Thursday, 9 June 2016

NBN: Australia still behind Canada despite policy effort

The latest Akamai State of the Internet report provides more evidence of Australia lagging behind Canada, despite almost a decade of policy effort. I have argued for a long time that this is a result of the inefficacy of Australia's centrally-controlled telecommunications policy regime.

This means that the NBN's lack of impact has little to do with good versus bad NBN models. I have stated previously that government control of the industry is the problem, not the solution, This view was guided by my research to explain why Canada had better broadband than Australia back in 2005.

My research compared telecommunications technology outcomes from the time of the telegraph to the present in Canada and Australia. The findings suggested that the problem lies with Australia's telecommunications policy framework. It was never a case of a lack of government interest or funding, but something fundamentally flawed in the centrally-controlled system that would make it a perpetual political football.

As Canada and Australia are well-suited to the 'most-similar' comparative approach, I adopted Mill's method of difference to explain why these two very similar countries had very different institutional arrangements with very different outcomes. All accepted methods of inferring causality.

I referred to the telecommunications policy frameworks as the monolith (Australia) versus the mosaic (Canada - see Wilson 2000: 25). Neither system was necessarily a result of deliberate design. Indeed, Canada's superior outcomes were the result of serendipity, in that its telecommunications policy framework grew organically, locally, and, as a result, constitutionally; the telecommunications powers reside with the provinces rather than the national government.

The reverse is true in Australia, but telling people about it does not prove popular.

In Australia, the broadband 'crisis', leading to the advent of the NBN, was a case of history repeating. The federal government alone would solve the crisis. After careful examination of the historical record, it was apparent that the broadband 'crisis' was no different than the telegraph crisis, the telephone crisis, the radio crisis, the television crisis, and so on, each appearing in similar magnitude when compared with Canada over time.

The State of the Internet report demonstrates that the federal government's faith in its own abilities has (again) not been justified.

Akamai provides an interactive and customisable graphical report generator known as 'connectivity visualizations' (incorporated in this post). Compared with my many years of 'empirical rummaging' (Skocpol 1995: 104) to build a chronological database of telecommunications technology penetration in Canada and Australia, using the Akamai visualiser is a cake-walk.

The Akamai blog provides a useful explanation of the metrics used in these visualisations.

First, I will return to some of my earlier, painstakingly put-together 'visualisations'. Here is the state of broadband speed in Canada and Australia in 2008:

Australia-Canada Broadband Speeds by Household 2008 (constructed from Akamai 2009 data)
This trend is remarkable in that it has not changed over time, given the extent of political focus and taxpayer investment in NBN. The latest Akamai report shows average connection speeds from the period above to the end of 2015:

Australia-Canada Comparison of Average Connection Speed 2007-2015 (Akamai 2016)

Second, the measurement of broadband penetration has changed, in that it is now possible to distinguish between unique IP addresses. But here are the figures for broadband penetration by household from 2004-2007:

Australia-Canada Broadband Penetration 2004-2007 (constructed from OECD data)
Below the Akamai visualiser shows the same trend continuing (note this is by number, not by percentage of population, but you can see that the trend is consistent, if not levelling out:

Australia-Canada Unique IP Addresses Comparison (Akamai 2016)
Finally, the percentage of services 15mbps and above, the speed Akamai regards as the minimum for 4K (or full Ultra HD quality video streaming) services:

Australia-Canada Percentage of Broadband Connections above 15mbps (Akamai 2016)
This last visualisation is the clincher. Despite NBN in whatever form, despite the investment of taxpayer dollars, Canada is not only (still) ahead of Australia, but increasing its lead.

This is not the fault of Mr Turnbull's NBN-lite. This has nothing to do with policy differences between the two major parties. This is the result of historical processes where the first principles of Australian telecommunications policy lie buried. Some time ago, I wrote an article for the now-defunct The Punch (19 May 2011), where the title was edited to read 'The NBN’s the culmination of 150 years of cock ups'. In hindsight, the title was closer to the truth.

But it is difficult to change an institution, and the way telecommunications is done in Australia is just that. The retort "Don't confuse me with your facts" comes to mind. Even Professor Reginald Coutts thinks the evidence is my personal 'fantasy'.

To channel Professor Julius Sumner Miller (1975), "Why is it so?"

These are all the hallmarks of a policy regime, an entrenched way of 'doing' policy that is difficult to change. The use of policy-based evidence rather than evidence-based policy is part of the regime's self-reinforcement. And that is why Australia is still behind Canada in broadband.


Miller, J.S. (1975). Why is it so? Light and modern physics. Sydney: Australian Broadcasting Commission.

Skocpol, T. (1995). Why I Am an Historical Institutionalist. Polity 38: 103–106.

Wilson, K.G. (2000). Deregulating Telecommunications: U.S. and Canadian Telecommunications, 1840-1997. Lanham, MD: Rowman & Littlefield.

Creative Commons License Except where indicated otherwise, Le Flâneur Politique by Michael de Percy is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 2.5 Australia License. Based on a work at Background image © @redshinestudio