The current battles being fought over net-neutrality were over before they began. Whether you regard it as a good thing or a bad thing the world already has a multi-tiered internet and it’s likely to become even more stratified in the coming years.

Most markets, or perhaps countries is a better grouping, depend on commercial organisations to deploy broadband access and to serve the consumers that sign-up. The internet service providers that governments and customers depend on to reach more and more consumers and in increasingly difficult locations (read ‘less likely to generate revenue’) are there to make money. They have a board of directors, they have shareholders and without their support and confidence they don’t survive for long.

So how do they make money? For years it was enough to gradually increase the scope of their network. They figured out where they could expect to sign up enough users to justify a deployment and then rolled out. Unfortunately this is a finite market and when it’s saturated the potential for revenue growth is severely diminished. Even in the global market with plenty of customers still to go for subscriber growth is slowing and with it the growth in access revenues.

Although the absolute numbers of new subscribers added is often stable or increasing quarter by quarter the growth in percentage terms is slowing. That’s a problem for a commercial organisation where stability just isn’t sexy enough. If you want your share price to increase then you have to demonstrate revenue and preferably margin growth.

To achieve this ISPs have to increase the average revenue they make from each customer and they have to do it on a regular basis. There are two primary channels, bandwidth and services.

Many ISPs are increasing the bandwidth they offer. Cable companies in particular, with the recent deployment in many markets of DOCSIS 3.0, are offering more downstream bandwidth. DSL has been able to up its game to an extent and fibre is where the future lies.

This presents a problem for net-neutrality in itself. If two consumers on the same node try to access a particular set of packets, be it video, P2P or any data then the subscriber paying for the faster connection won’t just receive more packets per second due to the bandwidth differential they will have priority over the slower connection by the very nature of the network architecture.

While this doesn’t explicitly contradict a basic tenet of net-neutrality since the subscribers are most likely paying different amounts for their connections it makes it difficult to distinguish from a two-tiered service. If you are prepared to pay more for your bandwidth you will receive your packets more quickly.

As mentioned if an ISP wants to grow its revenues the other straightforward option is value added services. Service providers are launching more and more services and what they really prefer is that the customer takes as many of them as they can be it VoIP, TV services, a VPN, security and so on. Not only does it mean you can charge a particular user more but it’s less likely that they’ll move to another supplier.

Value added services are increasingly important to ISPs.

ISPs in developed broadband countries take an even larger proportion of their revenues from value added services than the global chart depicts and this dependence will only increase over time.

The issue of value added service provision in a net-neutral market is more controversial than bandwidth. It’s universally accepted if you pay for a faster connection you will receive a faster connection or you vote with your feet and leave. The idea that a private company can act as a gatekeeper for different internet services and decide what their users can and cannot access is a source of bitter contention.

One solution for the ISPs is to impose a data limit, often termed a fair use policy (FUP), on a customers’ account. This has the effect of dividing their customers into two primary groups. The light user who browses web pages, emails, downloads the odd piece of music and so on and the heavy users who can max out their bandwidth over an extended period of time.

It just so happens that in most cases consumers who qualify as heavy users are using P2P in some form and so a FUP effectively discriminates against them.

However many ISPs already practice more direct forms packet control. VoIP is perhaps the commonest service that is blocked but P2P sites will sometimes fall foul of a particular ISP and if you take your TV feed from one supplier can you expect to be able to access channels from another?

If we remove ourselves from the moral maze and focus on the structure of the market we are faced with an inescapable truth. We can have as much net-neutrality as we are prepared to pay for.

For a market to exist where commercial organisations provide access to the internet then those organisations have to exist. In order to exist they have to make money or at least convince their directors and shareholders that they can make money and that next year they will make more than last year. If you restrict the ability of an ISP to make money and to increase the amount it makes from each user then that will necessarily have an effect on their survival and this would have knock on effects for customers.

There may be a situation where the market reaches a stable equilibrium if all suppliers are affected equally but this will just shift the point of competition from services to, for example, the amount an ISP can spend on advertising. In an homogeneous supply market consumers will buy the product they are aware of.

So for network-neutrality to work either government and/or communities pay for their own backhaul and last mile of supply, and many argue they already have, or you run the risk of a steadily reducing choice and ultimately the possibility of local monopolies in the access market.

In a world where we depend, for good or ill, on commercial organisations to provide services there has to be some freedom. Regulation can take us so far and is vital to counter some of the excesses and innovation (cf. derivatives) that are a by-product of intense competition. To try and squeeze this particular toothpaste back into the tube however would require an the expenditure of an amount of money and an appetite for market control that has yet to be displayed by consumers, businesses, voters or politicians in most major markets in the world.