Zero-rating practices in broadband markets
Cambridge, February 2017
On behalf of the European Commission (EC), Aetha developed a report providing an overview of zero-rating practices in Europe (as part of a joint project with DotEcon Ltd and Oswell & Vahida). The report also provides a review of the relevant legal and economic literature before discussing the potential benefits of zero-rating and the effects on competition. An initial framework for the assessment of zero-rating practices from a competition perspective, in addition to net neutrality considerations, was presented. A link to the report and a short summary of the research Aetha contributed to the report are provided below.
The full version of the report is available for download on Aetha's website. Please follow this link.
Summary of report findings
Zero-rating – the practice of excluding some traffic from overall data caps – has received a considerable amount of attention in the debate about net neutrality. However, it is permitted – under certain conditions – by European net neutrality rules.
In this report, our focus was on the benefits and competition of zero-rating, rather than net neutrality concerns. In order to achieve this, we conducted extensive research regarding zero-rated offers across Europe, including a survey of national regulatory and competition authorities. We then reviewed the available economic and legal literature regarding zero-rating, before developing a framework for the assessment of zero-rating practices on competition.
Zero-rating in mobile markets is becoming more prevalent across Europe as well as the USA. There was little zero-rating prior to 2012, but it is now increasingly common. Our research shows that the practice tends to be much more prominent in the case of mobile than fixed broadband services, simply because most fixed broadband plans are uncapped whilst mobile broadband plans typically have data caps.
The number of zero-rated offers varies significantly between European countries, and there does not appear to be any geographical or economic patterns in the use of zero-rating. Evidence from our detailed case studies partially supports the hypothesis that zero-rating is more common in countries where data caps are tighter – the case study countries with the lowest data caps, Portugal and Bulgaria, both have a large number of zero-rated tariffs.
The content categories that are most commonly zero-rated are social media, audio streaming, video streaming and communication (text), with data-light applications more commonly zero-rated than data-intensive applications. However, the nature of the zero-rated offers differs dramatically between countries. In some countries, it is more common for mobile operators to zero-rate one or two specific services (e.g. Facebook, as in Bulgaria and Germany), whilst in others mobile operators tend to zero-rate a wide range of services within a category (e.g. a group of audio streaming applications, as in Portugal and Sweden).
We have found little evidence of exclusivity/commercial arrangements between internet service providers (ISPs) and content providers regarding zero-rating. Where zero-rating is ‘exclusive’ to one operator, this is often because the application itself is exclusive to that operator or operator-owned (e.g. MobileTV is owned by Deutsche Telekom and is exclusive to Deutsche Telekom customers).
A review of economic and legal literature found that zero-rating is predominantly considered in the context of net neutrality rather than as a competition issue. However, much of the literature recognises that the effects of zero-rating arise from its impact on competition amongst ISPs and content providers, and therefore antitrust authorities may need to address zero-rating practices.
Regarding competition concerns from zero-rating, it is concluded that they may arise when there is reason to believe that competition amongst ISPs or amongst content providers is ineffective. In this case zero-rating might be considered to have potential exploitative or foreclosure effects. Foreclosure effects could stem from some form of exclusivity, either as a result of an agreement between the ISP and the content provider, or because the ISP is also the content owner.
Arrangements under which the ISP enjoys exclusivity in terms of zero-rating access to particular content would seem to be difficult to put in place for content that is available over the general internet, as it is easily replicable and does not require any consent from the content provider. Operator-owned content is different in the sense that it may be available exclusively to the customers of a particular ISP.
Arrangements under which an ISP guarantees exclusivity to a content provider would seem to require that content providers possess substantive market power and use it to foreclose competitors by imposing requirements on ISPs that potentially run against the ISPs’ interests. The research has not found any evidence for arrangements that would give exclusivity to a certain ISP or a specific content provider. Therefore these concerns may be of limited relevance in practice.
In summary, and acknowledging that information on the effective impact of zero-rating practices is very limited, there currently appears to be little reason to believe that zero-rating gives rise to competition concerns.