Barbara van Schewick
Mark Lemley & Lawrence Lessig
The Disclosure Project
Network Neutrality FAQ
Trying to figure out the network neutrality debate via the web is kind of hard. The wikipedia entry is too prone to fights and some of the other web sites are deliberately misleading. This web site is offered in the hope that it might help introduce matters and points of controversy in network neutrality. More details are found in the papers on the left.
Network neutrality is best defined as a network design principle. The idea is that a maximally useful public information network aspires to treat all content, sites, and platforms equally. This allows the network to carry every form of information and support every kind of application. The principle suggests that information networks are often more valuable when they are less specialized – when they are a platform for multiple uses, present and future. (For people who know more about network design, what is just described is similar to the "end-to-end" design principle).
(Note that this doesn't suggest every network has to be neutral to be useful. Discriminatory, private networks can be extremely useful for other purposes. What the principle suggests that there is such a thing as a neutral public network, which has a particular value that depends on its neutral nature).
A useful way to understand this principle is to look at other networks, like the electric grid, which are implicitly built on a neutrality theory. The general purpose and neutral nature of the electric grid is one of the things that make it extremely useful. The electric grid does not care if you plug in a toaster, an iron, or a computer. Consequently it has survived and supported giant waves of innovation in the appliance market. The electric grid worked for the radios of the 1930s works for the flat screen TVs of the 2000s. For that reason the electric grid is a model of a neutral, innovation-driving network.
The theory behind the network neutrality principle, which the internet sometimes gets close to, is that a neutral network should be expected to deliver the most to a nation and the world economically, by serving as an innovation platform, and socially, by facilitating the widest variety of interactions between people. The internet isn't perfect but it aspires for neutrality in its original design. Its decentralized and mostly neutral nature may account for its success as an economic engine and a source of folk culture.
Elaborating on this point – what the economic and social benefits of neutral information networks are -- has been one of the aims parts of my scholarship. In addition, other scholars such as Lawrence Lessig, Mark Lemley, Brett Frischman and Barbara van Schewick, have done much to develop these arguments.
The hard question, of course, is what exactly is “neutral?” But more on that further down.
Origins of the Debate
The Net Neutrality debate grew out of the concerns in the late 1990s about possible threats to the end-to-end nature of the internet. As documented by Mark Lemley and Lawrence Lessig in particular, the concern was that the vertical integration of cable firms with ISPs would prove a threat to the e2e design of the internet.
One suggested remedy was allowing consumers their choice of ISPs, usually called an "open access remedy."
Another idea was an anti-discrimination rule. This paper, Network Neutrality, Broadband Discrimination, written in 2002, argued that a discrimination rule was the best way, and in fact better than open access remedies, as a means to protect a neutral network. At some point thereafter the term network neutrality came into common usage.
The actual term "network neutrality," new or not, has a lot in common with alot of old ideas. The concept of a “common carrier,” dating from 16th century English common law, captures many similar concepts. A common carrier, in its original meaning, is a private entity that performs a public function (the law was first developed around port authorities). Furthermore, in networking, the “end-to-end” principle of network design is also a close cousin, if not the direct ancestor of network neutrality. David Isenberg’s lucid and well known “dumb pipe paper” is more or less the same idea.
Moreover, the basic economic problem found in the network neutrality debate (a form of "platform exclusion" or "vertical foreclosure") can be found in many other markets. In radio, for example, you have the problem of "payola" -- payments from the recording industries to radio stations, in exchange for playing their songs. Payola isn't great for music in the United States -- it is one of the reasons radio stations all sound the same.
However, I should point out that not all economists think that things like payola are actually a problem. They argue that if payola is inefficient then radio stations that don't accept payments will have an advantage, and would therefore stop, ergo, payola must be welfare maximizing.
I think it’s important to differentiate sharply between the principle of network neutrality and a network neutrality law. It’s a mistake to equivocate the design principle with proposed legislation of various forms. A neutral network might be designed without legal prodding – as in the original internet. In an ideal world, either competition or enlightened self-interest might drive carriers to design neutral networks.
However, when that isn’t the case—when carriers are interested in discriminating for one of various reasons -- matters get more difficult, and a law may be necessary.
To my mind, laws are most successful when they combat harmful behavior. To my mind, the basic justification for any law on network neutrality is an economic justification -- preventing behavior that may be narrowly beneficial for the carrier but that has negative spillovers for the economy and the nation.
The proposals I have supported focus on the following potential problems:
Blocking is the worst deviation from neutrality. Some economists might think it justified, but the basic problem is a distortion of competition between the blocked and unblocked companies.
2. Termination Monopoly Pricing
Since broadband service providers have a "termination" monopoly over the end user, they can use that to charge termination fees to those who wish to get access to the user.
3. “Playing Favorites” or MFN (Most Favored Network?) violations
Where carriers offer exclusive, preferential treatment to one application provider over others. Also distorting, though obviously less than blocking.
4. Transparency Failures
Where carriers fail to tell customers and application developers what, as far as they know, service they offer – i.e., estimated bandwidth, latency, etc.
In addition, a group have proposed rules on what can be called “Internet” service. Those might be considered a form of transparency regulation.
This is the hardest question for any advocate of network neutrality. I’ve always felt that carriers should be allowed to offer special services on purely private networks. That’s basically, for example, what the cable TV networks do.
The question then is whether it’s more neutral if carriers build networks with priority for certain content, or whether it’s more neutral to treat a bit as a bit.
I think the best, although still not ideal way to think about this problem is with the help of a private/public distinction. Private networks in this sense of the word are networks that aren’t interconnected with others. The cable TV network, described above, is a good example. On a private network, discrimination part of what gives the network its utility. By definition it is closed to outsiders, and that’s what makes it useful. The main point is that discrimination on a private network does have effects on the broader network – it doesn’t spill over.
Matters are different on a public network, or what is sometimes called an inter-network, or internet. On the internet, discrimination at one point can affect activities on other parts of the network. Hence
The best design principles leave carriers with a choice: build private networks that are discriminatory by nature (like cable, some versions of IPTV), or join the open network and play by the norms of neutrality. This is basically the status quo today.
What about the argument that the internet might profitably be improved – that the principles of network neutrality are leaving us with the network of the 1980s? Shouldn’t people in this field accept that deviations from neutrality might improve the public network, and not just private networks?
This point raises a technical question: for it depends on whether you think Quality of Service guarantees are possible across a large public networks. Most agree that they work on smaller networks, but how about the entire internet? Among others, the Internet2 research group has argued that QoS systems don't work well on public networks -- Andy Oram has a great piece explaining the development in their thinking.
If we assume that QoS generally doesn't work on public networks but does work well on private networks then we reach a common conclusion: the neutrality principle's main exception needs to be for private networks. It may be better for the entire network's design to distinguish between what's generally public, and what's private, and treat each network differently.
Relationship to Market Power
Another hard policy question is whether network neutrality rules (or laws) are called for in the absence of market power and concentration. As discussed above, network neutrality is first and foremost a principle. Therefore, in a market where vigorous competition exists, will the market itself solve any problems of discrimination?
I don’t want to try and answer the question on this web page – I address it more fully in my writings at left. But suffice to say, whether neutrality rules might still be useful given vigorous competition is an open question.
This short page can only introduce the concept of network neutrality and flesh out a few ideas. For more in depth work, read the works referenced in the margin.