Zero rating: a global threat to the open internet


Zero rating — the practice of offering internet users free access to some, but not all, of the internet, resulting in unequal access — is at the heart of the current debate over the future of the free and open internet. Around the world — from Delhi to Washington, Brasilia to Brussels — advocates, tech companies, and users are debating this crucial issue. But what exactly is zero rating? What is its technical impact on our use of the internet? What decisions have lawmakers and telecoms regulators already made regarding its use? And what happens next?

Keep reading to find out.

What is zero rating?

Zero rating is the opposite of Net Neutrality, the notion that all data on the internet should be treated equally. Net Neutrality is central to maintaining the internet’s potential for economic and social development, and to the  exercise of human rights such as the right to free expression. Its principles help ensure that that anyone, anywhere in the world, can receive and impart information freely over the internet, no matter where they are, what services they use, or what device they operate. Seen in this light, zero rating is a form of “network discrimination” — it deliberately sets up a system where “the internet” you get is different for different people.

Zero rating programmes manifest in different forms, the most frequent being “sub-internet” offers, where only a part of the internet is offered for “free”, and what we’re calling the “telco” model, where a telco prioritises either its own content or that of third parties. All forms of zero rating amount to price discrimination, and have in common their negative impact on users’ rights.

How does zero rating work technically? And how do zero rated programmes impact internet users?

Zero rating is all about control. Specifically, control over the user experience by the telecom carrier — and potentially its business partners. We can see this is true when we look at how zero rating is implemented technically. Technologically, it is about manipulation of the network, where you guide or force the user to change the way they would otherwise use it.

Right now, there seem to be two major models for implementing zero rating. There is the telco model, implemented by companies like Verizon, where the company gives preferential treatment to its own content, over whatever content might be independently created using its network. The second, and much more restrictive, model is the one used for sub-internet offers such as Facebook’s Free Basics programme and others, which orchestrate a tightly controlled “walled garden” network. Here, tech companies insert themselves in the middle of all communications in partnership with a telecom carrier, and dictate everything that users can and cannot do on the network.

The telco model: Quality of Service for me, not for thee

The telco model utilises some form of QoS (Quality of Service) protocol to ensure that its content is given preferential treatment, and therefore always appears “smoother” and more reliable than competing content. This content can be video, music services, or other applications.

Originally, QoS protocols were intended for internet users to dictate to the carriers what their preferences are for their own experience. Users could configure a router to pass information upstream to the carrier, designating which services mattered for them. The expectation was that carriers would comply with a “best effort” to fulfill the user’s preferences. The reality was that while some enterprise “users” configured QoS on their internet-connected routers, few individuals made use of the QoS features that existed (and still exist) in domestic networking equipment. Carriers have argued that because QoS features exist in the internetworking protocols, it implies that protocol designers intended to implement zero rating. But this assumption is simply not true. These features were never intended to empower carriers to force their preferences on users through zero rating programmes.

The sub-internet offers model: A middleman for the internet

The sub-internet offers model puts technology companies who partner with telecom carriers to provide such programmes in the middle of every network transaction. With current implementations of this model, users cannot do anything with any website or service without the company seeing their traffic and knowing what they are doing. By putting itself in the middle of every request and response over the network, the company can gain total access to the user’s behavior to build a detailed profile and have access to their communications history.

Not only does this model dictate that all of the user’s traffic go through the company, but the company needs the user’s traffic to be unencrypted at that interception point. Otherwise it cannot use the user’s data to build a profile. This model therefore does not allow end-to-end (strong) encryption. Such encryption would let the user break out of the zero rating “jail.” With end-to-end encryption, the carrier or provider offering a zero rating programme would not see what the user is doing, and this renders profile-building impossible.

We still don’t know what the long-term impact would be if more carriers adopt zero rating programmes. But given the restrictions and control such programmes impose, it seems likely there would be less innovation and less opportunity for internet users to participate in it. This is confirmed by the recent landmark empirical research report published by the Alliance for Affordable Internet (A4AI),  which found that “zero rating did not bring most mobile internet users online for the first time.”  Moreover, the report found that the vast majority of users (82%) prefer access to the full internet with time or data limitations, if restrictions had to be be imposed. This builds on other earlier initial research findings which had indicated that limiting access to the range of the internet impacted its perceived ability to materially impact the lives of the next generation of adopters, with low income users preferring even limited access to an open, unrestricted internet, versus the restricted experience that zero rating provides.

How are regulators dealing with zero rating? And what’s next?

There are several regulatory discussions taking place at the same time, all around the world. The Telecoms Regulatory Authority of India – TRAI — was one the first regulators to substantially look into the issue of zero rating. After extensive consultations, it passed what many have described as some of the strongest rules on zero rating — a comprehensive regulation restricting discriminatory differential pricing. However, after industry efforts calling for exemptions over this summer, TRAI announced a new consultation on free data pricing. While this consultation is taking place, the earlier differential data-pricing rules continue to stand in effect, remaining unchanged.

The US Federal Communications Commission (FCC) is next in line. Last year, the FCC adopted strong rules to safeguard Net Neutrality, but did not advance rules for zero rating, which at the time was a rather new phenomenon. Since then, US operators have launched a number of zero rating plans, which affect millions of people in the US. That’s why 70 companies and civil society groups, including Access Now, recently asked the FCC to be transparent about zero rating, creating a public process to inform evaluation of the existing zero rating plans.

Finally, in the European Union, we are approaching the last stage of establishing harmonised Net Neutrality rules. Next week, on June 6, the Body of European Telecoms Regulators (BEREC) will present the first draft of its guidelines for implementing the rules. Our hope is that BEREC will be inspired by the Netherlands, a country with a history of upholding strong Net Neutrality rules. There, the parliament approved a government proposal to prohibit zero rating. Such a ban is well within the scope of  the adopted EU Net Neutrality rules.

Once the guidelines are out, people in Europe can weigh in. There will be six-week comment period, letting anyone in Europe suggest improvements to the proposed guidelines. Access Now, together with the SavetheInternet.eu coalition, will help guide internet users through that process, providing a platform for sending comments to BEREC.

As we’ve seen, zero rating programmes are antithetical to Net Neutrality and have the potential to greatly damage human rights. We hope you join us as we continue our fight globally for an open, free, and secure internet for all.