The Role of Competition in Maintaining Net Neutrality

The rise of broadband internet for home-users within the last 10 years has caused important changes to the internet. Users of broadband services are able to consume and produce large amounts of internet content at very high bit-rates for a modest flat fee. At the same time, the number of ISPs serving broadband internet to home-users has dropped to two in most geographic markets. This is because the only providers that own the “last mile” of high-bandwidth cable to users’ homes are the existing telephone and cable providers – particularly within Canada and the US.

The combination of these effects (greater bandwidth for users, less competition for ISPs) has incentivized ISPs to change the service they are providing to home-users. In a recent article: Point/Counterpoint Network Neutrality Nuances by Barbara van Schewick and David Farber, van Schewick describes a number of scenarios in the US and Canada where ISPs have restricted their networks. On a general level there are three incentives for ISPs to impose restrictions.

1.    Further a monopoly – ISPs have an incentive to impose restrictions on their network to further profits for their own co-branded complimentary service or prevent a competitor’s substitute service. For example, ISPs such as AT&T and Verizon that offer co-branded services with Yahoo, may have an incentive to slow down traffic coming from competitors like Google.

2.    Conserve Bandwidth – ISPs have imposed network restrictions on certain protocols, ports, and applications so that in the short run high bandwidth consuming users don’t degrade the network to the point of excluding other users, and in the long run so that ISPs don’t have to invest as much in upgrading network capacity.

3.    Restrict Unwanted Content – ISPs have restricted specific internet content on their network based on their own internal policies and political views. For example, AT&T deleted words from a Webcast of a Pearl Jam concert in which the singer criticized George W. Bush.

Arguments why these Restrictions are Problematic

Within the article, van Schewick argues that these network restrictions are problematic and should be of concern because:

1.    It stifles new innovations. Both inventors and investors would be more hesitant to invest time and resources into new internet technologies for fear that it would be restricted by an ISP.

2.    It would reduce the ability of home users to freely choose what they want to consume making the market less efficient.

3.    It would truncate a home user’s ability to interact, educate, and engage in any discourse and idea.

A call for Net Neutrality

These changes and potential problems have spurred a call by many influential advocates for Net Neutrality. But what does Net Neutrality encompass? In the Point/Counter point article, Farber argues that:

The definition of net neutrality reshapes itself like lungs. It expands, drawing in causes ranging from freedom of speech to open access. Then it contracts, exhaling a lot of hot air, and starts all over again.

Perhaps a good working definition to start with is the one from savetheinternet.com

Put simply, Net Neutrality means no discrimination. Net Neutrality prevents Internet providers from blocking, speeding up or slowing down Web content based on its source, ownership or destination…. It protects the consumer’s right to use any equipment, content, application or service on a non-discriminatory basis without interference from the network provider. With Net Neutrality, the network’s only job is to move data — not choose which data to privilege with higher quality service.

Restoring Balance to the Internet

Assuming that the restrictions (e.g. traffic shaping) and the resultant effects (e.g. inefficient markets) are undesirable, how should the current system be changed to restore the ISP network to a less restrictive state?

While both van Schewick and Farber agree that some level of network neutrality is important, they disagree on how this should be achieved. Van Schewick argues that net neutrality laws are required. She justifies this claiming that more ISP competition and existing anti-competition laws will not discourage ISPs from restricting their networks to further monopolies, save bandwidth, and sensor unwanted content. Farber, on the other hand, argues that net neutrality laws are not necessary. Legislation in this area has been ineffectual in the past, would add an additional layer of complexity, and may actually restrict the development of the internet by restricting the free market choices networks can make. He says that further cooperation between institutions, ISPs and those concerned with the internet is needed. As well, ISPs need to provide more information to home-users on service restrictions. He also advocates for institutions to share more information on the health of the internet.

Farber’s argument and examples are persuasive. If net neutrality can be restored without legislation, this would be a simpler solution.  One of the alternatives to legislation is an increase in ISP competition. A number of advocates have called for increased competition in the ISP market.  van Schewick, however, finds issue with solely relying on the market to regulate ISPs.

One of the monopolistic practices ISPs have engaged in is blocking applications that compete with one of their other traditional business models. For example, in 2005, Madison River, a rural phone company in the US, blocked the internet telephone application Vonage because it threatened revenue from tradition phone services.

van Schewick argues that more competition won’t resolve this type of practice if all ISPs engage in the same practice and there is no other provider for users to switch to. As proof, van Schewick discusses how in many countries all mobile network providers block internet telephony applications to protect the revenue from their mobile voice services.  But this argument fails to take into consideration two things. One, are the countries in this example ones where there are many competitors within the same geographic market?  Two, assuming that the market is geographically competitive, is it actually  economically viable for any mobile carrier to offer this service?

Economic theory holds that because of the competition in a perfectly competitive market, the price an organization charges for a service will be driven down to the organization’s marginal cost of providing that service.  If the marginal revenue from allowing consumers to use internet telephony applications on the wireless network is not enough to offset the marginal cost of offering that service for all providers, then it doesn’t matter how competitive the market is, no provider will offer this service.  Logically this is desirable because it is the most efficient way to allocate resource. If providers were taking a loss their business would be unsustainable. At least with a competitive market, mobile providers will be striving to reduce costs and potentially provide access to internet telephony applications on their mobile services. Once one provider discovers how to make this economically feasible, then other competitors will need to as well or else face loss of customers. In a duopoly however, there is a less incentive for providers to reduce costs and price and improve service. It is more likely that both providers will engage in conscious parallelism and not undercut each other.

Secondly van Schewick argues that home-users do not have an incentive to switch if they do not realize their ISP restricts their preferred application and network.  In a competitive market, however, home users do not need to have the sophistication to detect network restrictions. Since competing ISPs want to attract customers away from their competitors, ISPs have a vested interest in investigating and exposing to home-users any restrictions their competition places on the network.

Thirdly, van Schewick argues that the cost of switching to a different ISP is significant and therefore a deterrent.  Within Canada nothing could be further from the truth.  Even with a duopoly, there are no cancellation fees except that the account will be terminated 30 days after notification, and most ISPs will send home users the high-speed modem free of charge via the mail – no technician required.  All the home-user has to do is plug in the device and input their password.

Lastly, van Schewick argues that ISPs don’t need to have a monopoly on the market to make discrimination and restrictions profitable; the increase in revenue from selling more copies of a co-branded service would be enough incentive to justify network restrictions. For example, even if Bell was competing against many ISPs and had only a small market share, Bell would still be able justify providing preferential access to Amazon music stores over other music providers if they were co-branded with Amazon. This is perhaps the most persuasive argument for net neutrality legislation. However, it is also a phenomenon of the free market that occurs frequently in other industries. The Air Canada Centre in Toronto has an exclusive arrangement with Pizza Pizza. Car manufacturers make exclusive deals with tire companies.  Loblaws carries President’s Choice products instead of Master’s Choice. The reality is that consumers frequently need to choose between products and services that are bundled with others they may have distaste for.  With enough competition, however, hopefully there would be a sufficient variety of options to satisfy the preferences of all consumers, even those who would be willing to pay a premium to have an open, unbundled gateway to the internet.

Despite van Schewick’s arguments, increasing competition in the ISP market still appears to provide a potential alternative solution to net-neutrality legislation. Perhaps the strongest argument for net neutrality laws not canvassed by van Schewick, however, is that restoring competition in the ISP market may not be possible. Currently, cable and telephone providers own the last mile of infrastructure that brings the internet to home-users. As Farber pointed out in his counter-point, existing legislative attempts in Canada and the US at forcing cable and telephone providers to relinquish control of this infrastructure has been difficult, expensive, and time consuming.   Without a forceful lever to restore competition, net neutrality laws may still be the next best option to keep the internet open.