Internet as a public good: A case for net neutrality

The internet has public good characteristics, and telecom liberalisation will not be sufficient to keep the internet healthy and growing.

Classic OPTE Project Map of the Internet 2005

Classic OPTE Project Map of the Internet 2005: Flickr

Knowledge is the currency of power today. Land, capital and unskilled labour will not matter as much as knowledge as ‘factors of production’ beyond a point in achieving growth and prosperity.

The internet is both the manifestation of the power of knowledge, and has allowed knowledge to have so much value by interlinking vast amounts of it. In the net neutrality debate, a lot of analogies have been drawn about what the internet is, on why you should either enforce neutrality or stick to market competition. People have compared the internet to highways, to cable TV, to milk cartons, to electricity grids and to many more things. Rather than improving clarity, metaphors are distracting here – as the internet is only very poorly comparable to most other things.

A basic notion about the internet is that it is a networked good, which means that its value increases with its size. As my colleagues Karthik Shashidhar and Saurabh Chandra have pointed out, the utility of a network increases as the square of the number of nodes on it. Any barriers in this network will fragment it, and reduce the overall value of the network. It can however be argued if the size of the network needs to increase, then there needs to be sufficient private gain for the network provider. Enforcing something like net neutrality, they argue, could come against that. This logic is certainly used in many other networks including that for cable TV.

The internet is all but a public good – which means that the marginal social benefit of consuming internet services is much larger than the private benefit of doing so. Human civilisation as a whole is better off when it is more interconnected, more radically networked, and capable of doing things previously unimaginable. In that context, the internet defies comparison to any other network previously built by humanity. After proto humans manage to walk upright, the quest for knowledge has been central to their progress. Since the invention of language and of writing, the internet might be the biggest radical jump in making the quest for knowledge easier.

It is true that the internet was not always open (remember AOLnet?) and that it took a series of accidents to get us where we are today. It is also true that future networks could be even more radical and we must not constrain our imagination to what is presently possible with the internet. Many argue that sufficient competition in the network provider space can ensure network health – that regulations enforcing net neutrality are both unnecessary and counter productive. My colleague Nitin Pai in fact bats for net neutrality because the ISP market is uncompetitive in India, with high entry barriers and intense regulation.

However, if  the internet is a public good – will competition ever be sufficient to ensure the vibrancy of the network? Will competition be sufficient to improve the effective network size? I would argue that it might fall short of the mark. Thus, regulations that enforce net neutrality may be necessary to prevent ‘walled gardens’ from springing up. Competition must certainly be encouraged, but restrictions must be placed both on differential pricing of data from different sources, and of content providers from directly paying for the data their consumers use. While this will lead to a few inefficiencies, it is a necessary trade-off to keep the internet as open and flat as possible.

PS. Read the takes of Pranay Kotasthane, Gautam John, Anupam Manur, Varun Ramachandra and Devika Kher on the net neutrality debate.

PPS. Also read Deepak Shenoy on how telcos aren’t really hurting and Nikhil Pahwa‘s useful definition of net neutrality.

, , , , , , , , , ,


  1. Brainstorming | Pertinent Observations - April 19, 2015

    […] And Pavan argued that competition is alone not sufficient: […]