-sided markets. Indeed, a practice that restricts competition from one side (ex: predatory pricing) may generate pro-competitive effects on the other side.
45 Oliver Budzinski & Annika Stöhr (2019) Competition policy reform in Europe and Germany – institutional
change in the light of digitization, European Competition Journal, 15:1, 15-54, DOI: 10.1080/17441056.2018.1555942.
p. 25 In addition, those markets are designed by peer-platform, creating a lot of value on the market. Thus, their action result in some efficiencies which makes it even more difficult to qualify a restriction of competition. However, some platforms and especially Airbnb can have negative networks effects due to its implication in multi-sided markets. The issue was raised as an example in the Google search case46 (in relation with the article 7 of the Regulation (EC) 1/2003). Sharing economy platforms can have indirect network effects and positive feedback effects on two-sided platforms, which creates barriers to entry and to the expansion of the market. Indeed, those online marketplaces are so powerful that it makes it more difficult for new platform to build themselves on the market. This issue is often linked with the “chicken and egg” problem. It means that a platform needs a multitude of suppliers to grow but to get a multitude of suppliers they need to attract a multitude of buyers47. However, buyers answer to the suppliers’ offer. Moreover, some other factors also have an impact as trust in the rating system or the growth path, which is difficult to handle in front of those giant platforms.
1.1.3 What are the consequences of spill-over effects in the platform economy?
Airbnb is the main example of what platform capitalism result in. The issue also lies in the influence that it has on other markets. Indeed, in Paris or Barcelona for examples, the number of properties rent through Airbnb has exploded, and some inhabitants even bought second, third properties for the scope of renting it through the platform. Thus, it has spillover effects on the market of long-term rental because those behaviors (on the short-term rental market), are the direct consequence of Airbnb’s action. Indeed, it leads to the raise of the rent and makes it more difficult for inhabitants to find affordable housing. An issue that is sometimes referred as the ‘Airbnb effect’48. It is the issue of the sharing economy and multi-sided markets in general, that reduce the number of goods in neighboring markets.
46 Case AT.39740 - Google Search (Shopping) [2017], European Court of Justice.
47 ‘Network effects aren’t enough’, A. HAGIU, S. ROTHMAN, Entrepreneurship, Harvard Business Review, April
2016.
48 ‘Unfairbnb: How online rental platforms use the EU to defeat cities’ affordable housing measures’, K. HAAR,
p. 26 However – and this is the main reason why sharing economy platforms are not caught – some spillover effects are positive. As it generates efficiencies, the platform cannot then be fined for restricting competition. Indeed, in a 2018’s paper49, M. Alyakoob and M. Rahman dealt with the multiple spill-over effects that platforms as Airbnb can have. They show that Airbnb has an impact on the growth of complimentary local services, focusing on the example of restaurants. As a matter of fact, Airbnb allows owners to offer accommodations in cities close to the main touristic attractions that are not always privileged by tourists in holidays. Thus, it indirectly induces a growth for businesses in those cities, because tourists are going to restaurants - among other things – in this neighboring city.
1.2 Is there an abuse in Airbnb’s conduct?
1.2.1 Airbnb’s action: an exclusionary conduct?
The issue with the sharing economy is that all sides of a market should be taken into account in determining the impact of a conduct. Airbnb is raising prices, but in the long-term rental market. In fact, the platform is giving indications on prices to short-term offerors, without strictly determining the price that should be applied. Thus, it is not directly exclusionary this way. However, Airbnb is a very attractive platform and influences the accommodations’ owners to propose their goods on the marketplace. Thus, it leads to a rarefication of accommodations on long-term rental market. However, when there is less offer on a market, the prices increase. Consequently, it excludes competitors that cannot compete by setting the same prices. This results in a possible exclusionary conduct which is not directly the consequence of Airbnb’s conduct since the platform is not active on the long-term rental market.
The question of predatory pricing could also be raised. Indeed, some accommodations’ prices are set particularly low. However, Airbnb only gives an indication and does not set the final price. It might then be difficult to catch Airbnb - once again – under this plea. It is not the case for all types of housing, however it could also be a reason why consumers could switch from traditional hoteliers, to the offers proposed on the platform. Indeed, they often put prices higher because of a set of constraints. This price structure and strategy cannot be reproduced by
49 ‘Shared Prosperity (or Lack Thereof) in the Sharing Economy’, M. ALYAKOOB, M. RAHMAN, SSRN
p. 27 traditional actors, which allows to say that this action could qualify as an unfair pricing practice. Indeed, private owners who offer their accommodation on the platform do not have to pay for the same costs – administrative for example – than the hoteliers. Furthermore, the issue has been raised that not only platforms as Uber, Lyft or Airbnb were involving in predatory pricing behaviors – the latter being less concerned since it only uses ‘smart pricing’ – but they also engage in regulatory arbitrage50 to avoid being fine in case an anticompetitive conduct was found51.
1.2.2 Airbnb’s action: an exploitative conduct? On the application of article 102 (a) TFEU
Article 102 TFEU (a) states that any abuse ‘directly or indirectly imposing unfair purchase or
selling prices or other unfair trading conditions is prohibited’. Thus, it refers to the notion of
exploitative conduct. However, this abuse is complex to recognize. Indeed, it supposes the application of unfair and/or excessive prices, two particularly broad notions. It has for scope to exploit consumers or suppliers. However, this is also a relevant question when it comes to those powerful sharing economy platforms. Indeed, the specific issue of the economic dependence often appears between the giant platform and small sellers that represent a big part of the users. They may depend on the platform to rent their accommodation – and indirectly to make ends meet for some of them. Indeed, the platform allows them to reach consumers.
However, the issue was raised in an article52 evaluating the possible outcomes of the Airbnb
Ireland decision53 when it was still a pending case. Christoph Busch thus developed the topic of the Uber test, an analysis method of the effects of a platform to determine its nature. In this case, if the platform was an information society services (ISS). The question was to determine if Airbnb would be qualified as an ISS or as an accommodation services provider just as Uber was a transportation services provider. This would change the whole Airbnb regime because it would have to align on the French authorization regime. The test was based on whether a platform –Uber – could be considered as just an intermediary, or a real market maker.
50 This is also known as ‘forum shopping’.
51 ‘What Uber, Lyft, AirBnB, Microsoft, and IBM Teach Us About Competitive Markets’, I. Welsh, August 2016. 52 ‘The Sharing Economy at the CJEU: Does Airbnb pass the ‘Uber test’? Some observations on the pending case
C-390/18 – Airbnb Ireland’, C. BUSCH, August 2018.
p. 28 Two elements need to be analyzed for this scope. First is the platform the reason why users can offer those services – meaning that without the platform landlords would not have to offer their accommodations, and customers to rent for short-term periods? Second, does the platform have a control over quality and prices of the offers proposed on the marketplace – meaning a decisive influence? This second issue is particularly important, since it can induce an exclusive power from the platform towards users that would not meet the criteria. However, exclusive behaviors prove an abuse of dominance situation. The issue is even worst in the case where the platform ticks the two boxes: since customers and suppliers can only interact thanks to the platform, this could result in exploitative abuses of suppliers.
Thus, the author first recognizes that Airbnb is a market maker since it allows a new range of short-term rentals to be offered on the platform whereas it was not the case before. It creates new supplies. However, Airbnb does not have a decisive influence since it does not determine the remuneration for the services offered – even if it has an influence on other criteria like the cancellation policy. Consequently, it is still difficult to assert that the platform has both exclusive and exploitative behaviors, thus abuse of its dominant position.
2 – Does the relation between Airbnb and its peers could allow the application of article 101 TFEU ?
2.1 Airbnb and the users: a vertical agreement?
It results from the previous developments that it is not yet possible to establish the abuse of dominance of Airbnb in the French market. However, article 102 TFEU is not the only legal basis that could be useful, since the relations between the platform and the landlords could be qualified as a vertical agreement. Thus, article 101 TFEU apply to the situation if the conditions are met. The question of the qualification of a vertical agreement between Airbnb and its peers – the users offering accommodations – has already been raised54. Indeed, a vertical agreement is defined by the Regulation on vertical agreement55 as an agreement or a concerted practice
54 ‘The application of EU Competition Law tot he Sharing Economy’, J. SAFRON, Stanford-Vienna European
Union Law working paper No 27, 2018.
55 Commission Regulation 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the
Functioning of the European Union to categories of vertical agreements and concerted practices, Official Journal L 102, 23.4.2010, p.1-7
p. 29 between two or more undertakings at different level of the production or distribution chain. It should concern the specific conditions that applies to consumers for the sell or purchase of goods and services. Following this regulation, Airbnb is at the top of the distribution chain, providing suppliers – the peers – on the downstream market with an intermediary platform where they distribute services. Moreover, it is even discussed whether the suppliers are not part of a horizontal agreement to possibly qualify a ‘hub-and-spoke’ situation – an issue that will be developed later. The only thing is that, if we can assume that they agree to contract with the platform56, they don’t have an agreement with other suppliers. Thus, it is not concretely possible to qualify a horizontal agreement between the peers.
However, in the Interstate Circuit57 case, the American Supreme Court ruled that there might be a horizontal agreement between downstream actors if it was organized by the actor in the upstream market. This horizontal agreement qualifies when there are several agreements at different steps of the distribution or production chain with the suppliers/retailers, acknowledging that other actors accepted the same contract conditions58. Nevertheless, all offerors agreed to Airbnb’s terms and conditions to start using the platform as an intermediary for distributing services. The situation fulfills the conditions and could allow the qualification. Indeed, Airbnb is in the upstream market, proposing its platform as an intermediary. In the downstream market, the platform concludes contracts with the private owners so they can offer their accommodations through it. Moreover, suppliers know that there are other suppliers at the same level, concluding the same contracts – at least they agree to the same conditions. However, Airbnb’s offerors do not really exchange information nor agree on the prices – even if they somehow align in the end. This is more the result of the market and its fluctuations and of price indications.
Furthermore, this is not always the outcome, since the Southern District of New York already qualified a ‘hub-and-spoke’ situation between Uber drivers, that were accused of price-fixing59. A ‘hub-and-spoke’ is an agreement that could be defined – following the Meyer v. Kalanick Supreme Court judgement – as ‘a situation in which there is a firm (the hub) that organizes
56 Indeed ‘the form in which that intention is irrelevant as long as it constitutes a faithful expression of the parties’
intention’, para 25 of the Guidelines on Vertical Restraint (2010/C 130/01).
57 Interstate Circuit, Inc. v. United States, 306 U.S. 208, 227 (1929).
58 Developped by M. ANDERSON and M. HUFFMAN in the article ‘The sharing economy meets the Sherman
act: is Uber a firm, a cartel, or something in between?’, Columbia Business Law Review, 2017.
p. 30
collusion in the upstream or downstream market (the rim) by interacting with each rim firm individually (spokes) in order to preserve its market share from shrinking in some way or maintain high profits’60. Consequently, it belongs to the categories of horizontal agreements involving exchange of information.
2.2 Hub-and-spokes and the interplay between upstream and downstream
market: a possible condemnation of Airbnb for restrictive vertical agreement?
Airbnb has generated more than €2 billion revenue in 2018, even more in 201961. However, the regulation on vertical agreement specifies that the block exemption will not apply if the connected undertakings within the association has a turnover worth more than €50 million. The turnover generated by the platform is largely exceeding this level. Thus, setting aside Airbnb’s market shares since it is more than complex to find them exactly, it would be necessary to look at the possible restriction that the platform or the buyers could realize. The question of whether Airbnb practices hardcore restrictions could be raised. Indeed, the limitation of the supplier’s sale price is a restriction that belongs to the blacklist of the block exemption regulation (BER). However, Airbnb only gives indication on the prices that suppliers should set according to certain criteria. Thus, it is not imposing a price. This is in the line of the discussion around the Airbnb Ireland case, wondering whether we could really say that Airbnb has a decisive influence on the control of quality and prices of the offers.
Furthermore, although it has never been judicially discussed yet, the question could be raised of Airbnb and its suppliers constituting a ‘hub-and-spoke’ agreement. Thus, the platform could be condemned as part of a vertical agreement, and the suppliers as part of a horizontal agreement. Indeed, the wording cannot be found in EU legislation but is considered so far as a horizontal collusion. However, at least in France, the collusion between suppliers exchanging information via Airbnb in the upstream market to set a price strategy is unlikely to be qualified. This could be where lies the limit between private owners activity and business operators.
60 How Meyer v. Uber Could Demonstrate That Uber and the Sharing Economy Fit into Antitrust Law’, N.
ANDREW PASSARO, 2017, highlighted in the case Spencer Meyer v Travis Kalanick, 15 Civ 9796; 2016 US. Dist, at 11-12.
61 ‘Airbnb’s $2 billion French market under pressure from lawmakers’, H. FOUQUET, Bloomberg.com, 29th of
