Home » CurrentHeader, Internet

Australians fight for Internet equality with non-neutral networks

By 10 February 2010 7 Comments

When it comes to Internet usage allowances, the Internet has not been kind to most of the world’s Internet users compared to American Internet users.  Americans get some of the most generous bandwidth caps in the 150 to 250 gigabyte (GB) range while most of the world gets pathetic usage allowances (see figure 1 below).  Even the smallest proposed usage cap by Time Warner of 40 GB per month dwarfs the typical usage caps implemented in most other first world nations.

Figure 1 – Comparison of average usage cap allowance in the world
Global usage caps for broadband access
Source: OECD, Comcast, AT&T, and TimeWarner

Despite the fact that the Time Warner proposal in its final form meant price reductions for the vast majority of its customers, had limited overage charges, and wavers on the first month of overage charges so that customers had ample warning, the proposal was extremely unpopular and even prompted one congressman to propose a ban on usage caps.

Yet as controversial as the 40 GB Time Warner usage cap was, it would be generous for the typical Australian broadband consumer who typically gets 5 to 20 GB of usage allowance for plans under $40 (AUD) per month.  Orcon Broadband in New Zealand for example offers a 1 GB plan for $70 (AUD) per month.  As foreign as this sounds for Americans, it has always been this way for the average Australian since the dawn of the broadband era over the last decade.

But why has the Australians and the rest of the world gotten such a rotten deal?  The reason is that the Internet has largely been American and to a lesser extent European centric because that is where most of the content lives.  All the most popular sites like Google, YouTube, Facebook, E-bay, Amazon, Yahoo, MSN, etc started off in the United States.  So if the Australians wanted to access that content, they had to come to the US to access that content which involves leasing or building extremely expensive fiber networks that run under the Pacific Ocean.  This meant huge transit costs to reach US and European based networks which means Australians have never been able to access high quality content at an affordable price.

Note: Just one hour of YouTube 720P HD content that averages 2.25 megabits per second (Mbps) would use over 1 GB of allowance and that would exceed the usage budget of many Australian broadband consumers.

Our friends from down under are understandably unhappy with the status quo and it appears that they are finally doing something about it.  Why should Australians pay long distance transit costs to the US for the honor of patronizing content companies?  Why not attract content companies to build up a presence in Australia so that Australians can enjoy the affordable usage allowances that Americans have had all along?

Australian broadband providers have figured out that simply switching to a dual-tier usage cap with a generous or uncapped limit for local destinations that don’t incur heavy transit costs and tight usage allowances on long distance transit networks provides the right incentive for content providers to come to Australia.  Imagine broadband for example has a “soft-cap” of 2 GB for International access and 28 GB for local access.  Orcon Broadband has promotional deals with no limits on YouTube.  This likely means that Google extended their edge cache to Orcon’s network with free or very inexpensive peering.

This “non-neutral” arrangement incentivizes content and application companies which are largely based in the US and Europe to extend their presence to Australian to better serve Australian customers.  These peering agreements are a win-win situation that reduces the content provider transit costs (typically $2 to $9 per Mbps per month) and turns the Australian broadband provider’s transit costs into revenue.  The Australian consumer is also a big winner because they can now access content at affordable local rates which gives them the same deal that Americans have always had.

Since everyone is a winner here, one would expect a happy ending but the Net Neutrality proponents have gone ballistic over this non-neutral arrangement.  Nate Anderson from Arstechnica argues these peering agreements harms the smaller content providers because how can they possibly compete with big content?  Anderson argues that all sites should live under the same cap regardless of whether they’re local or transit networks in order to preserve neutrality, but this would simply mean that Australians would be forced to live with unnaturally low usage allowances for all sites and it removes the incentive for content companies host content on Australian networks.  Why should such an artificial limit be imposed on local content for the illusion of neutrality?  Why should consumers be forced to pay through the nose in transit costs to reach content when it’s cheaper for content to come to the consumer?

Note: It is ironic that the Net Neutrality proponent’s favorite cause of municipal fiber networks (which were a disaster in reality) have the exact same kind of non-neutral arrangement that they oppose.  Municipal fiber networks allow local traffic to run at 100 Mbps but Internet traffic going to other networks (which could be same city, state, or country) get capped at 10 Mbps.  So if a content provider wanted to deliver content at 100 Mbps to the municipal fiber broadband customers, they have to essentially “peer” with the municipal fiber operator by purchasing a network connection from them.

Anderson’s arguments are similar to the backward logic used by the NYU study which seeks to “protect” content providers from cheaper and higher quality peering services even though it would force the content companies to pay more in transit charges.  These are the twisted arguments being presented to the FCC to get them to regulate and ban peering services.

Net Neutrality proponents are essentially arguing that content providers must be forced to continue paying high transit costs, consumers should be forced to pay high transit costs and/or live with consistently small usage caps, and broadband providers shouldn’t be allowed to make any money from dotcoms even though they employ ten times more employees than the dotcoms.  To make their arguments fly, they simply omit the facts and resort to scaring the public about the evils of non-neutrality.

The reality is that peering benefits the smaller content providers and it is an essential open bandwidth market that is more efficient from an engineering and economic standpoint.  It has transformed the Internet over the last few years and enabled on-demand streaming video on the Internet.  Without the ability to peer and bypass the financial and performance bottleneck of transit networks, large scale video streaming would have been impossible.  This is the true innovative nature of the Internet that must be preserved.


  • James said:

    Wait, 1GB of quota costs $70 in Australia? Since when!?

    I’m paying $70 per month to Adam Internet (my ISP), and I get 80GBs of quota. I just checked a couple of other popular ISPs, Internode and iiNet, and they get 50GB and 55GB respectively. I don’t know who the hell Orcon are, but that is NOT a representative sample of what broadband costs here.

    Hang on a second, I just checked the Orcon website, and it appears that they are a NEW ZEALAND company. New Zealand is not part of Australia!

    Unless I’m missing something, I think a correction needs to be made in this article…?

  • George Ou (author) said:

    James, thanks for pointing that out. New Zealand is part of the “continent” of Australia and that’s what I meant. Now you’re probably right that the prices aren’t as high in Australia, but the caps are typically very low for the plans that most people pay for which is typically lower than $70 AUD.

    But that’s a minor point James. The key point is that content providers should host more content in Australia so that they can get better usage allowances on local content.

  • James said:

    Yes, I definitely agree with the general point of the article.

    Another minor correction though: New Zealand isn’t part of the continent of Australia. According to the Wikipedia article you linked to, the Australian continent consists of Australia, Papua New Guinea, and portions of Indonesia. To be honest, didn’t actually know this either! I thought the continent of Australia consisted solely of the country of Australia.

    Still, New Zealand is a close neighbour to Australia, and the only other 1st world country in the area. There are a lot of similarities between the two countries, so your point still stands :)

  • George Ou (author) said:

    You’re right James, I should have said Zealandia.

    Good luck to you down there, it’s about time that you guys have better access to content.

  • Dash said:

    The problem here is that the rationale for capitalism depends on demand being based more on the quality of a product, not its availability. The point is to reward innovation, reward the better product, not reward the company that already has the resources to make their product more available, even if it’s an inferior product. The customer will generally take the cheaper product, unless the more expensive product is that much better, or that much better marketed.

    This sort of setup does benefit genuinely local companies, though once again, they are rewarded for being local, not for the quality of their product. It also benefits those already on top of the food chain. However, it penalizes a smaller, nonlocal company that is trying to reach a bigger audience with their superior product. If customers will be dissuaded from accessing their product, then they might as well throw in the towel. The local company is also unlikely to be able to compete with the large international company, because their local advantage isn’t an advantage if the international company can expend its immense resources to gain the same advantage, so they might as well give up too. Then all we’re left with is massive international corporations with a monopoly on a market that they’ve purchased, and not won through providing a better product. Prices go up, quality goes down, and customers have no choice.

  • George Ou (author) said:

    Dash, I don’t think you understand what’s going on here. It’s not restricted to local companies. All companies (local or foreign) are invited to cache and/or host their content in Australia for a reasonable (probably lower) price and at better performance. It’s a win/win for everyone.

  • Dash said:

    It’s clear how some content companies, and local CDNs, would benefit from this kind of arrangement, it’s not clear how it benefits the consumer.

    There are several relevant metrics that interact with the market in this situation. Off the top of my head, there’s
    1) The quality of the product that a company provides.
    2) The size and scope of the company.
    3) Whether or not they have a local presence.

    I’m saying that #1 should be the primary factor in market share, and that providing cheaper access to a company’s products when they have a local presence skews the market, increasing the importance of #3 as compared to #1. And when #2 can be used to purchase #3, that skews the market even further. Consumers are incentivized to patronize certain companies, independent of the quality of their product. It’s a barrier to free trade, similar to a tariff or import tax.

    I suppose this doesn’t really look like a problem until you look at the big picture. Sure, if it were only one country, maybe everyone who wanted to do business there could afford to peer locally. But there are a lot of countries in the world. How many businesses can afford to peer with all the local networks in the world?

    That’s why I said it creates a situation in which local companies, and big foreign companies who can afford a local presence, will profit and acquire market share regardless of the quality of their products, whereas a foreign startup is already at a disadvantage, even if they have a better product.

    The customer only wins if you assume that all they want is cheap content, and that any content will do. This is a very commercial view of content, and of the internet in general. You would reduce the ability of the individual to compete with established companies, and turn the internet into just another “big media” platform.