Net Neutrality Debate

I don't believe there is very conclusive evidence on either side of the net neutrality debate. There simply isn't consensus, as is demonstrated by this IGM Poll of economists. 

However, in the absence of such evidence, I feel that we should presume against net neutrality as a policy simply because it has not met its burden of proof. At the same time, I am very much willing to change my mind about this because I have only recently begun critically analyse the debate in the economic literature about net neutrality. The primary reason I am making this post is to learn. 

I am familiar with three main arguments against Net Neutrality: 1. it prevents the internalization of negative externalities, 2. it undermines the incentive that ISPs have to create new bandwidth infrastructure, and 3. It leads to inefficient resource allocation (ie dead weight loss). 

One: net neutrality prevents markets from internalizing negative externalities.

In economics, a negative externality is is the cost that affects a party who did not choose to incur that cost or benefit. Usually externalities are a market failure, that is, the free market is unable to internalize the externality and so government intervention is necessary (for example, CO2 emissions are a negative externality, the government can internalize the externality with a carbon tax). 

However, according to Coase Theorem, there are very specific situations in which the market can internalize externalities without the need for government intervention - namely that 1. there be well defined property rights and 2. there be sufficiently low transaction costs. I argue that broadband services meet all the requirements for a Coasian bargain to take place, and that net neutrality prevents that bargain from happening.

With net neutrality we would have content providers that users place low value on using too much bandwidth and content providers that users place high value on using too little bandwidth. If we allow bandwidth to be a marketable resource like any other, then content providers will bargain with ISPs in exchange for bandwidth. The resulting outcome is an optimal allocation of bandwidth to content providers.

Now you may be wondering "why can't content providers bargain with each other for bandwidth? Why assign the market power to ISPs?" Its theoretically possible for Netflix to negotiate and bargain with Google over how much bandwidth each provider uses. The reason why this doesn't happen is because of high transaction costs. In order for a Coasian bargain to lead to a pareto improvement in efficiency, you need to assign property rights (ie the right to sell bandwidth to CPs) to the agent with the lowest transaction costs, which would be the ISPs. Owen and Rosston explain why: 

There are several problems, however, in assigning the access right to upstream content suppliers. The first problem is that there are no significant transaction costs involved in LBBs selling access to content suppliers, whereas the reverse is not true. An LBB operator cannot purchase access rights from a set of sellers whose identities are unknown. Most would not come forward or even exist until a general offer of purchase were made. Many sellers would be fraudulent. Second, as noted above, some upstream suppliers may produce negative externalities on demand for overall delivered content, which they have no incentive to take into account. Well-known examples include spam, pornography, and other material whose very presence reduces consumer demand for connections. LBBs have a greater incentive than individual independent content suppliers to internalize this externality. The reverse problem exists for content producers that may generate positive externalities in demand.

Two: net neutrality undermines incentivizes to create new bandwidth infrastructure.

This argument is pretty simple. If bandwidth becomes a marketable resource, then ISPs will have an increased incentive to build more internet infrastructure simply so that they can sell more of it. Wallsten and Hausladen analyzed the effects of neutrality on broadband infrastructure in Europe and found that there "appears to be significantly less investment in next-generation networks. "

 

Three: a two-sided pricing model leads to a more efficient allocation of resources.

Gans and Katz outline several economic models which demonstrate that net neutrality leads to inefficiency. Their basic argument is that under a two-sided pricing model, ISP will have less incentive to use their monopoly power to raise the price of internet service. Cheng and Bandyopadhyay substantiate this claim with a more complicated game theoretical model indicating the social welfare implications of abolishing net neutrality are either positive or neutral in the long run.

I look forward to your views on this.