Digital Networks Act: What the EU Can Learn From the Swisscom-INIT7 Dispute
Main takeaways
- EU policymakers should take note that self-policing of internet connectivity works extremely well, with few disputes occurring around the world
- Swiss case shows that some telecom providers try to abuse their monopoly position at the expense of smaller ISPs, but regulators can deal with this with existing tools
- That is why the EU’s upcoming Digital Networks Act (DNA) should not introduce any type of settlement or arbitration mechanism, as it would only empower incumbents
In December 2024, Switzerland’s electronic communications regulator (ComCom) ruled that major telco Swisscom must provide zero-settlement peering to Init7, a local internet service provider (ISP), concluding a long-running dispute. Controlling 57% of the market, Swisscom is the dominant ISP in Switzerland, while Init7 holds a share of less than 1%.
Ultimately, the case is relatively simple. Swisscom was found to hold monopoly power over the termination of internet traffic to its subscribers and engaged in anticompetitive behaviour by attempting to extract higher fees from smaller ISPs. Following an investigation by ComCom, and in accordance with telecommunications legislation, settlement-free peering now has to be applied.
This decision is highly relevant, as similar anti-competitive behaviour by major telecom operators is reported across Europe. Moreover, the EU is preparing to review the European Electronic Communications Code (EECC) by the end of 2025. As the European Commission reviews the EECC, it should only rely on clear evidence provided by the market and past regulatory assessments. The IP interconnection market is functioning well, with the only disputes being caused by anti-competitive practices from big telecom companies.
Therefore, the Commission should drop the idea of introducing an arbitration or dispute resolution mechanism via the Digital Networks Act (DNA) for possible conflicts between telecom operators and other ISPs or content providers. This is because: there is no market failure justifying such regulation; existing tools can address the few disputes that do arise; it would unfairly benefit dominant telecom operators and effectively introduce network fees.
1. Why peering works – Until monopolists get greedy…
One of the reasons why the internet is so valuable is because anyone can reach almost any website in the world regardless of which ISP they use. ISPs in the same and nearby countries, which exchange a lot of traffic, can directly connect with each other. This is called ‘peering’ and is the approach Init7 preferred.
ISPs exchanging less traffic in a country can all come together at an internet exchange point (IXP) where they interconnect with each other. These IXPs are typically owned by third parties. Finally, ISPs in other countries can use larger telecoms companies who connect into IXPs and peering arrangements elsewhere. These are called ‘Tier One’ players and the service they provide is called ‘transit’.
ISPs tend not to charge each other, nor content and application providers (CAPs), for peering. If they are of broadly equal size then it is likely that they will pass as much traffic to another ISP as they will receive from it. However, ISPs do pay transit providers, as they offer broader connectivity to networks around the world.
This arrangement has worked well for many decades, but there are occasions when ISPs try to abuse this informal working arrangement. There is a monopoly problem: the ISP controls the only route to the subscriber. They can charge extra to let others’ data through to reach their customers – the exact complaint now raised against Deutsche Telekom by consumer organisations and digital rights groups in Germany. A large telecom operator might try this tactic to extract revenue from popular content providers and stifle ISP competition.
2. The myth of traffic costs – More data doesn’t mean more costs
Note the similarities here with the “fair share” payment demands, which Europe’s big telecom operators want to see incorporated in the DNA through an arbitration mechanism. Such a mechanism would effectively provide telcos with an institutionalised way to extract network fees from content providers to reach subscribers. As can be seen, this would enable anti-competitive and monopolistic behaviour by the telcos, and is based on a false premise.
In fact, costs for ISPs are very insensitive to traffic volumes. The major cost is the initial rollout of the broadband connection to the home, which may be a fibre or a cabled connection. These connections can carry traffic up to their maximum data rate, often many 100Mbits/s, without any material cost increase.
If all the subscribers linked to a local exchange generate more traffic, then the backhaul from that exchange may need to be upgraded, but this is rarely needed and the cost relatively low. Similarly, connections to peering points and higher speed ports into the routers that exchange traffic may need upgrading periodically and again this is a relatively low cost, often only a few thousand euros. Because of this insensitivity of costs to traffic volumes, most ISPs are content to have relationships with other ISPs where no money is paid between them.
3. Don’t fix what isn’t broken – Why the DNA should tread lightly
Disputes in internet connectivity are rare. When they do occur, they typically stem from market failures – often involving monopolistic telco practices, as illustrated by the recent cases in Switzerland and Germany.
Given this evidence, the foundational premise of the EU’s upcoming Digital Networks Act should be reconsidered. Proposals to introduce network fees, even through an arbitration mechanism, should be unequivocally rejected.
The DNA is intended to serve a clear purpose, that is to “boost secure high-speed broadband […] and incentivise and encourage investments in digital infrastructure.” In order to meet this objective, EU regulators must focus solely on constructive policies that advance these goals – and discard proposals that would undermine them, as they risk breaking the internet instead.
This article is the first in a series of guest posts authored by Professor William Webb. William is CEO of Commcisive, an independent consultancy providing advice and support across telecommunications matters. He was one of the founding directors of IoT start-up Neul and became CEO of the Weightless SIG. Prior to this Webb was a Director at Ofcom. He was President of the Institution of Engineering and Technology (IET) from 2014-2015, has published 20 books, has multiple honorary doctorates and awards, and often posts opinion pieces on LinkedIn.
