Why a metered Internet would actually be good

As the US telcos move towards bandwidth caps the justice of this has been a large topic of conversation in the media and on the web. One of the possible outcomes of all this is a metered Internet service where users pay per MB that they use. From some commentary I have heard in response to a positive article on this plan by John Dvorak in PC Magazine, this is not a popular option for all. A metered Internet would actually offer real benefits to users and would be likely to end up with a cheaper and better service.

The most often cited objection to this plan is the belief that once the infrastructure of the Internet exists then incremental bandwidth is essentially free. This is unfortunately not the case. Incremental revenue is not free, it is just really cheap which is not quite the same thing. While ISPs do not need to add extra capacity each time someone uses an extra MB but the equipment they have does have limited capacity. This means that the incremental cost does not have a nice gentle slope and instead has very large and steep steps. This is very important in understanding the economics of the situation.

Lets put a couple of things aside in this discussion though, there are ways that the telcos game the system to day that are not fair to consumers and they will probably try and game any proposed system. This should not be the key factor in determining the best solution though. The core solution should enable the telcos to make decisions that are best for us consumers and the policing regulations should be on the margin. The fact that the majority of the Internet infrastructure was funded with public money is not cogent to this argument.

In the current business environment and laws the companies that run our Internet backbones and access are required to make the maximum return for shareholders. If they are charging a flat rate for unlimited (or at least very large limit) traffic, the way for these companies to increase their revenue is to add more subscribers. There is no incentive for them to increase backbone speed when doing this, it is not required to gain the revenue and the marginal impact of each user added will not change the average utilisation rate of their infrastructure. Over time though these little additions become a large amount of data use and the system starts to strain.

The companies are then faced with a choice to upgrade their bandwidth or not. Upgrading will improve service for their existing customers, who are already paying them for what they are getting. Upgrading will also not ensure them to gain new customers, and may cost them customers with the inevitable service issues the upgrades will cause. In essence upgrading their systems will cost a lot for no short/medium term revenue increase so it will actually hurt the companies bottom line. An option that is more likely to keep the executives in jobs is to take actions that limit the amount of bandwidth people consume in ways that have low cost. Bandwidth caps anyone? Blame the users for any slowdown caused by the lack of capacity. Maybe also hit up the government for funding for upgrades, or stealth charge the consumers for upgrades by creating a tiered Internet and charging for ‘premium’ Internet traffic. All predictable outcomes from unlimited access plans.

In comparison if the Internet was metered like any other utility the motivation on the ISPs becomes very different. Suddenly their revenue improves with each MB people transfer over their network. If they let the backbone of the system slow down then it hurts their top line and slows their growth. In this case there is clear incentive for them to add capacity as they add users. The users themselves are not what drives the revenue, but the capacity they use, so if that capacity is not there then the revenue from the new users will not flow.

There would need to be tuning around the margins, like base flat rate plans for those that need to know what their bill will be each month in advance. Preferably there would be a base free component of a GB or two a month that would ensure everyone had free access to the essential Internet services like encyclopedias, email, bank sites etc but not enough to enable rich media or online gaming. There would also need to be a way to set your own limits on the monthly bill or even to have pay up front options like on mobile phones.

This system would not make us pay more, the same supply/demand factors that keep prices where they are now would cause prices to average the same as today, high users would pay a bit more and lower users would pay less. The economics of this are the same as they are for electricity, water, grapes or milk. Prices will be low enough to encourage consumers to consume while making sure there is profit for the company. The differences being that we would be in control of what we spent and the ISPs would be encouraged to help us download more at faster speeds. The system we have now encourages them to keep us fishes crammed into our current bowl even though we are growing and more fish keep coming in to join us.

2 thoughts on “Why a metered Internet would actually be good

  1. OK, so go back to the way phones used to work. By utilization. Of course, the phone companies finally figured out that by metering phone usage, it was actually costing them more in accounting and measuring that it was to just provide the service. Today I can call the whole US and Canada, and I pay nothing extra.

    VOIP? Nope, it’s a good old AT&T land line. Cell phones didn’t charge extra for long distance either, because they made the money off the minutes used, not the long distance.

    Going backwards to try and go forward doesn’t work.

  2. Phone plans are not the way they are now because they are easier for the phone companies, they are that way so it is harder for you to compare phone plans for which will be the cheapest for you and therefore make themselves more money on average. Scott Adams once called it a confusopoly.

    This works because we all have this little hard wired irrational logic in our brain that rates a variable future cost as a risk. We therefore prefer a set cost for future services regardless of whether this would mean a more expensive service for us.

    Phone plans that appear to give you things for free are not actually better as they hide the actual costs from you. This takes power away from you and confuses rather than informs your choices.

Comments are closed.