Page content

The Pros and Cons of the Sharing Economy

The Swedish Competition Authority organized an international seminar on sharing economy in Stockholm on the 11th of November 2016.

2016 year's Pros and Cons seminar addressed the issue of Sharing Economy.

Contributors (from the left): Ioannis Kokkoris, Vera Demary, Alexis Walckiers, Jacob Schaumburg-Müller, Mike Walker, Tommy Staahl Gabrielsen, Ariel Ezrachi, Arno Rasek, Sofia Ranchordas, Dan Sjöblom, Joshua Gans, Tara Koslov.

Summary of the seminar

The Pros and Cons seminar commenced with the Swedish Competition Authority’s Director, Dan Sjöblom welcoming the participants.

The moderator, professor Tommy Staahl Gabrielsen (Bergen University) took to the floor and introduced some of the seminars central themes. According to Staahl, the topic of the sharing economy is a very current one, affecting tax law, employment law, labour laws, and consumer protection. One of the major issues is whether actors in the sharing economy should be treated as normal service providers or as platforms in multisided markets. Another major issue is whether the sharing economy is truly about sharing or about pure commercial activities. Moreover, what is the best way to approach the sharing economy, through regulation or a laissez-faire approach? The professor concluded by remarking that most actors are currently struggling to understand the sharing economy. The European Commission and national competition authorities, including the Swedish Competition Authority, are issuing guidance and reports on the topic.

The seminars first paper was presented by Vera Demary (Cologne Institute for Economic Research). The paper’s focus is the relationship between the sharing economy and innovation. Demary started by defining the sharing economy narrowly, as including all economic activities among individuals that use assets instead of owning them. Such economic activities are characterised by digital, peer-to-peer intermediation. Thereafter, Demary defined innovation as the transformation of ideas into new and/or improved products, services or processes. Two types of innovation were identified: firstly, disruptive innovation is about overthrowing existing structures, business models and consumer bases in exchange for a novel way to cater to consumers’ needs; secondly, business model innovation is about the discovery of a fundamentally different business model in an existing business.

Demary pointed out that patents are not very important for the sharing economy. From a micro perspective, what matters is being the first actor bringing the peer-to-peer business model to the market. Often this is nothing more than cloning ideas and applying the model to new products, services, and locations. Some technological innovation may be present, often in the form of software to improve customer experience or simplify doing business.

According to models of disruptive innovation, many sharing economy platforms are still simply competitors of incumbents, not disruptive innovators. Analysing effects on competition from the competitive pressure exerted by sharing economy platforms on incumbents, a competition authority ought to focus on the peer-to-peer business model, which is characterised by network effects and economies of scale. Platforms can grow quickly and acquire a critical mass on both sides of the market, hence becoming big without needing to invest in assets. Market entry, or threat of it, fosters competition, pushing incumbents to innovate too.

When it comes to analysing competition between platforms, Demary focused on a few decisive factors, such as that the competition is for the entire market and the result depends on the ability to innovate and to raise capital for innovations. Innovation is important for differentiation, but some entrants find that a good alternative strategy to innovation is to hold back and let the leader innovate.

The paper’s commentator was Alexis Walckiers (Belgian Competition Authority). The commentator’s thesis was that there is nothing new to the sharing economy as individuals have always been exchanging. Walckiers expressed the opinion that reputational mechanisms can address most regulatory concerns on the sharing economy and in some cases (taxis, hotels) can even provide a better level of protection than existing laws. Nevertheless, reputational mechanisms may not be the appropriate tool to address health problems and financial regulation. Additionally, there are examples of platforms that are specifically designed to break the law, for instance music and film sharing platforms.

The role of competition authorities in the sharing economy is threefold according to Walckiers. Firstly, competition authorities need to be aware that platforms are prone to monopolisation, have high fixed costs and how marginal costs, are characterised by indirect network externalities. Secondly, competition authorities can engage in advocacy, to help explain the benefits of increased competition, foster review of existing regulations when they have become less useful and bring in the consumer perspective. Thirdly, competition authorities should manage expectations and acknowledge the limits of their mandate: they cannot weigh competition concerns against other motives for regulatory intervention; often only elected politicians can do that.

Professor Ioannis Kokkoris (Queen Mary) presented the seminars second paper. Kokkoris focused on competition enforcement challenges in the sharing economy. Firstly, the winner takes often the whole market; strong network effects create a tipping point. There are often barriers to multi-homing due to the inability to transfer reputation over platforms, resulting in locking in suppliers. Moreover, there is often an asymmetry of bargaining power between the platform and the suppliers. Secondly, there may be contracts that reference rivals’ contracts, as in the Apple iBooks case. Thirdly, MFN clauses are problematic, as suppliers cannot offer better prices on their own websites than the ones offered on the platform or on a competitor. Fourthly, some important mergers escape scrutiny, as they do not meet the notification thresholds, despite the platforms significant potential value. Lastly, price-fixing is problematic irrespective if it happens digitally. However, Kokkoris is sceptical about surge pricing as a form of a hub-and-spoke cartel.

Mike Walker (Markets and Competition Authority) was the paper’s discussant. Walker called for a non-interventionist approach to the sharing economy as from a competition law perspective it was not clear that these markets pose any problems. The disruption of traditional markets should be viewed as a good development, something evidenced by how incumbents are engaging in lobbying to preserve their interests. Agreeing with Kokkoris, Walker found it difficult to see pricing on online platforms as a form of hub-and-spoke cartel, since many of the platforms own no patents, are easily replicable, increase consumer choice and reduce prices. Algorithms may be used to collude online, but are only ways to facilitate cartels and make them more stable, not ways to create cartels. Moreover, algorithms may not be able to capture qualitative aspects of products and services.

Ariel Ezrachi (Oxford University) presented the seminars third paper, entitled “Can computers collude and discriminate?” Ezrachi presented two possible theories of harm regarding the anti-competitive conduct of sharing economy platforms.

The first theory of harm had to do with Article 101.1 TFEU collusion. Three examples were given. Firstly, online platforms could use an algorithm to facilitate a cartel. This does not pose any challenges for enforcement. Secondly, algorithms could be used, intentionally or unintentionally to align prices, leading to an artificial market price through a hub-and-spoke cartel. Thirdly, tacit collusion could become much more powerful and effective due to advances in online analytics. This tacit collusion might merit an effects-based approach. The second theory of harm regarded what Ezrachi called behavioural discrimination. Machines with artificial intelligence combined with big data and big analytics can be used to increase dynamic pricing and lead to almost perfect behavioural price discrimination, leading to users being locked in in ecosystems that reduce their freedom of choice.

Based on the above, Ezrachi considers that competition law will face conceptual and practical challenges in the near future and will need to identify the adequate level of intervention, taking into account privacy laws and customer empowerment.

Jacob Schaumburg-Müller (Danish Competition Authority) was the paper’s discussant. He agreed with the Ezrachi on most points but expressed also a hope that the sharing economy will mostly benefit consumers.

Joshua Gans (University of Toronto) presented the fourth paper, entitled “Sharing the competition”. Gans begun by observing that to understand the incentives to share something owned on a sharing platform, one has to take into account that some consumers who would have bought will instead use somebody else’s asset. This will lead to ownership becoming more dispersed, which may be efficiency enhancing if asset costs are low and under-utilization high.

According to Gans, regulators and competition authorities should look for four signals in the sharing economy. Firstly, as regulations change, competition authorities should question rules that favour current incumbents. Secondly, competition authorities should oppose new regulations that reduce interoperability, since that increases lock-in effects. As an example, Gans mentioned regulations that force drivers to be Uber’s employees. Moreover, competition authorities should scrutinise regulations that based on short-term pricing issues. Finally, they should challenge rules that enshrine a particular business model. Gans concluded by arguing that pre-emptive enforcement in the sharing economy is not a good idea.

Arno Rasek (Chief Economist, Bundeskartellamt) commented on Gans’ paper. Overall, Rasek found the paper’s results plausible and appreciated that these were based on rather simple assumptions. He agreed with the paper’s first recommendation, adding that we should question even regulations that impede sharing and hinder platform entry. He also agreed with second recommendation, as limiting interoperability has clear anti-competitive effects. Beyong advocacy there may be scope for antitrust enforcement, e.g. if the platform operator uses exclusivity clauses and/or platform parity clauses. Rasek also agreed with the third suggestion in principle, but found it of limited practical relevance. Finally, he agreed with the fourth recommendation, but found it challenging to communicate the idea to policy makers when there is currently no diversity of business models.

Sofia Ranchordas (Leiden University) presented the seminars fifth and last paper. The author identified two types of sharing in the sharing economy: true collaborative sharing, promoting intermediated, sustainable peer-to-peer exchanges on the one hand, and a quasi-sharing model that is more commercial in character.

There are a few reasons why regulation of the sharing economy may be legitimate, such as public safety, consumer protection, privacy and security, access to universal and non-discriminatory services, insufficiency of reputational mechanisms. The author argued for an initial grace period and temporary rules, as well as introducing new, specific and flexible rules rather than applying outdated rules.

According to Ranchordas, the truly sharing economy has both pros and cons. The former are increased diversity and lower prices. The latter is surge pricing, which may be anti-competitive. The quasi-sharing economy also has pros and cons. The former include convenience, flexibility, uniformity, quality of services, reliability of services, and efficient match of supply and demand. The cons include a growth of the informal economy, tax evasion, and the disappearance of original idea of sharing underused goods.

Tara Koslov (Federal Trade Commission) was the discussant of the last paper. She called for a nuanced approach that would create a regulatory framework capturing the difference between extreme cases, such as for instance renting a property a few weeks per year and owning several properties that are constantly rented out. Caution was called for regarding trying to fit the sharing economy into existing sectoral regulations, especially if those are actively designed to keep out new entrants. The benchmark for regulating the sharing economy ought to be a counterfactual in which no regulation exists, or where the current regulations apply. If reputational mechanisms can fill the gap, then regulation ought to be unnecessary. Moreover, regulators should keep in mind that perhaps some practices, for instance surge pricing, supress intrabrand competition, but they may increase interbrand competition. According to Koslov, whoever introduces regulation should have the burden to show that there is a legitimate risk to address and that the regulatory response to that risk is calibrated accordingly.

Arvid Fredenberg (Chief Economist, Swedish Competition Authority) closed the seminar by presenting next year’s topic, which will be “The Pros and Cons of Market Definition”.

Presentations from the seminar