Thank your for your subscribe
Oops something went wrong. Please check your entry

The economics of net neutrality

The Federal Communications Commission, through its president recently appointed by Donald Trump, has submitted a plan for the outright abolition of a historic principle in the operation of the Internet, known as “net neutrality” or “open Internet.” Specifically, its status as a “public service.” This status had, however, been strongly defended by the same FCC, under the leadership of its previous president appointed by Barack Obama. This is the principle defended by the European Union in its 2015 Regulation, which is still under discussion. This plan will probably be accepted by the FCC Board next month. The debate on net neutrality, never interrupted, is being revived with force. The terms of this document need an explanation, and so do its industrial stakes.

December 2017
lire en français
lire en français
Executive Summary

The Federal Communications Commission has submitted a plan for the outright abolition of a historic principle in the operation of the Internet, net neutrality. In other words, the principle that Internet service providers must treat all data the same, and not discriminate or charge differently by user, content, or publisher. Can economics help us make sense of the debate underlying this decision? Yes and no. In favor of discriminating data, Internet bandwidth is a resource that is not infinitely expandable and where congestion penalizes all users. But Internet service providers could leverage regulation to strengthen their monopolistic position. This is an empirical question we are dealing with. But to make matters more difficult it also has a political angle.

There are three parties involved around the Internet: content providers (Disney, Vivendi, Netflix... but also Facebook or Google, which are content intermediation platforms), the network operators (AT&T, Orange...) also known as Internet service providers or ISPs, and us, the consumers. What are the relations between these three parties?

One can bring things back to a familiar example, that of mail, a network that shares some of the properties of the web. It is also tripartite, bringing together the distributor or distributors who manage the network, the mail senders and those who receive them. In most countries, the postal network has the status of a public service, the primary consequence of which is the single price rule: it does not cost any more to send a letter from a large city or from the countryside. There is no tariff discrimination between users for the same service, no possibility of blocking a sender. This principle of neutrality applies to other goods, such as electricity and gas distribution, which are now highly competitive activities between private actors. This is the very principle that FCC is now rejecting. Not, should we add, without solid arguments.

The case for discrimination

Internet bandwidth is a resource that is not infinitely expandable and where congestion penalizes all users. It would not be efficient that Netflix, a large bandwidth eater, squeeze out the small user; or conversely, that a large number of small Internet users damage the quality of the films sent. Watching a movie requires a high speed of transmission while other services can wait a few milliseconds longer (email, Mr. Trump's last tweet, a holiday photo that we share on Facebook). Since this is a scarce resource, the best way to allocate bandwidth is still a free market, say the adversaries of neutrality, allowing ISPs to charge more for providers who want to offer quick access to their customers. Postal services, after all, charge a different fee for a letter delivered on D+1 rather than D+2.

ISPs, of course, fight for tuning prices, because this can become the key to their profitability. The large content providers, not surprisingly, prefer neutrality and hence the single price. This is the case of Facebook or Google, which are universally used by millions of people who are happy with the free access to the amenities offered, including the speed of access. The challenge for ISPs, today's simple suppliers of pipes, is to recover part of the rarity rent. If Net neutrality keeps drying up their profitability, they say, their investments in improving the network are penalized; but abolishing Net neutrality, other respond, would dry up their own profitability and penalize their investment in content.

No wonder whether both sides invoke the sacrosanct consumer, the one who is always summoned in serious moments: quality of listening will say one, quality of content will say the other. In this popularity contest, content producers are winning the day: you keep sentimental memories of the pioneering spirit of an Internet open to everyone; you see once again the encroachment of markets on an area that you thought you were escaping from.

If markets worked perfectly well, it would be possible to make investment in networks and investment in content compatible, for the benefit of the consumer. ISPs would offer two types of contracts, one for bandwidth-intensive content providers and the other for small users. The better profitability of ISPs would increase the supply of pipes and encourage new players to come in and thus ultimately weigh on prices for consumers. The increased demand would in turn stimulate the content supply.

Transparency vs. monopolies: guess who wins?

Is that correct? The heart of the matter lies in the possibility of establishing genuine competition in the ISP market. Fixed costs are very high, which gives them a natural monopoly position: the more customers they have, the more profitable the network in place, a network where the possibilities of new entrants are limited. And from their point of view, it is unfortunate to stick to a model where the only margin comes from the end user. Why not monetize the transportation service used by content providers? Economist referred to their business as a "two-sided market": since they are intermediaries between media companies and the end user, the margin could be captured on both sides. Is there a risk, should they be given tariff freedom, that they would catch an exorbitant share of the total margin?

This is an empirical question, that economists are investigating. Today, the vast majority of them say - including until recently the FCC economists - that it is better to keep the public service rule. With tariff freedom, the temptation would become strong for ISPs to artificially slow down the flow of the “slow lane” to increase demand for the “fast lane.” The ability to block a supplier if it does not approve the contract terms (what the FCC's plan allows) would put them in a dominant position. Could we see an ATT or Orange technically penalizing Whatsapp or Skype, which eat them a good portion of their international telephone traffic? Economists also note that neutrality has not prevented a massive investment in networks and excellent profitability for operators such as AT&T or Orange. They find it absurd and naive to think that the principle of "transparency" - which is now promoted by the FCC plan as a safeguard – can effectively replace the public service rule.

The former competitive regulation is “disrupted” by the technological shock.

But from empirical and technical, the question now takes a political turn. People are beginning to worry about the incredible profits that Facebook and Google are able to capture. There are also characteristics of natural monopolies. And in the face of this thriving digital economy, the regulator is struggling to find the legal tools to break down anti-competitive situations, as it has learned to do over time with Rockefeller and Carnegie. The former competitive regulation is “disrupted” by the technological shock. By the way, some people, for instance a President Trump annoyed by a Silicon Valley that is fiercely hostile to him, would happily see traditional industrialists thwarting the big digital platforms. Of course, this would only replace bad anti-monopoly regulations with bad anti-monopoly regulations, except that they are not the same monopolies.

If the FCC reform were to pass, the strategy of telecom operators would be disrupted. Some of them today, seeing the huge profits of GAFAs, boldly choose to combine content and container, and invest massively in media. Altice, led by Patrick Drahi, is the emblematic example in Europe. In doing so, they keep the old model, the one where the customer who pays is the end user, the one who buys his internet subscription at the same time, it is hoped, as the subscription to TV channels, a newspaper or the soccer league.

This would become useless and futile, if it is not already so today. With tariff freedom, it would be enough for the ISP to hold the plug firmly in the pipes to make the other side of its customer base pay for it, i. e., on a par with a Facebook, the content provider. All this takes us far away from the little Internet users who, free of any worries, watch news on their mobile phone.

Last but not least, a political point. The FCC's plan is huge because of the legal upheaval it represents. Not only the entire economy of telecoms and content providers, but also most consumer habits have been built on the assumption of net neutrality. Imagine the political charivari if the plan were to be voted on by the FCC. Such a decision would be challenged in court, so the FCC will probably have to be supported by some US Congress legislation. And the FCC's credit might already have been affected by this 180-degree reversal. This does not bode well for its ability to carry out the anti-monopoly struggles that day by day seem more necessary in the future.


Francois Meunier
President, Alsis Conseil, Associate Professor of Finance, ENSAE ParisTech