The public consultation period on the US Federal Communications Commission’s proposal to repeal net neutrality, the principle that all traffic on the Internet should be treated the same, ends today, July 17. With 8.5 million comments, this has been by far the most widely debated policy issue in FCC’s history. The rules under consideration for repeal were adopted only two years ago amidst intense controversy, and resemble those that were introduced in Europe with the 2015 Open Internet Regulation. Konstantinos Stylianou, Lecturer in Competition Law and Regulation at the University of Leeds School of Law, argues that the open Internet will survive even without strict net neutrality rules.
Say what you will about the merits of net neutrality; we are unlikely to reach consensus, much like academics and policy-makers have not reached consensus over the past 15 years ever since the term was put on the policy table. I have explored elsewhere the reasons why I think consensus has been elusive and what to do about it, but suffice it to say here that there are good arguments on both sides, and therefore any policy that categorically sides with one camp or the other is doomed to be wrong on the merits.
But the main problem in the news today is not that the current rules, which are demonstrably pro neutrality, are wrong or unbalanced; rather, the more fundamental question is whether we need rules in the first place at all. If not, then the repeal of the current rules should not spell the disaster that open internet activists fear. In fact, repealing superfluous rules would be good regulatory policy; the most basic premise of regulation is that rules of general applicability, like the ones on net neutrality, should be reserved only for generalised harm significant enough to justify the unavoidable cost that comes with regulation.
Far from it, not only has the case for net neutrality been weak from the beginning, but a look at all recent legislation reveals that net neutrality is largely a solution to a problem that does not exist. Very few documented cases exist on the docket of regulatory authorities as real-world examples of ISP practices violating the principles behind net neutrality. It is illustrative that FCC’s 2015 Report and Order which enacted the latest rules (the ones under consideration for repeal) did not cite any new evidence in that direction (para 75), but instead relied on evidence included in two previous attempts to enact relevant rules, the 2014 Notice for Proposed Rulemaking (paras 39-41) and the 2010 Report and Order (paras 35-36).
And even in those documents the evidence is scarce and unconvincing: the FCC cites the example of AT&T temporarily blocking Apple’s FaceTime app, which however, following an investigation, the FCC itself found that AT&T had legitimate reasons to block, the example of Verizon blocking certain tethering apps, which the FCC settled, relying not on net neutrality rules but on spectrum licensing rules, and a small number of incidents that date back to 2005-2008 (mainly the 2005 Madison River case and the 2008 Comcast BitTorrent case), which served as the main impetus behind net neutrality legislation in 2010. All in all, the entire record on the need for net neutrality rules contains fewer than five confirmed cases over the course of a decade.
All the while the EU delayed adopting net neutrality rules thanks to the perceived more intense competition among ISPs, which was seen to act as a safeguard against harmful practices. In a change of heart, relevant rules were adopted by virtue of the Open Internet Regulation in 2015, largely as a response to the findings of a report by the Body of European Regulators for Electronic Communications (BEREC) (recital 3), which showed that European ISPs engage in numerous and various traffic shaping practices (the report was also cited by the FCC in support of its own rules). A closer look, however, reveals that the report does not argue that these practices would amount to a violation of net neutrality principles (and therefore presumably relevant rules should be enacted); it merely documented traffic shaping policies without opining on their legitimacy or threat. Indeed, in its 2017 report on monitoring compliance with the Open Internet Regulation, Ofcom did not identify a single case of net neutrality violation in the UK, despite the continuation of the very practices that BEREC listed.
One could counter that the reason there are only a few incidents is precisely because of the deterrent value of existing net neutrality rules. But this is a feeble argument. Even bypassing the obvious chicken and egg problem, the timing and frequency of net neutrality violations proves nothing about the value of the relevant rules. The documented violations above are spread over a period of time that was only partially covered by net neutrality rules, which leaves little room for correlation between the rules and the violations (or lack thereof).
More importantly, though, even if one was somehow convinced that there is a generalised problem to be addressed, there is a wide spectrum of less onerous measures to be adopted compared to the rules currently in force. Self and co-regulation, regulatory threat, antitrust, and—my personal favourite—antitrust-like regulation, like the proposal put forth (but later rejected) by the FCC in its 2014 NPRM (paras 116-128), are all milder alternatives that are more proportionate to the level of harm currently posed by net neutrality violations.
I do not suggest that the broadband industry works flawlessly and that no oversight is needed. But the current rules, when juxtaposed with the documented threats, feel like an overkill. Regulatory agencies have spent far too much time quibbling about net neutrality, instead of turning their attention to more contemporary and more pressing issues, like harmonising spectrum, interconnection disputes, and connecting the unconnected. But this is the topic of another post.
This post gives the views of the author and does not represent the position of the LSE Media Policy Project blog, nor of the London School of Economics and Political Science.
In some ways you are right, this is a bit of a no win fight for either side; and if it tips in favor of one side or the other it might not really matter. But I believe you are still mostly wrong in how you arrived at your conclusions.
Your claims about what the monopoly edge access providers will or will not do cuts against a lot of history that says they will try to erect walled gardens to monetize their inefficient, balkanized vertically integrated models. Remember they only see a portion of demand as opposed to the core providers who have a complete view of demand.
It’s important to establish that the pro-competitive and equal access policies in the voice telecom sector in the 1980s-90s (and keeping voice and data separate) were what led to the commercial scaling of the internet/web in the US 5-10 years ahead of RoW.
The Republican FCC killed off that open, competitive model in the early to mid 2000s, leading to the final wave of consolidation and monopoly/oligopoly structure we have today.
At the same time, our multi-faceted approach to wireless competition led directly to Steve Jobs being able to single handedly resurrect equal access with the introduction of the iPhone.
Causality is not well understood in networks due to their complexity and the existence of externalities both of which contribute to unintended consequences.
So before you are so emphatic with your predictions keep these 3-4 “facts” in mind.
BTW, the flaw in NN (which I believe to be a contrived notion) is the same as the flaw in TA96 and any policy applied asymmetrically. At the end of the day fostering competitive, generative and sustainable digital network ecosystems that provide universally cheap broadband access is accomplished by: a) mandating interconnection as far down and out to the edge of the informational stack, and b) fostering and monitoring interactor settlements that clear supply/demand ex ante and at the margin both north-south and east-west.
The first is thought of as Title II, but it is not. The second cuts against the notion of a settlement free internet; which itself is monopolized and ossified directly as a result of the lack of interactor settlements. These issues are not new and have been around since the late 1890s-1930s. Happy to explore more of this with you. Where we end up with is a new market theory for networks called equilibrism.
Lots of thoughts in there. I just want to focus on two things: one is that I agree with you that in theory ISPs have the incentives and ability to violate net neutrality (as the FCC has argued convincingly). But what I’m saying is that, despite incentives and ability, ISPs have not exercised that power to violate net neutrality. And it’s not because of the existing rules.
The second thing is that policies of the 80s and 90s are no longer good guidance. Not only is there more competition (even the terminating monopoly you’re assuming is not true), but the nature of competition is different. I’ve explored that elsewhere if you’re interested: https://ssrn.com/abstract=2963648