Lets Have A Competative Internet Built On National Infrastructure

The Problem

Every week, there is at least one new article describing how the internet services in "the last mile" generally "aren't working" in America for various reasons. The Verge seems to be especially good at covering this, and there was even an episode on Last Week Tonight, with John Oliver from the Daily Show. It seems to me that the issues of eroding net neutrality and the lack of bandwidth at competitive prices are due to the lack of competition with ISPs serving the "last mile" to your home. This statement is the premise of this post, and if you disagree then you might as well stop reading.

Why There Is So Little Competition

At the moment in America (and a lot of other countries), the last-mile ISPs lay down their own physical infrastructure. These are often referred to as "cable companies" and the two biggest players are Comcast and Time Warner who deliberately don't compete. Laying down physical infrastructure to homes is immensely expensive, and most companies cannot afford to lay down lines to every home whilst "capturing" only a percentage as subscribers.

What's The Solution?

Having each provider lay down their own set of cables to each and every home is just plain silly. This is similar to laying down power cables to your house for each of the different electricity companies there are, and making them pay for it. Instead, the solution is to treat these physical cables like any other infrastructure, such as roads, railways, water pipes, and power cables. The government should be responsible for building/maintaining the infrastructure, or paying competing private companies to do this for them. The lines have a set capacity which can then be divided up and leased out to private ISPs. Some ISPs may decide that they only want to lease a small amount of capacity and run a budget service for their subscribers, whilst others may decide to lease a larger amount for a "premium" service. This is similar to how mobile operators work, whereby smaller operators operate by "piggybacking" off of the infrastructure of larger operators such as T-mobile and Vodafone. These smaller operators charge a smaller fee, but only have a fraction of the capacity to operate within. The government pays for development initially through taxes, but then through the money it receives through leasing existing lines.

A Working Example

British Telecom in the UK is the closest example that I know about. There are actually two different entities that make up what most people think of as "BT". These are BT Openreach, a group that is responsible for laying down and maintaining the physical infrastructure, and BT itself as an ISP. BT operates only on the lines that BT Openreach is responsible for, but so do hundreds of other providers, such as TalkTalk and Sky. If you are with any ISP in the UK, other than Virgin Media, then you are probably using these lines. The only difference is that a private organization is laying down the lines rather than the government, however many/most of the original infrastructure was originally built when BT was a national organization run by the government rather than a private company. In the spirit of competition, the UK government created grants that private companies could bid for in order to build infrastructure, but then pretty much gave them all to BT.

How Will Competition Solve Net Neutrality?

This largely depends on market forces. If the people see net neutrality as one of their most important factors when choosing an ISP, then they may be prepared to pay slightly more for an ISP that provides this, thus growing that ISP's market share and perhaps forcing others to do the same in order to compete. This is much like how the organic food market has grown, along with free-range eggs. In my opinion, it would be better to enshrine net neutrality in law, but when I mention this to some, they tell me to "Go live in China".

How Will Competition Solve Bandwidth Issues?

As more ISPs move onto a line, or ISPs try to grab more capacity, prices per megabit on those lines will increase. The government, or private company, that maintains the lines will then have enough incentive to lay down more lines in order to increase the capacity that they can sell off. It's more profitable to lease 20 Gigabits at an average of $1 per megabit, rather than 10 Gigabits at an average of $1.9 per megabit. (These are not real-world numbers). The only risks are that if run by the government, they might not act on this demand (incompetence/bureaucracy), in which case this is a political policy that should influence your vote when electing your representatives. Alternatively, if a private company is leasing the lines as a monopoly, it may decide that it is simpler to just let the price-per-megabit increase, rather than investing capital in laying down more lines to increase revenues, which is why I prefer the government based strategy.