A main argument in favor of network neutrality is that a discriminatory network distorts markets that depend on the network, and ultimately may slow national economic growth. For example, if a network favors search engine A over search engine B, A may become dominant even though B's technology is better. Similarly, if the network favors a usage of the network popular at a given time (say, Gopher, enormously popular in the early 1990s) that may slow the competitive arrival of a new usage (like the World Wide Web). In other words, a discriminatory network may "freeze" innovation (particularly application innovation) based on today's dominant applications. The link to national economic growth is as follows: many economists believe innovation is a major catalyst of economic growth, meaning that if a discriminatory network leads to less innovation, a country will grow more slowly. Opponents of neutrality regulation say this argument confuses a necessary form of discrimination, discrimination among packets with different latency requirements, with arbitrary discrimination between applications of the same type, and is therefore blind to network engineering requirements.

The arguments against network neutrality as a principle take three forms. The first and most common says that packet-level discrimination is absolutely necessary in order to provide Quality of Service on any packet network, and broadly-written regulations such as the Markey Amendment would ban it.

Another argument says that service bundling is necessary to encourage investment in the networks of the future. If a broadband carrier is, for example, allowed to charge more for a high-priority voice service when competing services such as Skype run at standard priority, the carrier's voice service will sound better, and this will limit Skype's appeal while increasing the carrier's revenue.

One response to this argument concedes that discrimination results in greater profit for broadband carriers -- but asks whether the costs in terms of application innovation are worth it. In the example, the operator is exploiting a bottleneck in the network to extend its monopoly from the network service to another sector entirely, in this case messaging applications. It also asks whether allowing discrimination is the best way to encourage network deployments. In other words, there may be less distortionary ways to encourage carriers to build out their network, such as using the tax code, or government funding.

Another argument against network neutrality relies on the economics of congestion. A neutral network is like a public good, leading to collective action or tragedy of the commons-like problems. Hence a provider may need to discriminate as between users or usage to ensure maximum network performance. For example, if some one use up too much bandwidth, this argument suggests, a network operator should be allowed to slow it down. The typical answer to this argument goes as follows. There may be more and less distortionary ways of managing bandwidth -- and blocking or disfavoring certain applications is more distortionary. A more neutral way of managing bandwidth is to manage bandwidth at the consumer side - i.e., to limit the users to, say, x gigabytes per month after which their transfer rate is reduced, instead of banning applications (systems of this type have been employed in other countries, e.g. Australia).

However, service providers in the U.S. have resisted these arguments, suggesting that they do not want to charge their users for using higher bandwidth, either because this might not be technically feasible or because it would be a major pricing change which might not be matched by competitors or accepted by the marketplace. Instead, service providers have been proposing to charge content providers to offset the higher bandwidth charges of the end users. This argument omits the fact that content providers already pay a service provider to host data on Internet for end users to consume.

Bandwidth limits aren't the central issue in this debate, latency is. Neutralists argue that charging for access to priority services amounts to "double taxation", but not persuasively. If standard service goes for one price, and enhanced service for another, it's consistent not to allow access to enhanced service to those who haven't subscribed to it.

The third argument is deregulatory. It says, network neutrality is a fine idea but will require government intervention, and intervention will invariably and inevitably lead to unintended consequences. This argument leads to a larger debate over whether government can ever act in useful ways, which is difficult to summarize. However, in response, some advocates of network neutrality have argued that FCC action alone is sufficient. Others argue that legislation will simply replicate common-carriage principles already long in place for many communications networks.

This argument can also be opposed on these grounds: Most high speed network providers are cable or telephone companies who are granted local monopolies by the government. Government granted monopolies must be regulated because if the monopolies act improperly market forces don't exist to correct the behavior. Evidence for this can be found in the self-same telephone monopolies work to eliminate competitors from reselling access to their networks -- networks paid for with public funds. The networks regulated by the Markey Amendment are not monopolies, however.

Some trends affecting the debate are:

  • The increasing use of Voice over IP, VoIP, and its latency requirements.
  • The increasing use of high bandwidth applications, such as online games, and music and video downloading.
  • Improvements in networking technology, which make providing broadband service, on the aggregate, cheaper.
  • The trend of governments funding the construction of high-speed networks in countries like South Korea, France, and for cities to build their own, wireless networks.
  • The increasing use of wireless home networks, which allow for neighbors to share an internet connection, thereby reducing revenues for the service providers. In urban areas this factor can be very large, with a great deal of people sharing one individual person's connection.