The developments in the case dealing with Facebook tracking its users beyond the social network have reached another level. In late June, it was confirmed that Facebook must curb data collection from users. According to the lead judge, “Facebook must give users the choice to reveal less about themselves – above all what they reveal outside of Facebook”. The issues associated with collecting personal data in the context of online tracking and profiling are normally considered to be solved under data protection rules; increasingly, however, they also feature as part of competition enforcement and markets regulation. In this light, the Facebook case is significant: it goes to the core of the GAFA’s business models and practices: that is, offering free services and generating profits based on users’ data and online advertising. While it is true that “no money changes hands when someone signs up for Facebook”, the data economy runs on a different currency: (already) in 2016, for example, “Facebook’s average revenue per user in the US and Canada was $62”. While Facebook insists people want a free service (enabled by amassing data and targeted ads) instead of a subscription model, studies show that users do place value on different privacy aspects – even if they agree with the terms and conditions of use when registering.
A recap of what has happened so far
As already discussed, after several years of investigating, the German Competition Authority (Bundeskartellamt) in early 2019 found that Facebook’s data collection practices were unfair and that its use of terms of service was in violation of German competition law regarding abuse of dominant position on the market for social networks. The abuse was committed “contrary to the rules of the [GDPR], making the private use of the network dependent on the authorisation to link the data relating to users and their devices generated outside facebook.com with the personal data generated by the use of Facebook itself without additional consent”. Facebook got one year to “stop ‘unrestrictedly collecting and using’ data and combining it with users’ Facebook accounts” without their agreement, so as to return Facebook users their sovereignty.
Facebook appealed against this decision and requested an interim relief; last August, the Düsseldorf Higher Regional Court ordered suspensive effect of the appeal and found that “Facebook’s terms and conditions were sufficiently accessible to the average consumer”, who can autonomously decide about (not) consenting to them. Thus, Facebook is not required to apply the restrictions on its data collection and use practices that were ordered by the Bundeskartellamt. The regional court expressed doubt as regards the legal basis of the decision of the competition authority, namely that violations of data protection rules could be instrumental to find a violation of competition rules. This was considered a blow to the innovative theory of harm pursuant to which the data collection as done by Facebook was unfair because it “invades its users’ constitutional right to privacy” and could not be considered a genuine choice.
Four years – and not an end in sight
Following the appeal by the Bundeskartellamt, last month, the German Federal Court of Justice (Bundesgerichtshof) issued a ruling; to many, Facebook’s “big loss” came as a surprise. This ruling was in relation to the order of suspensive effect of the appeal; the appeal against the Bundeskartellamt decision is still to be decided by the higher regional court. The court in June provisionally confirmed the allegation that Facebook abuses its dominant market position. In the preliminary proceedings, it decided against Facebook and established that “the suspensive effect of seeking a legal decision” ordered previously is no longer in place: the suspension of the “enforcement of a ban on Facebook combining user data into so called ‘super profiles’” without such users’ consent is now overturned and hence the prohibition by the competition authority can be enforced.
Autonomy gets centre-stage role
The harm is thus in the fact that with working competition, so in absence of Facebook’s dominant position, more options regarding the level of disclosure/use of personal data would be available. The court noted that the services offered to users “overlap and blend seamlessly” with the financing of the platform via online ads. As a dominant player on the market for social networks, “Facebook has a special responsibility to maintain the still existing competition”; here, access to data has to be regarded, as it has “great economic significance”. Because Facebook has at its avail such scope and quality of data, necessary to compete, its behaviour can lead to lock-in effects and negatively affect competition on the markets for social networking and online advertising, where, according to the court, Facebook needs not be dominant.
Public values and the contextualisation of market behaviour
This is the first case exploring whether data dominance is an issue for competition policy. In this sense, it is important in addressing the approach to governance of powerful digital platforms, especially in the light of current policy initiatives and debates. This, if anything, is a political debate. Perhaps more interesting, from an academic point of view, is that the developments in this saga cut straight into debates on the nature and objectives of competition law, which, traditionally, in “its economic character aims […] to enhance efficiency” (AG Wahl in Dunne, 2020). The court in Facebook understood it more as “protecting the autonomous decision-making process of market actors”. As Podzsun argues, competition law is not only about (measured or measurable) welfare effects, but also about protecting that making (market) decisions is done freely and independently. In this sense, the function of competition law also encompasses protecting consumers’ – as economic agents – control in making online choices freely and in a non-coerced way as an economic value (Podzsun, 2019).
Consumer choice is, of course, an important competitive parameter. However, the reliance on the notion of autonomy can also serve to open up the debate on the role of markets and private power. With digitisation, public values – namely those beyond economic efficiency – are affected and put under pressure. This development seems particularly pertinent (and obvious) in times of the current global pandemic, which has demonstrated our increased dependence on big tech and its involvement in what are essentially public services. In the age of data-intensive everything, autonomy as “the freedom to control the terms of one’s engagement with data markets, in terms of individuals defining for themselves how their data are used and what they enable,” could also be one of the important aspects of data justice (see Taylor, 2017). But the issue is, of course, much broader than individual control over data; data justice framework(s) also bring to light the questions of power, politics and interests, inequality and inclusion, governance, citizenship, etc. When it comes to the functioning of data markets (or markets revolving around data), the critical questions we need to ask ourselves therefore are what happens when dominant firms’ strategies go beyond market (power) effects and touch upon public values? What is the desirable effect of digitisation and digital innovation – and can economic regulation help promote it?
Yet, the case is far from closed; while Facebook must now comply with the competition authority’s data collection ban in Germany, the main proceedings in the appeal are still to follow. This leaves the outcome of the case very much out of sight as the proceedings will likely go on for years. There are also speculations (or calls) that the German court might make a reference for a preliminary ruling to the European Court of Justice regarding a uniform interpretation of EU law. While this would probably delay the resolution even more, it would bring a welcome clarification concerning this issue of fundamental importance.
Edit as of 28 August 2020: This post was written based on the court’s press release, media reports, and experts’ opinions. In late August, the German Bundesgerichtshof published its reasoning in the interim decision in the Facebook case (in German), which explains how the court constructed the theory of harm. This is not entirely reflected in the above post. A link to the decision and a summary with an excellent commentary of the court’s actual reasoning can be found here, pointing out some “very, very, very interesting aspects”.