We’re three days away from the FCC meeting at which the commission is scheduled to vote on the chairman’s Notice of Proposed Rulemaking to revise the Open Internet rules. I want to take this post to discuss what this NPRM is – and isn’t – about.
Let’s start with what the NPRM isn’t about. It isn’t about the future of the Internet. It is not about whether the FCC will or will not regulate business practices on the Internet. It is not about whether the FCC will or will not allow “fast lanes.” Anyone who tells you otherwise is advocating for a preferred outcome. In many ways that’s fine: most people interested in this NPRM have preferred outcomes, and many are so steeped in the issues (and their beliefs about them) that the “future of the Internet” rhetoric is useful shorthand. But for most folks — those who have never heard of an NOI, NPRM, FNPRM, R&O, or the APA — the shorthand is at best confusing.
Today, the FCC has broad authority to regulate the Internet. Under Section 706, if a firm were to implement a consumer-harming business model that violated the principles of network neutrality, the commission could bring an administrative action against that firm: the commission could sanction the firm and require it to forego the business practice. Indeed, under the generous deference afforded agencies such as the FCC, the commission could almost certainly take such action if it were presented with evidence that such business plans only harmed consumers indirectly, for instance by adversely affecting incentives for edge providers to innovate or invest in new technologies.
In other words, the “rules” that net neutrality advocates are demanding already exist. The Communications Act already gives the FCC power to investigate and take action against business practices that violate network neutrality, and instruct it to use that power when doing so is in the public interest, convenience, or necessity.
The Open Internet NPRM isn’t about anything that has grabbed the attention of the media or protesting public. The NPRM is about the decidedly unsexy legal question of how to implement the already existing statutory authority. Should the FCC implement clear ex ante rules that may harm consumers by foreclosing the development of pro-consumer business practices? Or should it instead rely on general standards that are enforced on a case-by-case basis as bad conduct arises? Importantly, in either case the FCC would need to bring an administrative action against a firm that violated neutrality principles. If the Commission implements broad ex ante rules, it risks that they will be invalidated by the courts – for a third time! – when they try to enforce them. If, instead, the commission relies on standards, the commission will have a slightly harder time challenging problematic conduct, but will face less risk that the entire Open Internet regime will be invalidated.
So the question of whether to implement clear rules or rely on case-by-case adjudication isn’t about whether to allow or proscribe “fast lanes.” It’s about how the FCC will take action against firms that harm consumers, the burden of proof that it will face when it takes such action, and the risks that it faces should it lose.
This is the sort of issue that only a lawyer can love – and even then it’s one for which many lawyers have little patience. Every law student learns during their first year (usually their first semester) about the blurry distinction between rules, standards, and principles. And then they spend the remainder of their time in law school struggling with this distinction as it plays out in various contexts. There is always a natural preference for rules, which provide clarity and the comfort of certainty; but they are also always under- and over-inclusive. Standards provide greater flexibility so to reduce the costs of this under- and over-inclusiveness, but they come at the cost of less certainty and greater costs when it comes time to enforce them. That is the entire debate over the Open Internet NPRM in a nutshell.
The same is basically true for arguments over reclassifying the Internet as a Title II (“common carrier”) service. Contrary to commonly held opinion, Title II services don’t prohibit discrimination. Title II isn’t a magic wand that requires common carriers to be neutral. It only bars “unjust and unreasonable discrimination.” Given the overwhelming body of technical and economic literature showing that discrimination can benefit consumers and facilitate new services — and that the Internet never has been “neutral” in the sense that most net neutrality advocates use the term — there is no reason to believe that reclassification would prevent firms from adopting business models to which net neutrality advocates object. Indeed, think about any services traditionally understood as common carriers: trains, roadways, parcel and post, electricity and other utilities, and even communications. Each of these services has always allowed for differential levels of service — and, when a customer of any of these services gets to be large enough, common carriage rules have always allowed individualized and preferential services to be negotiated in one form or another.
In other words, as with the rules vs. standards debate in the context of Section 706, the debate here is about the relative certainty afforded by reclassifying versus not reclassifying, and the relative costs of taking action against firms under either regime. Whether or not the FCC reclassifies, it would still need to take action against a firm engaging in potentially harmful conduct – and the arguments against and in favor of the firm’s conduct would be roughly the same under either regime. Under Title II, the FCC would likely have an easier time winning a case against a bad-acting firm. But the cost is that it would be harder for firms to develop new, pro-consumer, business models. Doing so would require receiving FCC permission under a process known as forbearance – a process that typically takes 12 to 15 months (or, in Internet-time, forever).
Finally, this isn’t the FCC’s only bite at the apple. The commission is free to change its rules, almost at a whim. So, should the commission adopt rules this year that prove to be too lax, it can change those rules next year (or even later this year!). Importantly, if the commission adopts rules that are too lax, it will be relatively easy to tell (firms will adopt harmful policies), and it will be easy for the commission to change the rules in response. Indeed, having direct evidence that stronger rules are needed, it will be easier to adopt tougher rules than it is today. On the other hand, if the commission adopts too tough rules, we are unlikely to ever see the harm – it’s hard to see when something good doesn’t happen. So, if the commission adopts too strong rules today, we’re probably stuck with them. We’ll all be poorer for it, though we may never know.
It’s very easy to get caught up in the passion and rhetoric surrounding network neutrality. But this issue before us right now – and before the commission on the 15th – isn’t about network neutrality. It’s not about whether the commission has the power to take action against firms that violate network neutrality principles. The commission already has that authority. The issue before us right now is the boring, technical, lawyerly question of how to implement that power. Should we rely on rules or standards; what are the relative costs of each; what are the relative risks of each; how do they affect burdens of proof.
My recommendation to the chairman, should he be reading this in a moment’s refuge from the angry mob outside his office: if the commission does not adopt the NPRM on Thursday, let the issue die there. You already have authority to take action against problematic conduct. The lack of a rule-making will make it harder to take such action (especially if the commission were to seek fines or other due process–protected sanctions) — but fundamentally, you still have the tools to protect the public interest. And you have more important things to do than tend to the care and feeding of your predecessors’ albatross.